Anda di halaman 1dari 446

Landscape Series 23

Marc Antrop
Veerle Van Eetvelde

The Holistic Nature of Landscape
Landscape Series

Volume 23

Series editors
Jiquan Chen, Center for Global Change and Earth Obser, Michigan State University,
East Lansing, USA
Janet Silbernagel,, Nelson Institute for Environmental
Studies, Madison, USA
Aims and Scope
Springer’s innovative Landscape Series is committed to publishing high-quality
manuscripts that approach the concept of landscape from a broad range of per-
spectives. Encouraging contributions on theory development, as well as more
applied studies, the series attracts outstanding research from the natural and social
sciences, and from the humanities and the arts. It also provides a leading forum for
publications from interdisciplinary and transdisciplinary teams.
Drawing on, and synthesising, this integrative approach the Springer Landscape
Series aims to add new and innovative insights into the multidimensional nature
of landscapes. Landscapes provide homes and livelihoods to diverse peoples; they
house historic—and prehistoric—artefacts; and they comprise complex physical,
chemical and biological systems. They are also shaped and governed by human
societies who base their existence on the use of the natural resources; people enjoy
the aesthetic qualities and recreational facilities of landscapes, and people design
new landscapes.
As interested in identifying best practice as it is in progressing landscape
theory, the Landscape Series particularly welcomes problem-solving approaches
and contributions to landscape management and planning. The ultimate goal is to
facilitate both the application of landscape research to practice, and the feedback
from practice into research.

More information about this series at

Marc Antrop • Veerle Van Eetvelde

Landscape Perspectives
The Holistic Nature of Landscape
Marc Antrop Veerle Van Eetvelde
Department of Geography Department of Geography
Ghent University Ghent University
Ghent, Belgium Ghent, Belgium

ISSN 1572-7742 ISSN 1875-1210 (electronic)

Landscape Series
ISBN 978-94-024-1181-2 ISBN 978-94-024-1183-6 (eBook)
DOI 10.1007/978-94-024-1183-6

Library of Congress Control Number: 2017949530

© Springer Science+Business Media B.V. 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with
regard to jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer Science+Business Media B.V.
The registered company address is: Van Godewijckstraat 30, 3311 GX Dordrecht, The Netherlands

Our Common Landscapes for the Future

The solution of the ecological crisis through a transformation towards sustainable

development is the most urgent challenge in our time. It is not just a question of
CO2 and the conversion towards renewable energy. The reestablishment and
conscious development of our common landscapes at all spatial scales for the
combined benefits of biodiversity, our cultural heritage and the preservation and
development of ecosystem services will probably be the most comprehensive and
necessary social task in the future if a conversion towards a sustainable develop-
ment shall succeed in due time. This is so because it will be necessary to ensure that
the potential material wealth of modern society will be transformed into a good,
fruitful and healthy life of mankind, acknowledging our multifaceted needs and our
universal dependence on a continuous cooperation with nature in all its heteroge-
neous variability. Our common landscapes express the unification of the heteroge-
neous structures and functions constituting our physical environments, a basic
condition of our wealth and of the good life. We must learn this lesson – even if
an ever-growing ideology of individualism tends to lure us to ignore it.
A holistic view at the landscape, expressing the intimate relationship between
man and nature in the understanding and use of the landscape, has been a classical
role of geography. In Europe, it developed primarily within the German tradition
based on von Humboldt and Ritter and within the French tradition of regional
monographs developed by Vidal de la Blache and others. Each of these traditions
emphasised different components of the relationship but always as part of an
integrated unity. In Belgium, geography developed under the inspiration of both
traditions giving rise to sophisticated synthesising trends within a holistic approach.
Such efforts were however in general not a success within the academic world
during large parts of the twentieth century, although a certain positive response
from the complex problems facing land management in the real world could be
recognised but also sometimes misused for ideological purposes. Geography as a

vi Foreword

synthesising holistic activity went more and more isolated in an academic commu-
nity, still more dominated by constant specialisation and partitioning of research
within science, social science and humanities in a targeted search for universal
laws. Geography disintegrated as a discipline, and holistic geography was more and
more seen as a mere descriptive activity, a tradition that in fact was considered a
hindrance for the establishment of general laws, often considered laws working
independently of time and space. Also outside geography, studies of spatial differ-
entiation at all levels became a rare field of interest, although a remarkable part of
spatial theories and methods being promoted later, in fact, has been developed by
individuals and schools in a variety of disciplines within this period. In addition,
most applied sciences of relevance for land use, such as agronomy, forestry and
physical planning, developed more and more with the presumption of a homoge-
neous space that made redundancies about spatial differentiation.
Changes occurred however from the 1960s provoked by the upcoming environ-
mental crisis. This manifested itself in a myriad of spatially differentiated problems
related to the violent physical processes brought in during the post-war era that
included extensive land use changes, industrialisation and urbanisation based on
fossil fuels and non-circular flows of matter in nature and society. A strong need
was reintroduced for studies of spatial differentiation at all levels to many disci-
plines. At the same time, the development of computer technology drastically
changed the technical possibilities for managing spatial analysis, and development
of remote sensing techniques offered the possibility to deliver huge quantities of
spatially differentiated data to be used for both spatial analyses and syntheses.
This provoked the development of new, often interdisciplinary organised fields
of interests to study spatial, chorological relations associated with environmental
problems to find sustainable solutions that could be adapted to different landscape
conditions. New disciplines developed such as landscape ecology, restoration
ecology, environmental studies, cultural ecology, etc., but spatial trends came
also to the forefront again in many established academic disciplines, as well as
within practically applied disciplines such as agronomy, forestry, landscape archi-
tecture and physical planning.
Especially in old-developed countries characterised by marked cultural land-
scapes, this renewed landscape interest also resulted in holistically oriented land-
scape initiatives at the political level, such as the implementation of the European
Landscape Convention. The convention recognises all kinds of landscapes at all
spatial scales, including everyday and ruined landscapes, as a common concern for
everybody, and recommends integrated landscape planning and management as a
common social responsibility for all citizens. Its influence can be followed, e.g. in
the endeavour to integrate agricultural and environmental policy within the
European Union – although not yet with any remarkable success.
Especially within the many new interdisciplinary fields of interest that have
evolved in the wake of the ecological crises, geographers trained within the holistic
traditions of their discipline have got a special synthesising role not only concerning
theoretical and methodological aspects of spatial analysis but also concerning the
linkage between the natural and cultural aspects of landscape development, trends
Foreword vii

in land use and land cover as well as social and historical aspects of landscape
perception. Geographers from Ghent University have shown a remarkable ability
and perseverance to integrate these different aspects of the landscape relevant for
the future use and protection of our environment. Marc Antrop and Veerle Van
Eetvelde have been at the forefront for this development, so it is both logical and
very welcomed that they collect and publish their experiences in a book about the
holistic nature of landscape.
There is no single “objective” way to introduce “the holistic nature of land-
scapes”. In this book, they start by the many different histories of landscape
research, originating from very different landscape practices and different related
opinions on how landscapes can be perceived and understood, which basically
influences the approach to research concerning the way landscapes as material or
perceived units can be experienced and expressed linguistically for different pur-
poses. They continue with the presentation of landscapes as units characterised by
dynamic processes linked to spatial patterns being sustained but also changed in
different time perspectives. They present different approaches to landscape assess-
ment for practical, aesthetic and mental use purposes and as a source of inspiration.
And they end up in the discussion of the establishment of the common endeavour of
taking practical care of the landscape in the future.
However, especially during the last decades, this new agenda of a growing
responsibility concerning the conscious transformation of our landscapes as an
important part of a future transition towards sustainable development has been
challenged by the reintroduction of an alternative agenda, namely, the agenda of a
neoliberal globalisation, closely related to the renewed demand on an open market
pushed forwards by the World Trade Organization. These two agendas are now
running their own individual life almost independently from each other. The
globalisation agenda is driven by technological and economic renewal, dominated
by traditional economic power, although the introduction of more “sustainable
technologies” is gaining ever-increasing importance also in this agenda. In com-
parison, the agenda on sustainable development is more defensive and with less
influence on the present rapid landscape changes. The agenda also differs in the fact
that globalisation is oriented towards an open market with the individual producer
and consumer in focus, whereas the agenda of sustainable development is oriented
towards collective goals, such as nature protection, pollution, common land use,
social justice, etc.
Also, when it comes to planning and management of the landscape, the two
agendas cannot be properly paralleled, since the globalisation agenda at the polit-
ical level is accomplished almost without any spatial or geographical dimension,
whereas the sustainability agenda is closely related to the handling of the differen-
tiation in the material and cultural environment apprehended at different spatial
The two agendas tend to affect abstract academic thinking in different ways:
Where the sustainability agenda seems to have promoted a certain showdown with
the tradition of general non-spatial academic thinking, on the contrary, the global-
isation agenda seems basically to maintain it.
viii Foreword

This is a growing problem since abstract conceptual thinking continues to build

up barriers to the understanding of spatial relations in landscapes and through that
also the necessary common solutions concerning their protection and use.
The Swedish geographer Torsten Hägerstrand, mostly known for his theoretical
studies on spatial innovation and time-geographical studies, ended his carrier as a
strong supporter of an integrated geography that could serve the study of the
necessary changes in land use and landscape management related to a transforma-
tion towards a future sustainable development. Despite the growing interest in
spatial studies, he was very critical of the contemporary trends towards abstract
academic thinking especially within the social sciences that make it difficult to
relate concepts to the material world, always organised and conditioned in a spatial
way which makes the landscape crucial both in nature and society. He once
expressed the problem in such a simple and basic way that it almost got a daring
character: “The most important is not to regard humans as ‘society’ or ‘business’, or
something else abstract, but on the other hand, as physical entities that are con-
stantly somewhere, being involved in more or less concerted action. They carry
their knowledge and assessments with them. Such do not float freely in some kind
of cultural space. It is only when you stop unzipping your concepts from the specific
landscape context that you are able to see how the element you are studying is
incorporated into changing neighbourhoods. It is important because touch is the
most elementary relationship of existence”.1
Thus, the challenge of sustainable development is closely related not only to the
ability but also the will to adapt to the multiple varieties of specific
“neighbourhoods” in the form of ever-changing spatial landscape conditions that
cannot be ignored but presupposes a wise interpretation of the very complex natural
and social processes of our common landscapes. Here we cannot rely on general
political truism.
But what is a wise interpretation? What is wisdom? This question was already
answered by Heraclitus of Ephesus for more than 2500 years ago: Wisdom is to
speak the truth and to act obeying nature.

Professor Emeritus Jesper Brandt

Roskilde University, Denmark
June 2017

Torsten Hägerstrand: Natur och Samhälle. NordREFO 1983:1.

1 The Holistic Nature of Landscape – Landscape

as an Integrating Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Climbing the Mountain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The Landscape Is in the Eye of the Beholder? . . . . . . . . . . . . 5
1.3 Dealing with “The Whole That Is More Than
the Sum of its Parts” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 The Structure of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 The History of Landscape Research . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 A Complex Story . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 The Early Beginnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.4 The Emerging Scientific Research – The Landscape
as Object of Study of Geography . . . . . . . . . . . . . . . . . . . . . . 13
2.4.1 The Societal Context . . . . . . . . . . . . . . . . . . . . . . . 14
2.4.2 The Landscape Concept in National
Geographical Schools . . . . . . . . . . . . . . . . . . . . . . . 15
2.5 Landscape from the Air: Aerial Photography
and Historical Geography . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.6 The Loss of Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.7 The Humanistic Approach and the Revival of Landscape
Ecology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.8 The ‘Landscape Crisis’ and the Shift Towards
Applied and Transdisciplinary Landscape Studies . . . . . . . . . . 22
2.9 Landscape in n-Grams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3 The Multiple Meanings of Landscape . . . . . . . . . . . . . . . . . . . . . . . 35
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Origins and Etymology of the Word Landscape . . . . . . . . . . . 36

x Contents

3.3 Subtleties of Language – Landscape Versus Land . . . . . . . . . . 39

3.4 Landscape with Adjectives . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4.1 Natural and Cultural Landscape . . . . . . . . . . . . . . . 43
3.4.2 Rural and Urban Landscape . . . . . . . . . . . . . . . . . . 45
3.4.3 Ordinary and Spectacular Landscapes . . . . . . . . . . . 46
3.4.4 Landscape and the Beautiful, the Sublime, the
Picturesque and the Pictorial . . . . . . . . . . . . . . . . . . 46
3.4.5 Ephemeral and Seasonal Landscapes . . . . . . . . . . . . 49
3.5 Formal Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.5.1 Cultural Landscapes in the UNESCO
World Heritage Convention . . . . . . . . . . . . . . . . . . 50
3.5.2 The European Landscape Convention . . . . . . . . . . . 51
3.5.3 Shifting Landscape Research Since
the Coming of Formal Definitions . . . . . . . . . . . . . . 55
3.6 Elements for an Operational Definition of Landscape . . . . . . . 57
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4 Approaches in Landscape Research . . . . . . . . . . . . . . . . . . . . . . . . 61
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2 Ways of Seeing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2.1 The Bird’s-Eye Perspective: Landscapes
at a Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2.2 The ‘Interior’ Perspective: Being in the
Landscape – Lookouts and Composite
Landscapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.2.3 The Inner Perspective: Mindscapes
and Visualisations . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.2.4 Landscape as Meta-Reality: The Transcendental
Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.3 Disciplines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.3.1 Geography and Historical Geography . . . . . . . . . . . 71
4.3.2 Landscape Ecology . . . . . . . . . . . . . . . . . . . . . . . . 72
4.3.3 Historical Ecology . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.4 Archaeology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.5 Environmental Psychology . . . . . . . . . . . . . . . . . . . 74
4.3.6 Landscape Architecture . . . . . . . . . . . . . . . . . . . . . 75
4.3.7 Economics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.4 Inter- and Transdisciplinary Approaches . . . . . . . . . . . . . . . . 76
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5 Basic Concepts of a Complex Spatial System . . . . . . . . . . . . . . . . . 81
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.2 Holism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.3 Scale and Heterogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.3.1 A Source of Conceptual Confusion . . . . . . . . . . . . . 89
5.3.2 The Spatial Scale . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Contents xi

5.3.3 The Temporal/Time Scale . . . . . . . . . . . . . . . . . . . . 90

5.3.4 The Organisation and Planning Scale . . . . . . . . . . . 90
5.3.5 Landscape Heterogeneity Is Scale Dependent . . . . . . 90
5.4 Discrete Objects and Continuous Phenomena . . . . . . . . . . . . . 91
5.5 Landscape in Layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.6 The Map Is Not the Landscape, Nor Is Its Representation . . . . 92
5.7 Borders, Fuzziness, Gradients and Ecotones . . . . . . . . . . . . . . 94
5.8 Interaction Between Spatial Patterns and Processes . . . . . . . . . 95
5.9 Connectivity and Connectedness . . . . . . . . . . . . . . . . . . . . . . 96
5.10 Multifunctionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
5.11 Reading a Palimpsest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6 Sensing and Experiencing the Landscape . . . . . . . . . . . . . . . . . . . . 103
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.2 Landscape Perception, Experience and Preference . . . . . . . . . 104
6.3 Seeing the Landscape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.3.1 The Human Vision . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.3.2 Concepts and Definitions . . . . . . . . . . . . . . . . . . . . 108
6.3.3 Landscape as a Scene: Depth Layers,
Viewing Sectors and Skyline . . . . . . . . . . . . . . . . . 111
6.3.4 Photographic Perspectives . . . . . . . . . . . . . . . . . . . . 113
6.3.5 Panoramas, Vistas, Eye-Catchers and Landmarks . . . 118
6.4 Conditions of Perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.4.1 Standing Where? . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.4.2 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
6.4.3 Atmospheric Perspective and Skylight . . . . . . . . . . . 121
6.4.4 Time and Lighting Conditions . . . . . . . . . . . . . . . . . 122
6.5 Gestalt-Principles and Perception . . . . . . . . . . . . . . . . . . . . . . 123
6.5.1 Visual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
6.5.2 Soundscapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
6.6 Experiencing the Landscape . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.6.1 Landscape Experience Assessment . . . . . . . . . . . . . 128
6.6.2 Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.6.3 Research Models and Paradigms . . . . . . . . . . . . . . . 133
6.7 Experts and Laypeople Experience the Landscape Differently . 135
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
7 Landscape Dynamics and Evolution . . . . . . . . . . . . . . . . . . . . . . . . 141
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
7.2 The Landscape is Dynamic . . . . . . . . . . . . . . . . . . . . . . . . . . 142
7.3 Landscape Genesis and History . . . . . . . . . . . . . . . . . . . . . . . 143
7.3.1 From Traditional to New Landscapes . . . . . . . . . . . 144
7.4 Contemporary Driving Forces and Processes of Landscape
Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
7.4.1 A Global View – Driving Forces . . . . . . . . . . . . . . . 147
7.4.2 Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
xii Contents

7.4.3 Networking: The Network Society

of the Information Age . . . . . . . . . . . . . . . . . . . . . . 156
7.4.4 Calamities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
7.5 Models and Processes of Urbanisation . . . . . . . . . . . . . . . . . . 158
7.5.1 Urban or Not Urban, That’s the Question . . . . . . . . . 160
7.5.2 A Process of Diffusion: From Urban Sprawl
to Functional Urban Areas . . . . . . . . . . . . . . . . . . . 161
7.6 Concepts and Models to Study Landscape Change . . . . . . . . . 165
7.6.1 The Biography of a Landscape . . . . . . . . . . . . . . . . 165
7.6.2 Space-Time Models to Study Change . . . . . . . . . . . 165
7.6.3 Time Depth and Landscape Paths/Trajectories . . . . . 167
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
8 Analysing Landscape Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
8.2 Decomposing the Landscape: Models for Analysis . . . . . . . . . 178
8.2.1 Model 1: Element, Component, Structure . . . . . . . . 178
8.2.2 Model 2: Point, Line, Polygon, Surface . . . . . . . . . . 180
8.2.3 Model 3: Patch, Corridor, Matrix, Mosaic . . . . . . . . 180
8.2.4 Model 4: Mass, Screen, Space . . . . . . . . . . . . . . . . . 181
8.2.5 Model 5: Landmark, District, Path, Node, Edge . . . . 183
8.3 The Map Is Not the Landscape . . . . . . . . . . . . . . . . . . . . . . . 187
8.3.1 Mapping Landscape Features . . . . . . . . . . . . . . . . . 187
8.3.2 Size and Scale Dependency in Choropleth Maps . . . 188
8.4 Everything Is Related to Everything Else . . . . . . . . . . . . . . . . 188
8.4.1 Vertical and Horizontal Relations . . . . . . . . . . . . . . 189
8.4.2 Correlation and Coherence . . . . . . . . . . . . . . . . . . . 190
8.4.3 Landscape Heterogeneity and Diversity:
Applications of the Information Theory . . . . . . . . . . 193
8.4.4 Networks: Connections and Fragmentation . . . . . . . 195
8.5 Landscape Metrics and Indicators . . . . . . . . . . . . . . . . . . . . . 200
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
9 Building Blocks of the Landscape . . . . . . . . . . . . . . . . . . . . . . . . . . 209
9.1 A Substrate Carrying Human Life . . . . . . . . . . . . . . . . . . . . . 209
9.1.1 The Terrain as Foundation: The Natural Settings . . . 211
9.1.2 Cultural Building Blocks: The Human Impact . . . . . 214
9.2 People Inhabit the Landscape: Settlements . . . . . . . . . . . . . . . 216
9.2.1 The Settlement Site . . . . . . . . . . . . . . . . . . . . . . . . 218
9.2.2 Principles for a Settlement Typology . . . . . . . . . . . . 220
9.3 People Use the Land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
9.3.1 Organizing Territory and Landscape . . . . . . . . . . . . 232
9.3.2 Small Elements Give a Characteristic Touch . . . . . . 241
9.3.3 People Name Landscapes: Toponyms . . . . . . . . . . . 250
Contents xiii

9.4 Mosaics and Borders: Interactions Between Places . . . . . . . . . 250

9.4.1 Settlement Patterns: Clustered or Scattered? . . . . . . 250
9.4.2 The Multiple Estate . . . . . . . . . . . . . . . . . . . . . . . . 252
9.4.3 Territorial Mosaics . . . . . . . . . . . . . . . . . . . . . . . . . 252
9.4.4 Evolution of Settlement Patterns and Territories . . . . 256
9.4.5 Hierarchy of Settlements . . . . . . . . . . . . . . . . . . . . . 259
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
10 Identifying, Mapping and Assessing Landscapes . . . . . . . . . . . . . . . 265
10.1 From Identification to Monitoring . . . . . . . . . . . . . . . . . . . . . 265
10.1.1 Making Spatial Units . . . . . . . . . . . . . . . . . . . . . . . 266
10.1.2 Typology and Chorology . . . . . . . . . . . . . . . . . . . . 267
10.1.3 Basic Methodological Approaches . . . . . . . . . . . . . . 270
10.1.4 Adding the Third Dimension . . . . . . . . . . . . . . . . . . 274
10.1.5 Dealing with Borders . . . . . . . . . . . . . . . . . . . . . . . 277
10.1.6 A Hierarchy of Landscape Units . . . . . . . . . . . . . . . 278
10.2 Landscape Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
10.2.1 Atlases, Catalogues, Observatories . . . . . . . . . . . . . 280
10.2.2 Generic Traditional Landscape Types . . . . . . . . . . . 284
10.2.3 Mapping New Landscapes . . . . . . . . . . . . . . . . . . . 286
10.2.4 Mapping the Visual Landscape . . . . . . . . . . . . . . . . 287
10.2.5 Mapping the Mindscape . . . . . . . . . . . . . . . . . . . . . 292
10.3 Landscape Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
10.3.1 Attributes, Variables, Indicators and Criteria . . . . . . 294
10.3.2 Assigning Values: What Is Significant
and Important? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
10.3.3 Landscape Character Assessment . . . . . . . . . . . . . . 298
10.4 Landscape Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
11 The Artist’s Landscape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
11.1 Landscape as a Source of Artistic Inspiration . . . . . . . . . . . . . 311
11.2 Landscape Painting – The Imagined Landscape . . . . . . . . . . . 313
11.2.1 Landscapes That Are Not the Landscape . . . . . . . . . 313
11.2.2 The Challenge of Perspective . . . . . . . . . . . . . . . . . 313
11.2.3 Setting the Scene: A Brief Overview of Important
Historical Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
11.3 Making a Statement – From Vision to Landscape Design . . . . 330
11.3.1 Marking the Land . . . . . . . . . . . . . . . . . . . . . . . . . . 330
11.3.2 From Garden Design to Landscape Architecture . . . 330
11.3.3 Principles of Landscape Architecture . . . . . . . . . . . . 332
11.3.4 Dimensions in Landscape Design Styles . . . . . . . . . 336
11.3.5 A Brief Overview of the Succession
of Garden Styles . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
xiv Contents

11.4 Making Places – Evolving Landscape Design Styles . . . . . . . . 359

11.4.1 A Prototype: Vaux-le-Vicomte . . . . . . . . . . . . . . . . 359
11.4.2 A Model: Versailles . . . . . . . . . . . . . . . . . . . . . . . . 364
11.4.3 The Up-scaling to Urban Planning . . . . . . . . . . . . . . 366
11.5 Practices in Contemporary Landscape Architecture . . . . . . . . . 371
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
12 Bringing It All Together – Taking Care of the Landscape . . . . . . . 377
12.1 Speaking for the Landscape . . . . . . . . . . . . . . . . . . . . . . . . . . 378
12.2 Who Is Competent? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
12.3 Landscape Planning Is a Spin-Off from Spatial Planning . . . . . 381
12.4 Planning a Complex and Highly Dynamical System . . . . . . . . 383
12.5 Subsidiarity and Fragmentation . . . . . . . . . . . . . . . . . . . . . . . 383
12.6 Planning at the Landscape Scale – Landscape as Integrating
Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
12.7 Landscape Qualities, Values and Services . . . . . . . . . . . . . . . 387
12.7.1 The Intrinsic Value of Landscape and the Question
of ‘Right’ and ‘Wrong’ . . . . . . . . . . . . . . . . . . . . . . 387
12.7.2 Assigning Landscape Values: Many Decisions
to Make . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389
12.7.3 Respect Our Common Heritage: The Past
Is Important for the Future . . . . . . . . . . . . . . . . . . . 390
12.7.4 Criteria to Assess Holistic Qualities
of the Landscape . . . . . . . . . . . . . . . . . . . . . . . . . . 392
12.7.5 Shaping the Future: Landscape Quality
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
12.7.6 Landscape Services . . . . . . . . . . . . . . . . . . . . . . . . 399
12.8 Principles to Set Goals for the Landscape . . . . . . . . . . . . . . . . 400
12.8.1 Attempt Sustainable Development . . . . . . . . . . . . . . 400
12.8.2 Stimulate Multifunctionality Wisely . . . . . . . . . . . . 401
12.8.3 Reduce All Kinds of Fragmentation . . . . . . . . . . . . . 402
12.8.4 The Endless Feedback Loop Between Functioning
and Spatial Structure . . . . . . . . . . . . . . . . . . . . . . . 404
12.8.5 Interesting Diversity, Safe Order and a Distinct
Character and Identity . . . . . . . . . . . . . . . . . . . . . . 405
12.9 Some Methodological Issues of the Planning Process . . . . . . . 406
12.9.1 Top-Down and/or Bottom–Up? . . . . . . . . . . . . . . . . 406
12.9.2 Visualisation and the Immersion in Virtual
Landscapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
12.9.3 Participatory Mapping: Bringing Knowledge
of Locals and Experts Together . . . . . . . . . . . . . . . . 409
12.9.4 Making a Diagnosis of the Actual Landscape . . . . . . 409
12.9.5 Choosing the Type of Management . . . . . . . . . . . . . 410
Contents xv

12.10 A Strong Forward-Looking Action to Enhance, Restore or

Create Landscapes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 410
12.10.1 Dealing with the Uncertain Future . . . . . . . . . . . . . . 410
12.10.2 Uncertainty, Risk, Hazard and the Precautionary
Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
12.10.3 Scenarios for Future Landscapes . . . . . . . . . . . . . . . 414
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
Chapter 1
The Holistic Nature of Landscape – Landscape
as an Integrating Concept

Abstract The authors use some historical narratives to illustrate how the concept
of landscape implies a holistic nature that intimately links the real tangible world
with its experience in the eye of the beholder. Holism makes that landscape can
serve as an integrated concept between a wide variety of perspectives to study
it. This is the basic idea of the book and is developed in different chapters, each
devoted to these perspectives.

Keywords The discovery and invention of the landscape

1.1 Climbing the Mountain

As a child, native from the Low Countries, born in the city of Ghent, the landscape
appeared to me as a stroboscopic film between gaps in the ribbon building along the
roads of Flanders. The landscape was a flat area filled with rows of coppiced trees
and scattered farms. We didn’t use the word ‘landscape’, rather referred to it as ‘the
outside’ (den buiten in Flemish), meaning everything beyond the city, where
farmers worked and lived and what the English call ‘countryside’. Later, when
travelling around, I took the habit to climb hilltops, searching for lookouts where
the landscape revealed itself more excitingly. These were places of reflection, of
meditation. I saw the landscape as an intriguing book written in a language I didn’t
yet understand. Climbing heights seem an instinctive behaviour and is done for very
diverse motives: to localize oneself better in the environment, to find the way, to
study geography, landscape or nature, to challenge oneself, to admire the beauty of
the scenery, etc.
Scholars who are discussing the landscape as a source of aesthetic contemplation
often refer to the account of Italian poet Petrarch of his ascent of the Mount
Ventoux in France in the fourteenth century (Krebs 2014; Cosgrove 1998; Lemaire
1970). Jakob Burckhardt (1878) declared Petrarch “a truly modern man” and
claimed that he was the first to climb mountains for pleasure. Other scholars
studying landscape used this event as a marker of the new metaphysical world-
view emerging with humanism in the Renaissance and a new awareness and way of
seeing landscape and nature.

© Springer Science+Business Media B.V. 2017 1

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_1
2 1 The Holistic Nature of Landscape – Landscape as an Integrating Concept

I was curious to know what Petrarch thought of the landscape he saw. In a letter,
written around 1350 AD to his former confessor, published in his Epistolae
familiares (IV,1), he describes his ascent on April 26th, 1336 (Sadlon 2007).
Petrarch wrote:
My only motive was the wish to see what so great an elevation had to offer. At first, owing
to the unaccustomed quality of the air and the effect of the great sweep of view spread out
before me, I stood like one dazed.

While describing the panorama, he contemplated his life and described his
I turned my eyes toward Italy, whither my heart most inclined. The Alps, rugged and snow-
capped, seemed to rise close by, although they were really at a great distance; the very same
Alps through which that fierce enemy of the Roman name once made his way, bursting the
rocks, if we may believe the report, by the application of vinegar. I sighed, I must confess,
for the skies of Italy, which I beheld rather with my mind than with my eyes. An
inexpressible longing came over me to see once more my friend and my country. . .

Then, he started reading a passage in St. Augustine’s Confessions he carried with

him, discussing why people wonder about the grandeur of nature and were not
considering themselves. Petrarch was abashed:
“I was satisfied that I had seen enough of the mountain” and turned his “inward eye upon
myself” reflecting on the vanity of human “appetites which spring from earthly impulses.”

I was surprised that Petrarch did not use the word landscape at all. He essentially
described his inner thoughts and feelings engendered by his experience of scenery
and the grandeur of nature, much in the sense of the romanticists later on. No
characterization and no explanation of the landscapes he saw. I discovered that the
historical accuracy of the account is questionable and that Petrarch was certainly
not the first to climb the Mont Ventoux. Jean Buridan (Johannes Buridanus), a
scholar of natural philosophy at the University of Paris, climbed the Mont Ventoux
before 1334 “in order to make some meteorological observations.” (Moody 1942).
This reminded me of Alexander von Humboldt. On June 23rd, 1802, he climbed
with his companions the volcano Chimborazo in the Andes, which was then
believed to be the highest mountain on earth. All the way up to just below the top
– stopped by altitude sickness – he was making measurements of temperature, air
pressure and the blueness of the sky, and collecting rocks and species. Although
they climbed higher than any European recorded in history, in none of von
Humboldt’s writings there was any reflection on this extraordinary and tedious
physical achievement, rather extensive discussions of the measurements were
made, profound analyses and many interpretations, resulting in beautiful a
Naturgem€ alde, literally meaning ‘nature painting’, that gives a holistic synthesis
of the mountain system of the tropical Andes, relating it to similar environments all
over the world (Fig. 1.1).
Andrea Wulf (2015) considers the climbing of the Chimborazo as the moment
von Humboldt “began to see the earth as one great living organism where
1.1 Climbing the Mountain 3

Fig. 1.1 A ‘nature painting’ (Naturgem€ alde) by von Humboldt showing a holistic synthesis of the
Andes mountain system related to similar mountains in the world. The title reads: “Geography of
the Plants in the Tropic Lands: a Naturgem€ alde of the Andes, based on Observations and
Measurements for different variables made from 10 degrees latitude north to 10 degrees south,
in the years 1799–1803, by Alexander von Humboldt and A.G. Bonpland.” The tables on the left
and the right give the observations and measurements according to the elevation. In the right table
there is also an extensive ‘geognostic’ description of the Tropics. On the figure, plant species are
written at the appropriate altitude and slope locations, as well as comparative information about
other observations elsewhere in the world

everything was connected, conceiving an old new vision of nature that still influ-
ences the way that we understand the natural world.” (p. 2). However, setting off for
South America on June 5th, 1799, von Humboldt wrote in a farewell letter his
intentions of this voyage:
I shall try to find out how the forces of nature interact upon one another and how the
geographic environment influences plant and animal life. In other words, I must find out
about the unity of nature. (Nicolson 1995, p. ix)

However, it was more than studying the natural environment, the plants and
animal life. His writings are also full of comments about people, races, languages,
history and politics. Von Humboldt innovative contribution to natural sciences and
(landscape) ecology avant la lettre is generally known, although rapidly
overshadowed by Darwin’s revolutionary theory and later forgotten because of
political reasons. His contribution to humanistic thought, especially concerning the
relationship between culture, history and the environment is less well known (Wulf
2015; Bunske 1981). More than half a century later, in the preface of “COSMOS: A
4 1 The Holistic Nature of Landscape – Landscape as an Integrating Concept

Sketch of the Physical Description of the Universe, Vol. 1”, he elaborated his
This general picture of nature, which embraces within its wide scope the remotest nebulous
spots, and the revolving double stars in the regions of space, no less than the telluric
phenomena included under the department of the geography of organic forms (such as
plants, animals, and races of men), comprises all that I deem most specially important with
regard to the connection existing between generalities and specialities, while it moreover
exemplifies, by the form and style of the composition, the mode of treatment pursued in the
selection of the results obtained from experimental knowledge. The two succeeding
volumes will contain a consideration of the particular means of incitement toward the
study of nature (consisting of animated delineations, landscape painting, and the arrange-
ment and cultivation of exotic vegetable forms), of the history of the contemplation of the
universe, or the gradual development of the reciprocal action of natural forces constituting
one natural whole; and lastly, of the special branches of the several departments of science,
whose mutual connection is indicated in the beginning of the work. (von Humboldt 1858,
p. 11)

His vision was truly holistic avant la lettre (the concept ‘holism’ only emerged in
1920) and modern. But, again, I was surprised as nowhere in his extensive writings
von Humboldt used the word landscape as we do now, although a very concise
definition of landscape was attributed to him and widely copied: ‘Landschaft ist der
Totalcharakter einer Erdgegend’ (Zonneveld 1995). In Ansichten der Natur, von
Humboldt (1807) referred often to the (physiognomic) character of the landscape in
the sense of scenery and in Cosmos he considered landscape painting one of the
three “incitements to the study of nature”. Or, as Ritter put it later: “Landscape is
nature which is one moment present for a feeling and alert observer” (“Landschaft
ist Natur, die im Augenblick f€ ur einen f€ uhlenden und empfindenden Betrachter
a€sthetisch gegenw€ artig ist” – Ritter 1989).
Climbing the mountain is more than reaching a top and having a nice
all-encompassing prospect of the landscape. Climbing is a process starting at the
bottom and preparing the ascent often with the help of local people. Before having a
bird’s eye perspective, one is in the landscape interacting with others living there.
Exploring the landscape is zooming in and out, and is about changing scales and
perspectives all the time. Thus, the landscape becomes the whole that is more than
the sum of its composing parts. It integrates nature and culture, science and
aesthetics, geography and history, humans and their environment at all scales and
in all aspects. The landscape is where natural and social sciences integrate, a place
where space and time meet. A fascinating, but complex phenomenon to study, and
that is the challenge we take in this book.
1.2 The Landscape Is in the Eye of the Beholder? 5

1.2 The Landscape Is in the Eye of the Beholder?

Clearly, the landscape has to do with the geographical environment in a holistic

sense but is also a source of aesthetic experience and artistic expression. Von
Humboldt emphasised the importance of the aesthetic ‘incitement’ that landscapes
arise. In a footnote of his preface to Cosmos, he explained:
In order to depict nature in its exalted sublimity, we must not dwell exclusively on its
external manifestations, but we must trace its image, reflected in the mind of man, at one
time filling the dreamy land of physical myths with forms of grace and beauty, and at
another developing the noble germ of artistic creation. (Cosmos, Vol. I, footnote 13).

A commonly used quote in relation to landscape aesthetics and valuation is

“beauty is in the eyes of the beholder”, paraphrasing the British philosopher David
Hume “Beauty is no quality in things themselves: It exists merely in the mind which
contemplates them; and each mind perceives a different beauty” (Hume 1758).
Although the controversy of the expression (Jóhannesdóttir 2016), the same idea is
found implicit in the formal definition in the European Landscape Convention:
“Landscape means an area, as perceived by people. . ..”. This leads to the debate of
landscape as an objective or subjective phenomenon and of an object of study in
natural and human sciences.
However, perceiving is much more than seeing. “A glance is different from a
stare, a sight is different from a vision”, said Denis Cosgrove (2002) and “most
languages make a fundamental distinction between seeing and looking. The former
suggests the passive and physical act of registering the external world by eye; the
latter implies an intentional directing of the eyes towards an object of interest.” The
“ways of seeing” (Cosgrove 2002) determine whether the landscape is studied as an
object in natural sciences, as a social construct, or as an aesthetic object. Each of
these ways has its proper definitions, vocabulary and methods. Each way demands
proper skills and specialization.
Most interesting is that perception, aesthetics and artistic expressions also have a
holistic nature, similar to the ‘ecological’ holism of landscape. Often the German
concept of Gestalt is used in this context. Christian von Ehrenfels introduced the
notion of “Gestalt” in psychology in his essay “On Gestalt qualities” (1890)
(Wagemans 2015). He argued that, because we can recognise two melodies as
identical even when no two notes in them are exactly the same, a melody must have
a “Gestalt quality” that is immediately given and which is more than the sum of the
composing elements. Between the two World Wars, visual “Gestalt-principles”
were formulated to explain how we interpret and give meaning to complex patterns,
and thus helping the learning process.
6 1 The Holistic Nature of Landscape – Landscape as an Integrating Concept

1.3 Dealing with “The Whole That Is More Than the Sum
of its Parts”

Both holism and the Gestalt-principles lost their popularity soon after their intro-
duction, in particular when rapid technological innovations transformed scientific
research. This occurred in geography and ecology, allowing quantification and
modelling, which were considered a more objective scientific approach. After the
Second World War, research disciplines specialised and the divergence between
natural and human sciences continued. Landscape studies became considered
old-fashioned and the interest in the landscape was gradually lost and so the holistic
As a geographer, doing ‘landscape science’, embedded in regional geography, it
was a difficult time. Important choices had to be made. In some institutions,
landscape studies were found in the natural sciences and in other in the human
and social sciences. Landscape studies were fragmented over very different disci-
plines: physical geography, ecology, regional geography, historical geography,
archaeology, humanistic geography, sociology, psychology and planning.
The choice between a holistic and a reductionist approach was then a choice
between ‘soft’ and ‘hard’ sciences, depending also on the methods and technologies
that were available. Aerial photography was popular and offered a new, holistic,
bird’s eye perspective on the landscape, which made Carl Troll say, “air photo
interpretation is to a large extent landscape ecology” (Troll 1939). Jan Zonneveld
recognised two approaches to “landscape as a whole” (Zonneveld 1999). One was
pragmatic and consisted in subdividing landscape hierarchically in spatial units,
which were treated as “black boxes”, allowing to make meaningful assessments
even when not all the parts could be studied in detail. This “pragmatic-holism”, as
Ies Zonneveld called it, proved to be very useful in land evaluation (Zonneveld
2005). The other way was to model landscape composed of thematic layers using
data from very diverse disciplines. The aim was, after the analysis of the themes, to
recompose the “whole” in various ways. Later, the growing computer power and
the general availability of digital data allowed geographical information systems to
achieve the task and to perform complex spatial analysis permitting new forms of
reductionist approach and creating new kind of landscape visualisations.
Considering the landscape as a holistic, dynamic and hierarchical system has
often been criticized being a naive and unrealistic fantasy (Naveh 2000). Never-
theless, this approach proved to work, while highly specialised disciplines were not
able to deal with the increasing complexity and faster transformation of our world.
The only solution lies in a transdisciplinary approach where all actors are involved.
The landscape is becoming an integrating concept. Understanding landscape is
about changing perspectives and scales.
1.4 The Structure of the Book 7

1.4 The Structure of the Book

An attempt to make a comprehensive overview of the diverse approaches to

landscape in a holistic perspective is a risky endeavour. Not only because of the
multiple meanings the concept landscape has and the many approaches that are
possible, but also because of the growing number of actors involved and the
increasing speed of landscape changes and attitudes. The risk to miss new ‘ways
of seeing’ is real. We have to apologize for these gaps and missing links in advance.
Instead of converging to a holistic synthesis, landscape research and policy are
widening and diverging like branches of a fast growing tree. How to avoid we can’t
see the forest for the trees? How to integrate “generalities and specialities”?
The primary holistic nature of landscape will be used as the leading principle in
the discussion of the different approaches, risking that we leave out some of the
depth and detail each of them can offer.
Each chapter deals with one approach or theme and is made as a stand-alone
unit, with proper abstract and references. However, none of the approaches is
completely distinct from another. To guarantee the coherence of the narrative
within a chapter, we kept some overlap and hence some repetition may occur
between different chapters.
Speaking and writing about landscape proceeds sequentially, while a holistic
phenomenon demands parallel processing. Therefore we choose to use visual
language besides written language. This is why there are so many figures with
extensive captions in the book.
Landscape research s.l. has a long history, which reflects the different
approaches and disciplines involved. Chapter 2 gives an overview of this history.
We follow the development of the European culture, as this shows clearly how the
concept landscape became a subject of scientific research, which today is global.
The history shows how a multitude of related concepts emerged, adding to the
complexity. It also shows the impact of technology on the creation of new “ways of
seeing” and new modes of dealing with the landscape.
Today, we use formalised definitions of the word landscape in research and in
particular in policy. In linguistic terms, the word landscape has multiple meanings
and shows subtle differences between tongues. Chapter 3 explores the etymology
and linguistic properties of the word in order to understand each other better. Also
here we start from the origins of the word landscape in European languages, as the
root of the widely used and globalised scientific concept landscape today. Although
English is nowadays dominant in scientific research, problems in translation and
understanding with other landscapes are still important.
The complexity of landscape and its multiple meanings make that different
approaches are not only possible but also necessary. Chapter 4 gives an overview
the different disciplines studying the landscape. Each of them focuses on a more or
less specialized view, deepening the expertise and knowledge, but often loosening
the holistic perspective. Hence, there is a growing need for inter- and transdisci-
plinary approaches.
8 1 The Holistic Nature of Landscape – Landscape as an Integrating Concept

Many concepts related to landscape as a complex system are used as well. Some
are essential to understand the holistic nature of the landscape and define some
specific analyses to describe holistic properties. The most common ones are
presented in Chap. 5.
The landscape is intimately linked to an observer. The landscape is sensed and
perceived, and thus creates a psychological realm. The landscape becomes a
mindscape and a social construct and this influences how we interact with the
external physical world. It determines our attitude towards the landscape and the
environment. Chapter 6 explores the landscape perception and experience of
landscape by humans.
Chapter 7 looks at the landscape as a dynamic phenomenon. Landscapes evolve
and have a history linked to the continuous transformation of nature, society and
culture. Landscape is a palimpsest showing fragments of different superimposed
time layers. Change is an essential property of the landscape. It implies adaptation
and hence the need to understand processes and actions. A challenge nowadays is
how to handle the increasing pace and magnitude of the contemporary changes that
also transform values and functions.
Chapters 8, 9 and 10 discuss three ways to reduce the complexity of the
landscape, attempting to analyse “the whole that is more than the composing
parts”, to understand its heterogeneous composition and hierarchical structure.
Maps and even better aerial photography reveal spatial patterns that configure
landscapes in various characteristics ways. Analysing landscape patterns developed
into an important branch of landscape research introducing landscape metrics and
several techniques to describe different types of patterns. Chapter 8 gives a sum-
mary of the attempts of describing holistic characteristics.
Landscapes are geographical spaces shaped by landform and filled with objects
of very different nature. Chapter 9 discusses these building blocks of the landscape
at three spatial scales. Settlements and their initial territories are presented here as
holons that define different landscape types and regions.
Geographical space is characterised by mosaics of different landscape types
organised in a hierarchical manner according to scale. Some landscape types occur
in different spaces and their context varies accordingly. Chapter 10 presents the
principles and main methods of making landscape classifications and evaluations.
Chapter 11 discusses the artist’s landscape, i.e. landscapes that are conceived
mentally, created through visual representations or landscaping. These are the
domains of landscape painting, historically closely associated with landscape
architecture. This chapter focuses on representing, designing and constructing
landscapes from impressions, emotions and visions, often with the intentions of
making some kind of statement.
Chapter 12 deals with landscape policy, planning and management, aiming to
integrate all necessary knowledge for conserving, restoring, enhancing and creating
landscapes. The landscape is seen the common heritage of humanity and important
for human well-being. What qualities and values are important and how to assess
these in participation process. The main questions are: who speaks for the land-
scape, what are the different competences involved and how to formulate goals in a
transdisciplinary process?
References 9


Bunske, E. V. (1981). Humboldt and an aesthetic tradition in geography. The Geographical

Review, 71(2), 127–146.
Burckhardt, J. (1878). The civilization of the renaissance in Italy. Translated by S. Middlemore,
Penguin Classics, 1990, 416 pp.
Cosgrove, D. (1998). Social formation and symbolic landscape. Madison: Wisconsin University
Cosgrove, D. (2002). Landscape and the European sense of sight – Eyeing nature. In K. Anderson,
M. Domosh, S. Pile, & N. Thrift (Eds.), Handbook of cultural geography, Chapter 12
(pp. 249–268). London: Sage.
Hume, D. (1758). Of the standard of taste. Essay XXVI in essays and treatises on several subjects.
London: A. Millar. edition 1777.
Jóhannesdóttir, G. R. (2016). Landscape and aesthetical values: Not only in the eye of the
beholder. In K. Benediktsson & K. A. Lund (Eds.), Conversations with landscape, Chapter 8
(pp. 109–123). London: Routledge.
Krebs, A. (2014). Why landscape beauty matters. Land, 3, 1251–1269.
Lemaire, T. (1970). Filosofie van het landschap (p. 224). Bilthoven: Ambo.
Moody, E. A. (Ed.). (1942). Iohannis Buridani Quaestiones super libris quattuor De caelo et
mundo. Cambridge, MA: Medieval Academy of America.
Naveh, Z. (2000). What is holistic landscape ecology? A conceptual introduction. Landscape and
Urban Planning, 50, 7–26.
Nicolson. (1995). A Historical Introduction. In: Alexander von Humboldt (ed.), Personal Narra-
tive of a Journey to the Equinoctial Regions of the New Continent. Abridged and Translated by
Jason Wilson and a Historical Introduction by Malcolm Nicholson, London: Penguin Books,
pp. 311.
Ritter, J. (1989). Landschaft. Zur Funktion des Ästhetischen in der modernen Gesellschaft. In:
J. Ritter (ed.), SubjektivitÄt. Sechs AufsÄtze. Frankfurt am Main. pp. 141–163.
Sadlon, P. (2007) The Ascent of Mont Ventoux. Translated in English. http://petrarch.petersadlon.
Troll, C. (1939). Luftbildforschung und Landeskundige Forschung. Erdkundliches Wissen.
Schriftenreihe für Forschung und Praxis, Heft 12, F. Wiesbaden: Steiner Verlag.
von Ehrenfels, C. (1890). On ‘Gestalt Qualities’. English translation of “Uber €
‘Gestaltqualit€ aten’”., Vierteljahrsschrift fur wissenschaftliche Philosophie, 14, 1890,
von Humboldt, A. (1807). Ansichten der Natur. eBook-Original ausgabe, eClassica, Aurabooks.
von Humboldt, A., 1858. Cosmos – a Sketch of the Physical Description of the Universe. vol.
1, Translation Otte, E.C. Baltimore/London: John Hopkins University Press.
Wagemans, J. (2015). Historical and conceptual background: Gestalt theory. In J. Wagemans
(Ed.), Oxford handbook of perceptual organization. New York: Oxford University Press.
Wulf, A. (2015). The invention of nature – Alexander von Humboldt’s new world (p. 473).
New York: Alfred A. Knopf.
Zonneveld, I. S. (1995). Land ecology (199 pp). Amsterdam: SPB Academic Publishing bv.
Zonneveld, J. I. S. (1999). Landscape synthesis in monitoring global change. In: Moss, M. R., &
Milne, R.J., Landscape synthesis. Concepts and applications. Landscape system analysis in
environment management. Working Group of the International Association for Landscape
Ecology, 1–10.
Zonneveld, I. S. (2005). The land unit as a black box: A Pandora’s box? In: Wiens, J.A., & Moss,
M.R. (eds.), 1999. Issues in landscape ecology. Studies in Landscape Ecology (pp. 331–345).
Cambridge, MA: Cambridge University Press.
Chapter 2
The History of Landscape Research

Abstract The study of the landscape as an object of research started in Europe

during the Renaissance and the Age of Discovery. The first pictorial representations
of landscapes emphasized its visual character and scenery. A new style of garden
design and urban planning emerged. The scientific research of landscape started
with the systematic descriptions during the naturalistic explorations. Landscape
became a core topic in geography and was seen as a synthesis between interacting
natural and cultural components of a region defining its characteristic identity.
Different national schools in landscape research developed. After the First World
War, aerial photography gave a completely new perspective on the landscape
revealing its holistic character. Consecutive specialisation in science resulted in
the loss of the holistic synthesis. At the end of the twentieth century, landscape
research followed very different approaches and interdisciplinary integration was
lacking. Increasingly faster environmental changes became manifest in the land-
scape, putting the landscape on the political agenda. International conventions
stimulated transdisciplinary approaches involving public participation in the
twenty-first century. N-grams of the corpus of five million digitized books by
Google, allow the analysis and evolution of landscape related terms since the
seventeenth century.

Keywords Landscape research • Landscape disciplines • Landscape science •


2.1 Introduction

This chapter describes the history of the use of the word ‘landscape’ since its origin
in western culture and the emergence of its different meanings, as well as the
research that developed around the concept. Six phases are recognized. Next, the
occurrence of the word ‘landscape’ and related concepts in the corpus of digitized
books by Google since the seventeenth century, using the N-grams viewer.

© Springer Science+Business Media B.V. 2017 11

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_2
12 2 The History of Landscape Research

2.2 A Complex Story

The history of – what we now call – landscape research followed different paths and
several important conceptual changes occurred. To understand this development we
must consider the multiple meanings the word landscape has in different languages,
as well as the shift in these meanings in relation to the evolving society and
technology. As the etymologic roots of the word landscape are found in Western
Europe, the perspective of this review will start here. It is the path that leads to the
development of scientific and professional disciplines regarding the landscape.
Following consecutive phases can be recognised (Antrop 2013):
1. the early beginnings: the proto-scientific phase (before the nineteenth century)
2. the emerging scientific research: the landscape as object of study of geography
(nineteenth-early twentieth century)
3. landscape from the air: aerial photography with its holistic perspective and the
emergence of historical geography (approximately 1930–1960s)
4. the loss of synthesis (approximately 1960–1980s)
5. the humanistic approach and the revival of landscape ecology (approximately
6. The ‘landscape crisis’ and the shift towards applied and transdisciplinary land-
scape studies (since approximately 1990)
Figure 2.1 gives a graphical overview of the history of landscape research from
the perspective of western culture where it originated. It places ideas, concepts,
disciplines, methods and technology and exemplary key people and networks on a
time line. No geographical differentiation is attempted to show regional differences.
Different typographic styles represent these different aspects and are explained at
the bottom of the figure. Bold numbers on the left indicate different phases that can
be recognized. Arrows indicate different parallel development paths leading to the
actual research domains: (A) garden design – landscape architecture – landscape
design, (B) academic disciplines, and (C) landscape as heritage protection, conser-
vation and management.

2.3 The Early Beginnings

Dealing with the landscape as an object of study started in Europe during the
Renaissance and the Age of Discovery. In the fifteenth century appeared the first
pictorial representations of landscapes, emphasizing the importance of its visual
character and scenery. These paintings are the only pictorial information source
about the landscapes of that period, but cannot be regarded as photographic
registrations. The landscape was used as an expression of human ideas, thoughts,
beliefs and feelings. The creation of imaginary landscapes appeared almost simul-
taneously with a new style of garden design and urban development and lifestyle.
2.4 The Emerging Scientific Research – The Landscape as Object of Study of. . . 13

Fig. 2.1 The history of landscape research (explanation see text) (After Antrop 2013)

Garden architecture and urban planning made a branch of practitioners from which
contemporary (landscape) architecture and town planning developed.
Simultaneously, the discovery of new worlds demanded new methods for
describing and depicting in a systematic ‘scientific’ way the exotic landscapes
and people. New techniques were developed such as cartography.

2.4 The Emerging Scientific Research – The Landscape

as Object of Study of Geography

The scientific research of the landscape initiated however from philosophers of the
Enlightenment, in particular with the lectures Immanuel Kant (1724–1804) gave
between 1756 and 1796 on (physical) geography and anthropology, disciplines he
considered ‘pragmatic knowledge of the world’ (Weltkenntnis) (Elden 2011).
The empirical research of landscape started with the systematic descriptions
during the naturalistic explorations, such as the ones made by Alexander von
Humboldt (1769–1859) and Charles Darwin (1809–1882). These explorations,
and many others, offered data and methods for empirical landscape research. A
short and very concise definition of landscape was attributed, but not proven, to
Alexander von Humboldt: ‘Landschaft ist der Totalcharakter einer Erdgegend’
14 2 The History of Landscape Research

(Zonneveld 1995). This definition implies that landscapes express regional diver-
sity and that landscape is a holistic phenomenon, perceived by humans. Although
von Humboldt was a pioneer in biogeography, physical geography and climatology,
he always stressed in his writings the human, historical and cultural aspects in the
landscape and above all the aesthetical qualities, which he considered even as
mentally healing (Bunske 1981; Nicholson 1995). In these early days of natural
sciences, the distinction between different disciplines was vague and methods were
mostly descriptive, which made Alexander von Humboldt say:
In classical antiquity the earliest historians made little attempt to separate the description of
lands from the narration of events the scene of which was in the area described. For a long
time physical geography and history appear attractively intermingled. (von Humboldt

The growing importance of the landscape concept in Europe from the

mid-nineteenth century cannot be understood without considering the development
of botany and zoology, history and geography, and geographical education in
particular, and the context of many important changes related to the
industrialisation, the exploration of new territories, often colonies, the economic
globalisation and the formation of nation-states.

2.4.1 The Societal Context

The explorations raised also the interest of the broad public for nature, landscape
and geography. In 1830 the Royal Geographical Society (RGS) was founded in
Britain and in 1888 the National Geographical Society (NGS) in the USA. The
colonization and industrial revolution, and many associated processes, such as
urban sprawl, the enclosure of common land and the ‘agricultural invasion’ by
the import of new products, up-scaling and mechanisation, created new landscapes
that erased existing ones. Landscape became also popular in arts, in particular in
painting and gardening. The American Society of Landscape Architects (ASLA)
was already founded in 1899. Scenic and symbolic meanings became more impor-
tant and the idea of ‘national landscapes’ symbolising the identity of the nation
emerged (Schama 1995; Olwig 2002), as well as the concept of a ‘national park’.
Around the beginnings of the twentieth century, the loss of nature and encroach-
ment on the countryside initiated movements of protection of monuments, sites,
nature and landscapes in most Western countries. Landscape became accepted as
common heritage and laws for protecting it were issued. Exemplary is the founda-
tion of the National Trust (NT) in 1895 in Britain.
These profound changes were reflected in education and research. In Britain the
geographical education was reformed in the perspective of the growing demand for
knowledge in natural sciences, the colonies and Britain becoming the most impor-
tant economic power in the world. ‘Geographical literacy’ was popular in education
and social consciousness of the latter half of that century (Maconochie 1989).
2.4 The Emerging Scientific Research – The Landscape as Object of Study of. . . 15

Landscape, however, was not specially emphasized. New educational principles

became important, in particular the ones of Pestalozzi (Elliott and Daniels 2006).
Geographical education had to start from direct terrain observation, i.e. studying the
landscape of the surroundings, and gradually extend to broader, more abstract
scales. This inspired also the development of a pedagogic methodology and it is
significant that Carl Ritter dedicated the first volume of Die Erdkunde im
Verh€altniss zur Natur und zur Geschichte des Menschen (Geography in Relation
to Nature and the History of Humankind, 19 volumes written between 1816–1859)
to Pestalozzi (Leopold et al. 1971). In France, the defeat in the Franco-Prussian war
of 1870–1871 was partially ascribed to the poor geographical knowledge and map
reading skills of the military, and resulted in an educational reform supporting the
“école française de géographie” of Vidal de la Blache, then a geography teacher at
the E´cole normale (Boulanger 2002).

2.4.2 The Landscape Concept in National Geographical


Most theoretical thinking about geography and landscape in the academic world
happened in Germany from the mid-nineteenth century to the Second World War
(Wardenga 2006). From the mid-1880s, a movement developed for the evaluation
of the purely descriptive regional geography leading to what is known now as the
typically German L€ anderkunde (‘knowledge of the lands’) as fostered by Alfred
Hettner (1859–1941). The purpose was to profile geography as a chorological
spatial science and to create up-to-date images of the world in the perspective of
the nation-state debate, the significance of colonies, both important in an industrial
society becoming increasingly globalised. L€ anderkunde boomed and changed with
the outbreak of the First World War to provide knowledge useful in the political
debates concerning territorial claims. Wardenga points to the significant role of the
“so-called Landschaft (landscape) concept” in this shift of emphasis. The concept
Landschaft was hitherto already in use in school education as an easy introduction
to more abstract geography at a wider scale, including L€ anderkunde. Indeed,
Landschaft meant “a specific section of the earth’s surface which could be per-
ceived as a harmonious whole consisting of different natural and anthropogenous
factors.” (Wardenga 2006). The term was also familiar in everyday language, which
made it easy to communicate in the new context. The landscape was seen a real
“spatial entity, a harmonious, individual whole, an absolute coherence, as an
organism.” (Wardenga 2006). Clearly, at the end of the nineteenth century, German
geographers began to use landscape studies as an alternative to von Richthofen’s
‘chorography’ and ‘chorology’.
While Hettner became the leading theoretician of the L€ anderkunde, it was Alwin
Oppel (1849–1929), also a German geographer, who introduced in 1884 the term
‘Landschaftskunde’ (‘landscape science’) (Troll 1950). Theoretical concepts and
16 2 The History of Landscape Research

mainly descriptive methods of this ‘Landschaftskunde’ were developed mainly in

Germany and Scandinavia. The prominent Friedrich Ratzel (1844–1904),
influenced by Darwin, made the foundations of human and political geography
introducing ideas such as Lebensraum (‘living space’, territory), the organic state
theory and environmental determinism. He has been credited to introduce about
1890 the term Kulturlandschaft as the part of the initial Naturlandschaft, which was
transformed by human activity. His ideas were certainly popular amongst German
scholars and politicians but also influenced foreign geographers such as the French
Vidal de la Blache and the American Ellen Churchill Semple. For the methodology
of landscape science, the work of Siegfried Passarge (1866–1958) was important.
He wrote a three volumes manual Die Grundlagen der Landschaftskunde (Passarge
1919/1921), followed by Vergleichende Landschaftskunde (Passarge 1921/1930),
which systematically describe mainly physical landforms.
In all European countries, the nationalist thinking influenced the development of
science and arts, which is also reflected in the geographical landscape studies,
topographical maps and paintings (Cosgrove 2004). In Germany, the strong links
of geography with the nationalistic politics and Nazi ideology and its questionable
scientific praxis, lead, after the collapse of the Third Reich, ultimately to the
discrediting of the discipline and the L€ anderkunde and Landschaft in particular
(Wardenga 2006).
In France, Paul Vidal de la Blache (1845–1918) was influenced by Friedrich
Ratzel (1844–1904) but developed his own approach, which later on was called
‘possibilism’ in opposition to the environmental determinism of Ratzel. He stated
that the environment sets limitations and offers opportunities (‘possibilities’) for
cultural development. His methods were based on a combination of cartography
and field observations using descriptions and annotated sketches, in particular
concerning the geology and geomorphology, as well on texts for the historical
aspects. The synthesis was achieved in vast regional monographs. Characteristic
was the recognition of the importance of the local community and the lifestyle of
the inhabitants (‘genre de vie’) in organising the landscape (‘paysage’), which
resulted in a regional differentiation not only based upon natural conditions but
also upon culture, traditions, settlement patterns and social territories. He consid-
ered the landscape as a holistic unity, which was expressed in a characteristic ‘pays’
(‘land’) (Claval 2004). The description of regions became synthetic ‘tableaux’ of
idealistic landscapes. The essence of his approach is well summarised in a paper he
presented at the Ninth International Congress of Geography in Geneva:
Since the educational geography left the cabinet office where it locked itself too voluntarily
up and started to observe nature directly, the interpretation of landscapes became one of its
main topics. [. . .] The analysis and synthesis both have a role. The analysis attempts to
recognise heterogeneous characteristics, which make the landscape composition; and as
past and present causes interfere in relief forms, this kind of interpretation is somewhat an
exegesis. But otherwise this landscape forms a whole of elements that enchain and
coordinate; its interpretation demands rational perception of the living synthesis that is
presented to our eyes. (Vidal de la Blache 1908).
2.4 The Emerging Scientific Research – The Landscape as Object of Study of. . . 17

The ‘interpretation demanding rational perception’ resulted in idealised land-

scapes and criticism on Vidal de la Blache’s work was that he often filtered
‘disturbing’ new elements out. His ‘tableaux’ emphasised the coherence of the
landscape as generated by natural conditions and human activities. The influence of
Vidal de la Blache was important not only in France, where the ‘vidalien’ school
remained important until the 1960s, but it influenced also Mediterranean landscape
studies and studies in Britain.
The Finnish geographer Johannes Gran€o (1850–1913) took a somewhat partic-
ular position. In his book Reine Geographie (Gran€o 1929), he made the distinction
between the ‘Nahsicht’ and the ‘Fernsicht’ or ‘Landschaft’. The ‘Nahsicht’ (‘prox-
imity’) is the surroundings that can be experienced by all senses, while the
‘Landschaft’ is the part that is mainly perceived visually. He developed descriptive
methods for the study of both, creating systematic typologies. He was also a pioneer
in photography and introduced this technique of recording in natural sciences,
mastering it as an artist (Jones 2003). Working in Estonia and Finland, most of
his work remained unknown until his book was translated into English as ‘Pure
Geography’ in 1997 (Gran€o and Paasi 1997).
Also, very little work was translated of the extensive landscape research in the
former Soviet Union and Eastern European. Most remained unknown for the
landscape researchers in the Western Europe and America until recently (Pedroli
1983; Wiens et al. 2007). In general, the concept of landscape was related to the
German concept ‘Landschaft’, in particular to the physical landscape. Also, con-
cepts as typology, biogeoecology, geobotany and phytotopology and phytocoenosis
were commonly used in the context of landscape research, which makes the
approach more appropriately landscape ecology (Pedroli 1983). Pioneering work
was done in the field of hierarchical landscape classification, referred to as typology
and topology (chorology) (Vinogradov 1967; Sochava 1971). Concepts and
methods of these approaches were used later in land evaluation (Zonneveld 1995).
Ellen Churchill Semple (1863–1932) introduced Ratzel’s environmental deter-
minism in the USA. However, it was Carl Ortwin Sauer (1889–1975) who devel-
oped the (German) concept of the landscape as a territory. He made it the corner
stone of the cultural geography as presented his famous article The Morphology of
Landscape (Sauer 1925). His focus is primarily on geography:
The term ‘landscape’ is proposed to denote the unit concept of geography, to characterize
the peculiarly geographic association of facts. Equivalent terms in a sense are ‘area’ and
‘region’. Area is of course a general term, not distinctively geographic. Region has come to
imply, to some geographers at least, an order of magnitude. (Sauer 1925)

However, Sauer sees the landscape as a holistic entity and as a complex system:
The objects which exist together in the landscape exist in interrelation. We assert that they
constitute a reality as a whole that is not expressed by a consideration of the constituent
parts separately, that area has form, structure, and function, and hence position in a system,
and that it is subject to development, change, and completion. (Sauer 1925).

Also, the landscape is a mental construction based on successive observations

from different angles:
18 2 The History of Landscape Research

. . .landscape is not simply an actual scene viewed by an observer. The geographic land-
scape is a generalization derived from the observation of individual scenes. (Sauer 1925).

This ‘geographic landscape’ is very similar to the ‘tableaux’ of Vidal de la

Blache. Cosgrove (2004) noted that “Sauer added to his essay a brief section
referring to the ‘aesthetic’ dimension of landscape, in which he claims that however
analytic and comprehensive the formal study of landscape morphology might be,
there will always be a dimension of landscape that lies ‘beyond science’, and which
cannot be approached through formal study but only via avenues of poetry and art.”
(Cosgrove 2004). This dimension would give rise to humanistic approaches in
landscape studies.
In his notorious work The Nature of Geography, Richard Hartshorne (1939,
1961) considered landscape as a territorial concept too confusing and redundant
with concepts as region and space, which he considered preferable alternatives
(Muir 1999). However, Sauer’s vision resulted later in the first important sympo-
sium on human impact on the environment in a global perspective. The symposium
resulted in a voluminous book Man’s Role in the Changing Face of the Earth
(Thomas 1956), which not only gave a comprehensive overview but also was
remarkable for its interdisciplinary approach.
Summarising, the landscape became a core topic in geography and was seen as a
unique synthesis between interacting natural and cultural components of a region,
making its characteristic identity. To study the landscape, information was gathered
from field observations, maps, literature, sketches and later terrain photographs.
Methods were developed for the detailed description of landscape elements and for
making typologies and idealised models. Theoretical debates about the nature of
landscape became important in the first half of the twentieth century, in particular in
Germany. Different national schools developed, with different emphasis on natural
or cultural landscape, on territory and community, on chorology (region) and
chronology (history).

2.5 Landscape from the Air: Aerial Photography

and Historical Geography

After the First World War, aerial photography gave a completely new approach to
the study of landscape. The bird’s eye perspective revealed clearly the holistic
character of landscape. Complex patterns became visible reflecting hierarchies of
spatial scales, suggesting that multiple processes were involved. This made Carl
Troll say that ‘Luftbildforschung ist zu einem sehr hohen Grade
Landschafts€ okologie’ (‘air photo interpretation is to a large extent landscape ecol-
ogy’) (Troll 1939), and thus introducing the term landscape ecology, which he
called an ‘Anschauungsweis’ (‘a way of seeing’). Aerial photography also opened a
new view on our past as many unknown archaeological and historical features were
2.6 The Loss of Synthesis 19

detected, giving a boost to historical geography and initiating landscape

After the Second World War, landscape research was still mainly descriptive,
resulting in regional monographs, mainly as the result of doctoral theses. The
emphasis was on landscape classification (chorology and typology), on landscape
genesis, both natural and historical, and landscape as the basis for the regional
identity. In this context, the Permanent European Conference for the Study of the
Rural Landscape (PECSRL) was created in 1957 and is the oldest organized group
of landscape researchers in Europe (Helmfrid 2004). In the same period, landscape
architects organized themselves in an international, professional federation (IFLA
founded in 1948).
In the UK and Ireland, the focus was more on the archaeology and history and
historical geography became important with the pioneering work of Henry Clifford
Darby. The interest for the landscape grew faster and became more important for
the general public than for academic scholars (Taylor 2006). A milestone was the
book The Making of the English Landscape by William Hoskins in 1955 (Hoskins
Nature protectionists also gained interest in landscape and its protection became
their mission. Soon protected natural areas were embedded in larger environments,
such as the ‘Areas of Outstanding Natural Beauty’ (AONB) in Britain (1956), the
‘Naturparke’ in Germany (1957) and the ‘Parcs Naturels Régionaux’ in France

2.6 The Loss of Synthesis

Continuous specialisation in science and the introduction of quantitative techniques

changed research profoundly in the 1960s–1970s. Most important was the ‘new
orientation’ in geography, aiming at more explanation based on theory and model-
ling. New techniques of spatial analysis laid the foundation of geostatistics.
Regional geography and landscape studies became old-fashioned and were dropped
as object of study by most geographers, and in many countries geography split in
social and physical geography. Jan Zonneveld called it the ‘gap in geography’ and
predicted that other new scientific approaches would regain the lost synthesis
(Zonneveld 1980).
The collapse of the status of geography in Germany, in particular in Western
Germany, lead to a crisis in the ‘Landschaftskunde’ with endless theoretical dis-
cussions about definitions, losing all societal significance of the discipline. Paffen
(1973) summarized this in Das Wesen der Landschaft. Meanwhile, the theoretical
basis for landscape science continued to develop in Eastern Europe, as
‘Landschaftslehre’ (Neef 1967).
The ‘gap in geography’ was rapidly filled and landscape research took off again
from different sources. In 1972 the ‘Working Community Landscape Ecological
Research’ (‘Werkgroep Landschapsecologisch Onderzoek’, WLO) was founded in
20 2 The History of Landscape Research

The Netherlands in an attempt to restore landscape as a concept of synthesis and to

promote interdisciplinary research. The WLO grouped geographers, biologists,
ecologists, agronomists, planners and several other disciplines. It launched the
journal Landschap (Zonneveld 2000). Another approach to landscape research
came from historical geographers and archaeologists. Work in the UK was impor-
tant here, such as the series publications of The Domesday Geography edited by
Darby (Darby and Campbell 1962; Darby and Welldon Finn 1967; Darby 1971;
Darby and Terrett 1978; Darby and Maxwell 1978). In 1967 the Landscape
Research Group (LRG) was founded, publishing the journal Landscape Research.
In the framework of the Permanent European Conference for the Study of the Rural
Landscape (PECSRL), the first important syntheses at a European scale were
realised, such as a common terminology (Uhlig and Lienau 1972) and an overview
of field systems and settlement forms (Lebeau 1979).
Simultaneously, a philosophical and humanistic approach to landscape emerged
from the Berkeley-school in the USA and from several British geographers. They
emphasised the importance of landscape perception and landscape as a social
construct with narratives and symbolic meanings. Important exemplary publica-
tions are Topophilia (Tuan 1974) and Past time, present place: Landscape and
memory (Lowenthal 1975).
The general settings of the chorology, typology and genesis of the traditional
rural landscapes were already in place before the 1970s when the scientific interests
shifted. From the 1970s on, satellite remote sensing offered again a new perspec-
tive, which offered – forced by the low resolution of the first generation satellites – a
small scale and more global view on the world with an emphasis on the land cover,
environment and processes of change. Important was also the new data format that
came with digital raster images. Towards the end of the 1970s, the fast development
of computers made applications of statistical modelling possible and pattern rec-
ognition and image classification remapped the landscapes mainly based upon land
cover. It took 20 years before the power of personal computers and the lowered
price of detailed digital data made a fully quantitative analysis of landscape patterns
possible, at the same time enhancing the loss of holistic synthesis.
The economic recession, consecutive energy crises and increasing environmen-
tal problems made it clear that the problems became too complex to be handled by
non-concerted actions of different specialized disciplines (Moss 1999). Environ-
mental impact assessment (EIA), including landscape, was first enacted in the USA
in 1969. It stimulated the development of new methods for studying the landscape,
such as the Leopold matrix, which attempted a qualitative expert assessment
(Leopold et al. 1971). It lasted until 1985 before the European Union introduced
an Environmental Impact Assessment Directive, which included ‘landscape and the
(visual) surroundings’ as one aspect to be studied. The procedure and context
necessitated the development of adapted methods for landscape analysis, which
needed to be fast and detailed at a local scale.
2.7 The Humanistic Approach and the Revival of Landscape Ecology 21

2.7 The Humanistic Approach and the Revival

of Landscape Ecology

The reactions on the decline and efficiency of landscape research were multiple. In
1982, the Dutch ‘Working Community Landscape Ecological Research’ (WLO)
organised an international ‘brainstorming’ in Veldhoven (The Netherlands). The
participants were geographers, ecologists and some landscape architects and plan-
ners. Here, landscape ecology, as conceived by Troll, was revitalised and landscape
research got a new input by the ‘Geo€ okologie’ and ‘Landschaftslehre’ coming from
the East-European countries (Tsjallingii and de Veer 1982). The North-American
ecologists rapidly accepted the new interdisciplinary approach (Forman 1990;
Barrett et al. 2005). In 1982, the International Association of Landscape Ecology
(IALE) was founded, promoting interdisciplinary landscape research, with a
renewed interest to holism and systems theory, and considering multi-scale hierar-
chical structures and dynamics of functions and change (Forman and Godron 1986;
Naveh and Liebermann 1994; Antrop et al. 2013). Besides the existing, more
national journals such as Landschap and Landscape Research, two new interna-
tional journals were introduced: Landscape and Urban Planning in 1986 and
Landscape Ecology in 1987.
Simultaneously, the humanistic and historical approach to landscape continued
to develop. Important exemplary publications from this period are The Iconography
of Landscape (Cosgrove and Daniels 1988), Recovering the Substantive Nature of
Landscape (Olwig 1996), Understanding Ordinary Landscapes (Groth and Bressi
1997), and The History of the Countryside (Rackham 1986).
Meanwhile, landscape architects and garden designers in Europe, inspired by the
IFLA, founded in 1989 the European Federation of Landscape Architects (EFLA) ,
later becoming a regional chapter as IFLA Europe in 2007. In an attempt to improve
their education and training and making their profession more scientifically based,
different schools in landscape architecture created in 1991 ECLAS, the European
Conference of Landscape Architecture Schools, meeting annually, followed by the
creation of the thematic web-based network LE:NOTRE in 1996.
Summarizing, at the end of the twentieth century different approaches in land-
scape research could be recognised. Landscape ecologists focused on the relations
between spatial patterns of land use and ecological processes, which they tried to
model quantitatively using landscape metrics. Historical geographers and archae-
ologists focused on the time dimension and the genesis of the landscape and its
meaning as heritage. Humanistic and cultural geographers focused on the landscape
as a mental and social construct with important symbolic meanings and as a carrier
of narratives and symbolic values (Cosgrove 2002; 2003). Separately, landscape
architects and design practitioners focused on the scenery. Each of these approaches
used their proper definitions, concepts and methods, but a full interdisciplinary
integration was still lacking.
22 2 The History of Landscape Research

2.8 The ‘Landscape Crisis’ and the Shift Towards Applied

and Transdisciplinary Landscape Studies

Toward the end of the twentieth century, the pace and magnitude of societal and
environmental changes increased hugely, causing rapid and dramatic transforma-
tions of the landscape. The philosopher Ton Lemaire (2002) used the term ‘land-
scape crisis’ to denote the feeling of uncertainty and discomfort many people have
because they cannot cope with the increasing faster changes they experience in the
landscape. Although the value of the natural and traditional rural landscapes as
heritage was already recognised in law and a specific policy in many western
countries since the early twentieth century was in place, the on-going devastating
changes still caused severe threats on this natural and cultural capital. Interest in the
landscape grew again and several international congresses and conventions fostered
new actions and stimulated new research.
The first landmark was the recognition of cultural landscapes as a new category
in the World Heritage Convention of UNESCO in 1992. An important momentum
in Europe to put the landscape on the political agenda was the First Assessment of
Europe’s Environment (the ‘Dobřı́š Assessment’) in 1995 (European Environmen-
tal Agency 1995). Chap. 8 of the report is devoted to landscapes and defines
concisely the problem:
The richness and diversity of rural landscapes in Europe is a distinctive feature of the
continent. There is probably nowhere else where the signs of human interaction with nature
in landscape are so varied, contrasting and localised.
Despite the immense scale of socio-economic changes that have accompanied this
century’s wave of industrialisation and urbanisation in many parts of Europe, much of
this diversity remains, giving distinctive character to countries, regions and local areas.
(European Environmental Agency 1995).

The report manifestly links the diversity of the landscapes in Europe to the
characterisation of the European culture and identity, making it a political issue.
No explicit definition of landscape is given, but the report directly inspired the
Council of Europe to elaborate the European Landscape Convention, as shown in
the preamble (Council of Europe 2000). Undoubtedly, the European Landscape
Convention (ELC) stimulated new landscape research, inspired interdisciplinary
and transdisciplinary projects and fostered inter-regional and international
co-operation, even beyond Europe.
Other conferences and conventions stimulated indirectly more efforts in land-
scape policy. The Aarhus Convention in 2001 stimulated the participation of the
population in decision-making; The Faro Convention of the Council of Europe in
2005 emphasised the value of cultural heritage for society.
Campaigns by the Council of Europe, such as ‘Europe: a common heritage’ and
programs such as ‘European Pathways to the Cultural Landscape’ (EPCL
2000–2003) in which archaeologists took a particular interest (Clark et al. 2003),
resulted in new approaches and concepts as ‘landscape archaeology’ and ‘geo-
2.8 The ‘Landscape Crisis’ and the Shift Towards Applied and. . . 23

The Lisbon Treaty of the European Union of 2007 states: “The Union shall
respect its rich cultural and linguistic diversity, and [...] ensure that Europe’s cultural
heritage is safeguarded and enhanced” (Art.2.3) (EUR-Lex 2007). Gradually, also
the European Commission took initiatives for funding programs for preserving and
enhancing cultural heritage, which included also cultural landscapes. Conservation
was increasingly aiming to preserve and to enhance a whole cultural landscape
rather than an isolated site and also becoming more people-centred. The values of
cultural landscape are now seen also in the context of sustainability, economic
benefit and welfare. The concept of landscape services – as a particular form of
ecosystem services – gained importance (de Groot et al. 2010).
Although the main driving forces of the landscape transformation were identified
as urbanisation and industrialisation, increased mobility and mechanisation,
up-scaling of agriculture and polarisation of activities all in a global context, still
little was known how these forces really transform the landscape at the local scale
(Swaffield and Primdahl 2006; Pinto-Correia et al. 2006; Antrop and Van Eetvelde
2008). It became gradually clear that a sole academic interdisciplinary approach
was insufficient to cope with all the issues related to landscape in society. The
landscape is too complex and too many stakeholders are involved in making the real
changes in the land. Insiders and laypeople needed to be included in participatory
processes for managing and planning landscapes (Opdam et al. 2001; Selman
2006). The need for a transdisciplinary approach grew, meaning the integration
scientific knowledge and applications significant for society and involving public
participation (Naveh 1991; Tress et al. 2003, 2006). Also, the need for faster
exchange of research findings grew. Many Internet sites emerged, as well as online
open access journals such as Living Reviews in Landscape Reviews. New concepts
and methods were introduced to denote the new challenges: sustainable landscapes
(Haines-Young 2000; Wu 2013), multifunctional landscapes (Brandt and Vejre
2004a, b), landscape character assessment (Swanwick 2004), historic landscape
characterisation (promoted by English Heritage), landscape paths and trajectories
(Käyhk€ o and Skånes 2006), landscape economics (Price 2013; van der Heide and
Heijman 2013).
A divergence in landscape research can be recognised in this period. It was mainly
caused by a different emphasis on pure academic and policy-oriented and applied
research. This became in particular clear between Europe and the USA and is well
illustrated by the rise and fall of the landscape metrics. The quantitative description
of landscape patterns using spatial analysis and modelling developed during the
1980s and was stimulated by the development of GIS and the growing availability
of digital datasets. The fast development of landscape metrics and indicators was
promoted by specialised software, such as Fragstats (McGarigal and Marks 1995).
The central paradigm of landscape ecology, i.e. that spatial patterns and ecological
processes in the landscape are related, enhanced the idea that these processes could
be controlled and steered by adapting the landscape patterns through land use/land
cover planning and that these changes could be measured by indicators. Exemplary
of this thinking was the creation of ecological networks that could help to preserve
biodiversity. After a spectacular growth of the use of landscape metrics and
24 2 The History of Landscape Research

indicators, since 2004 a decline could be noticed (Uuemaa et al. 2009). However,
the quantitative methods involved were very sophisticated, the results were too
abstract, not transparent and above all uncertain to be useful for policymakers in
participatory planning. Attempts to reformulate landscape metrics as policy-
significant indicators were not successful for similar reasons. Important was also
the lack of critical thresholds and absolute values, which made these indicators
rather useless to evaluate effects of policies and impacts of decision-making. The
use of landscape metrics and modelling remained only interesting in pure academic
and theoretical research. Consequently, in Europe, in contrast to the USA, the use of
landscape metrics in applied landscape research dropped. The specific problems
that landscapes are facing in Europe are much more complex and a lot more
stakeholders are involved. The natural diversity of landscapes is intimately linked
to the cultural diversity and a long history. Thus, ecological issues, historical and
cultural ones are equally important. In Europe, landscapes are not considered as
mere configurations of spatial patterns and ecological processes, but form also a
homeland, are common heritage and contain narratives and symbolic values as
well. All these are difficult to be modelled and measured by quantitative indicators
alone and a more holistic and interdisciplinary approach is obvious. So, landscape
characterization developed, resulting in landscape descriptions and classifications
in various forms and supported by all kind of landscape representations, such as
(GIS)-maps, iconography, photography and 3D visualisations, but also narratives.

2.9 Landscape in n-Grams

The research field of ‘culturomics’ makes a quantitative analysis of culture, in

particular, based upon expression in language. This was performed on landscape
related concepts as occurring in millions of digitised books that are available today.
Google developed a web-based application, the Google Books Ngram Viewer,
which allows analysing the base corpora of approximately five million books
created from the more than fifteen million books digitised by Google, which is
only 4% of all the books ever published since the 1500s. Only books are considered
and no periodicals. Michel et al. (2011) describe the method of the creation of the
base corpora and how this database can be used. For most of the following analysis,
the English corpus was used based on the 2012 OCR and metadata. English is the
most extensive corpus, but in some cases also other languages were used for
Several interesting analyses could be performed to reveal the popularity of
landscape related n-grams over time and their relative behaviour. Although the
results of the Google Books Ngram Viewer are restricted to a selected corpus and
need to be interpreted with prudence, some interesting patterns of ‘being busy with
landscape’ emerge.
An n-gram is a contiguous sequence of n items from a given sequence of text or
speech. The items can be phonemes, syllables, letters, words or base pairs. “land-
scape” is, for example, a 1-gram (or unigram), while “landscape ecology” is a 2-gram
2.9 Landscape in n-Grams 25

Fig. 2.2 N-grams of “landscape” and “landscape ecology”

or (bigram). Typically, n-grams are collected from a text or speech corpus, such as a
collection of books published over a given time (Fig. 2.2). The Google Books Ngram
Viewer produces a graph showing the percentage a given n-gram over the selected
years in the corpus of a selected language.
The graph shows that “landscape” occurred occasionally in books since the
seventeenth century and its proportion in the corpus of all English books increased
strongly since 1750, with a dip in the first half of the twentieth century (note the dips
caused by the two World Wars) and a steep rise and the end of the century to reach
almost 0.003% of occurrence in the corpus in 2008, which is the last year of
updating the corpus. The bigram “landscape ecology” emerges rather recently
with a very low frequency (exactly 0.0000125% in 2002) compared to “landscape”
(0.0028856%). To express its variation more clearly in the graph its frequency is
scaled by a factor of 100. Then, it becomes clear that the relative frequency of both
n-grams in the corpus of the books analysed is decreasing since 2000. Similar
looking timelines do not indicate a necessarily correlation between the n-grams,
rather express the moment a n-gram became more or less popular.
The landscape is a concept with very different meanings according to the
language and the interest for the landscape did not start everywhere at the same
time and depended on other technological end societal developments. With the
Ngram Viewer, it is possible to compare the occurrence of n-grams in the corpora of
different languages. Following example compares “landscape” in the corpora of
2012 for English, French, German, Italian, Spanish, Russian and simple Chinese
over the period 1600–2008 (Fig. 2.3).
Before 1700, exceptional peaks occur mainly linked to books related to garden-
ing and painting being short living ‘bestsellers’. With the emerging sciences and
landscaping practices in the middle of the eighteenth century, a continuous interest
in the landscape becomes noticeable, starting in Germany and followed by other
Western countries. The German “Landschaft” reaches a peak with the Second
World War and then drops significantly. In other European countries, the proportion
26 2 The History of Landscape Research

Fig. 2.3 N-grams of the word “landscape” in different languages

Fig. 2.4 N-grams of landscape related disciplines

of landscape related books increased more rapidly from the second half of the
twentieth century and seems to drop since 2005. Similarly, the Chinese contribution
rises exponentially. Note also the difference between the timelines in American
(eng_us) and British English (eng_gb).
A comparison of different landscape related disciplines needs some scaling of
low frequencies in order to give a meaningful graph (Fig. 2.4). Landscape related
disciplines are scaled by a factor 100, techniques by different factors (aerial
photography enlarged 20 times, remote sensing five times and GIS two times) so
they can be more easily compared in the graph. Aerial photography starts with the
First World War, remote sensing comes in the second half of the 1960s and GIS
takes off in the 1980s. Landscape ecology emerges in the 1960s and its popularity
increased steeply in the mid-1980s. Most other landscape related disciplines come
later, except landscape architecture. Landscape architecture goes way back before
the year 1900, as geography, and its frequency shows most variation over time.
2.9 Landscape in n-Grams 27

Fig. 2.5 N-grams of “landscape architecture”, “gardening” and “landscaping”

Fig. 2.6 N-grams of applied disciplines related to landscape

The domination of “landscape architecture” deserves a more detailed analysis

(Fig. 2.5). Landscape architecture goes back to the eighteenth century but occurs
ten times less than “gardening”, which is dominant in the eighteenth and nineteenth
century. The concept “landscaping” emerges in the Interbellum of the twentieth
Concerning disciplines related to applied landscape research (Fig. 2.6), “land-
scape design” is the oldest mentioned with the highest frequency (although ten
times less frequent than “landscape architecture”). “Landscape planning” starts in
the early twentieth century. Most other disciplines became relatively more impor-
tant after 1970 and often show ups and downs through time. The concept “land-
scape policy” gets some minor significance at the end of the twentieth century.
It is interesting to see how (and where) the debate about “natural landscape” and
cultural landscape” evolved (Fig. 2.7). The English “cultural landscape” and “nat-
ural landscape” is five times less frequent than the German counterparts. The
28 2 The History of Landscape Research

Fig. 2.7 N-grams of cultural and natural landscape terms

Fig. 2.8 N-grams of related basic landscape concepts

importance of the “Kulturlandschaft” and “Naturlandschaft” rises with the debate

about the landscape as the main subject in German geography during the first half of
the twentieth century. The concept “Kulturlandschaft” is far more popular than
“Naturlandschaft”, in particular during the rise of the Nazi ideology. In English, the
popularity of cultural and natural landscape increases after the Second World War,
and the different evolution in British and American English for “cultural landscape”
is noticeable.
Similarly, the timelines of common landscape related concepts, such as “land-
scape”, “nature”, “heritage”, “landscape scenery”, “intrinsic value”, “holism”,
“Gestalt” and “systems theory” can be compared (Fig. 2.8). “Nature” always had
ten times more books published than “landscape” and was most popular during the
end of the eighteenth and during the nineteenth century. The philosophical debate
of the “intrinsic value” gained interest in the same period. “landscape scenery” was
most popular in the nineteenth century and dropped in the twentieth century.
2.9 Landscape in n-Grams 29

Fig. 2.9 N-grams of techniques and methods related to landscape analysis

Fig. 2.10 N-grams of concepts related to landscape evaluation and visualisation/visualization

“Gestalt” became rapidly popular in the beginnings of the twentieth century, while
“holism” appears in the second half of the century together with “ecology” and
“systems theory”. The interest for “heritage” follows the popularity of “landscape”.
After the Second World War different methods can be compared (Fig. 2.9). Most
popular was “land evaluation”, which was using aerial photography to make
thematic maps, in particular in developing countries, for applications in soil sci-
ence, agronomy and rangeland management. “landscape evaluation”, and later
“landscape assessment” used the same methodology, but at a more restricted
scale. “landscape mapping” had a more stable trend and “landscape typology”
developed in particular with the historical and regional description in the 1960s.
Land evaluation was taken over with the exponential development of GIS and the
availability of spatial data. GIS stimulated also the use of “landscape metrics” and
the development of “landscape visualization” (while “visualisation” is clearly less
represented) (Fig. 2.10).
30 2 The History of Landscape Research


Antrop, M., & Van Eetvelde, V. (2008). Mechanisms of recent landscape transformation. In Ü.
Mander, C. Brebbia, & J. F. Martin-Duque (Eds.), Geo-environment and landscape evolution
III (pp. 183–192). Southampton: WIT Press.
Antrop, M. (2013). A brief history of landscape research. In P. Howard, I. Thompson, &
E. Waterton (Eds.), The Routledge compagnion to landscape studies (pp. 12–22). London:
Antrop, M., Brandt, J., Loupa Ramos, I., Pinto-Correia, T., Padoa-Schioppa, E., Porter, J., & Van
Eetvelde, V. (2013). How landscape ecology can promote the development of sustainable
landscapes in Europe – The role of the European Association for Landscape Ecology (IALE
Europe) in the 21st century. Landscape Ecology, 28, 1641–1647.
Barrett, G. W., Barrett, T. L., Wu, J. (2005). History of landscape ecology in the United States
(190 pp). New York: Springer.
Boulanger, P. (2002). La géographie militaire française: 1871–1939 (pp. 35–41). Paris:
Brandt, J., & Vejre, H. (Eds.). (2004a). Multifunctional Landscapes. Theory, Values and History.
Vol. I (p. 276). Southampton: WIT Press.
Brandt, J., & Vejre, H. (Eds.). (2004b). Multifunctional Landscapes. Monitoring, diversity and
management. Vol. II (p. 292). Southampton: WIT Press.
Bunske, E. V. (1981). Humboldt and an aesthetic tradition in geography. The Geographical
Review, 71(2), 127–146.
Clark, J., Darlington, J., & Fairclough, G. (2003). Pathways to Europe’s landscape. European
Pathways to Cultural Landscape (EPCL) (114 pp),
Claval, P. L. (2004). The languages of rural landscapes. In H. Palang, H. Sooväli, M. Antrop, &
S. Setten (Eds.), European rural landscapes: Persistence and change in a globalising envi-
ronment (pp. 11–40). Dordrecht/Boston: Kluwer Academic Publishers.
Cosgrove, D. (2002). Landscape and the European sense of sight – eyeing nature. In K. Anderson,
M. Domosh, S. Pile, & N. Thrift (Eds.), Chapter 12: Handbook of cultural geography
(pp. 249–268). London: SAGE Publications.
Cosgrove, D. (2003). Landscape: ecology and semiosis. In H. Palang & G. Fry (Eds.), Landscape
Interfaces. Cultural heritage in changing landscapes (pp. 15–20). Dordrecht: Kluwer Aca-
demic Publishers.
Cosgrove, D. (2004). Landscape and landschaft (pp. 57–71). Lecture at the “Spatial Turn in
History” Symposium German Historical Institute, GHI Bulletin, No.35.
Cosgrove, D., & Daniels, S. (1988). The Iconography of landscape. Cambridge: Cambridge
University Press.
Council of Europe. (2000). European landscape convention. Firenze, 20.10.2000 http://conven
Darby, H. C., & Campbell, E. M. J. (Eds.). (1962). The domesday geography of South-East
England. Cambridge: Cambridge University Press.
Darby, H. C., & Terrett, I. B. (Eds.). (1978). The domesday geography of Midland England (2nd
ed.). Cambridge: Cambridge University Press.
Darby, H. C., & Maxwell, I. S. (Eds.). (1978). The domesday geography of Northern England (2nd
ed.). Cambridge: Cambridge University Press.
Darby, H. C., & Welldon Finn, R. (Eds.). (1967). The domesday geography of South-West
England. Cambridge: Cambridge University Press.
Darby, H. C. (1971). The domesday geography of Eastern England (third ed.). Cambridge:
Cambridge University Press.
de Groot, R. S., Alkemade, R., Braat, L., Hein, L., & Willemen, L. (2010). Challenges in
integrating the concept of ecosystem services and values in landscape planning, management
and decision making. Ecological Complexity, 7, 260–272.
Elden, S. (2011). Reintroducing Kant’s geography. Albany: State University of New York Press.
References 31

Elliott, P., & Daniels, S. (2006). Pestalozzi, Fellenberg and British nineteenth-century geograph-
ical education. Journal of Historical Geography, 32, 752–774.
EUR-Lex. (2007). Access to European law.
European Environmental Agency. (1995). Europe’s environment: The Dobrˇíš assessment. Copen-
Forman, R. T. T. (1990). The beginnings of landscape ecology in America. In I. S. Zonneveld &
R. T. T. Forman (Eds.), Changing landscapes: An ecological perspective (pp. 35–41).
New York: Springer Verlag.
Forman, R., & Godron, M. (1986). Landscape ecology (619pp). New York: Wiley.
Gran€o, O., Paasi, A., & Gran€ o J. G. (Eds.). (1997). Pure geography (M. Hicks Trans., 191 pp) .
Baltimore/London: The John Hopkins University Press.
Gran€o, J. G. (1929). Reine Geographie. Eine methodologische Studie beleuchtet mit Beispielen
aus Finnland und Estland. Helsinki, Helsingfors, Acta Geographica, 2(2), 202.
Groth, P., & Bressi, T. W. (Eds.). (1997). Understanding ordinary landscapes (p. 272). New
haven: Yale University Press.
Haines-Young, R. H. (2000). Sustainable development and sustainable landscapes: Defining a new
paradigm for landscape ecology. Fennia, 178(1), 7–14.
Hartshorne, R. (1939). The nature of geography. A critical survey of current thought in the light of
the past. Annals of the Association of American Geographers, 29(3–4), 173–658.
Hartshorne, R. (1961). The nature of geography (504 pp). Reprinted with corrections from the
Annals of the Association of American Geographers.
Helmfrid, S. (2004). The permanent conference and the study of the rural landscape. In H. Palang,
H. Sooväli, M. Antrop, & S. Setten (Eds.), European rural landscapes: Persistence and change
in a globalising environment (pp. 467–482). Dordrecht/Boston: Kluwer Academic Publishers.
Hoskins, W. G. (1955). The making of the English landscape (third ed.). London: Hodder and
Stoughton. 1988.
Jones, M. (2003). Human geographical landscapes: J.G. Gran€ o’s approach to landscapes as
scientist and artist. In O. Gran€o (Ed.), Origin of landscape science. J.G. Gran€ o an a new
pure geography for a new state (pp. 71–98). Turku: The Turku University Foundation.
Käyhk€o, N., & Skånes, H. (2006). Change trajectories and key biotopes—Assessing landscape
dynamics and sustainability. Landscape and Urban Planning, 75(3), 300–321.
Lebeau, R. (1979). Les grands types de structures agraires dans le monde (162 pp). Paris, Masson,
troisième édition.
Lemaire, T. (2002). Met open zinnen. Natuur, landschap, aarde (358 pp). Amsterdam, Ambo.
Leopold, L., et al. (1971). A procedure for evaluating environmental impact (Vol. 645).
Washington DC: US Geological Survey Circular.
Lowenthal, D. (1975). Past time present place: Landscape and memory. The Geographical Review,
65(1), 1–36.
Maconochie, A. K. (1989). Geography: An international Gallup survey. Survey of findings. McGill
Journal of Education, Book Review, 24, 003.
McGarigal, K., & Marks, B. J. (1995). FRAGSTATS: Spatial pattern analysis program for
quantifying landscape structure (p. 351). Portland: U.S. Department of Agriculture, Forest
Service, Pacific Northwest Research Station.
Michel, J. B., Yuan Kui Shen, Y., Presser Aiden, A., Veres, A., Gray, M. K., Brockman, W., The
Google Books Team, Pickett, J. P., Hoiberg, D., Clancy, D., Norvig, P., Orwant, J., Pinker, S.,
Nowak, M. A., & Aiden, E. L. (2011). Quantitative analysis of culture using millions of
digitized books. Science, 331(6014), 176–182.
Moss, M. R. (1999). Fostering academic and institutional activities in landscape ecology. In J. A.
Wiens & M. R. Moss (Eds.), Issues in landscape ecology (pp. 138–144). Snowmass Village:
International Association for Landscape Ecology, Fifth World Congress.
Muir, R. (1999). Approaches to landscape (310 pp). London: MacMillan Press.
32 2 The History of Landscape Research

Naveh, Z., & Lieberman, A.. (1994). Landscape ecology: Theory and application (360 pp).
New York: Springer.
Naveh, Z. (1991). Some remarks on recent developments in landscape ecology as a transdisci-
plinary ecological and geographical science. Landscape Ecology, 5, 65–73.
Neef, E. (1967). Die theoretische Grundlagen der Landschaftslehre. V.E.B. Haack Gotha.
Nicholson, M. (1995). A historical introduction in: Von Humboldt, A., 1814. Personal narrative of
a journey to the equinoctial regions of the new continent (Abridged and translated, 311 pp).
London: Penguin Books.
Olwig, K. R. (1996). Recovering the substantive nature of landscape. Annals of the Association of
American Geographers, 86(4), 630–653.
Olwig, K. R. (2002). Landscape, nature and the body politic: From Britain’s Renaissance to
America’s new world (299 pp). Madison: University of Wisconsin Press.
Opdam, P., Foppen, R., & Vos, C. (2001). Bridging the gap between ecology and spatial planning
in landscape ecology. Landscape Ecology, 16(8), 767–779.
Paffen, K. (Ed.). (1973). Das Wesen der Landschaft. Wege der Forschung (Band XXXIX, 514 pp).
Darmstadt: Wissenschaftliche Buchgesellschaft.
Passarge, S. (1919–1921). Grundlage der Landschaftskunde (3 vol). Hamburg: Friedrichsen.
Passarge, S., 1921–1930. Vergleichende Landschaftskunde (5 vol). Berlin: Reimer.
Pedroli, B. (1983). Landscape concept and landscape and rangeland surveys in the Soviet Union.
ITC Journal, 4, 307–321.
Pinto-Correia, T., Gustavsson, R., & Pirnat, J. (2006). Bridging the gap between centrally defined
policies and local decisions – Towards more sensitive and creative rural landscape manage-
ment. Landscape Ecology, 21(3), 333–346.
Price, C. (2013). Reaching the economics of landscape. In P. Howard, I. Thompson, & E. Waterton
(Eds.), The Routledge compagnion to landscape studies (pp. 308–321). London: Routledge.
Rackham, O. (1986). The history of the countryside. London: J.M. Dent.
Sauer, C. O. (1925). The morphology of landscape. In J. Leighly (Ed.), Land and life (University of
California Publications in Geography, 2; Reprinted in, 1974) (pp. 315–350). Berkely: Univer-
sity of California Press.
Schama, S. (1995). Landscape and memory (p. 652). New York: A. Knopf.
Selman, P. (2006). Planning at the landscape scale (213 pp). London: Routledge.
Sochava, V. B. (1971). Geography and Ecology. Soviet Geography, XII, 277–293.
Swaffield, S., & Primdahl, J. (2006). Spatial concepts in landscape analysis and policy: some
implications of globalisation. Landscape Ecology, 21(3), 315–331.
Swanwick, C. (2004). Landscape character assessment. Guidance for England and Scotland.
Cheltenham: The Countryside Agency, Scottisch Natural Heritage.
Taylor, C. (2006). The making of the English landscape and beyond: Inspiration and dissemina-
tion. Landscapes, 2, 96–104.
Thomas, W. L. (Ed.). (1956). Man’s role in the changing face of the Earth (1193 pp). Chicago: The
University of Chicago Press.
Tress, B., Tress, G., Fry, G., & Opdam, P. (Eds.). (2006). From landscape research to landscape
planning: Aspects of integration, education and application, Wageningen UR Frontis Series, v
12. Dordrecht: Springer.
Tress, B., Tress, G., van der Valk, A., & Fry, G. (Eds.). (2003). Interdisciplinary and transdisci-
plinary landscape studies: Potential and limitations, Delta Series 2 (192 pp). Wageningen:
Delta Program.
Troll, C. (1939). Luftbildforschung und Landeskundige Forschung. Erdkundliches Wissen.
Schriftenreihe für Forschung und Praxis, Heft 12. Wiesbaden: F.Steiner Verlag.
Troll, C. (1950). Die geographische Landschaft und ihre Erforschung (pp. 163–181). Studium
Generale, 3 Jg., H.4/5.
Tsjallingii, S. P., & de Veer, A. A. (1982). Perspectives in landscape ecology. Wageningen: Centre
for Agricultural Publishing and Documentation.
References 33

Tuan, Y.-F. (1974). Topophilia: A study of environmental perception, attitudes and values.
Englewood Cliffs: Prentice Hall.
Uhlig, H., & Lienau, C. (1972). Internationale Arbeitsgruppe für die geographische Terminologie
in der Agrarlandschaft.
Uuemaa, E., Antrop, M., Roosaare, J., Riho, M., & Mander, Ü. (2009). Landscape metrics and
indices: An overview of their use in landscape research. Living Reviews in Landscape Research
3,1. URL (cited on <date>):
van der Heide, M., & Heijman, W. J. M. (Eds.). (2013). The economic value of landscapes
(329 pp). Routledge Studies in Ecological Economics.
Vidal de la Blache. (1908). Translated from the electronic version on
Vinogradov, B. V. (1967). The landscape concept and its use in the study of grassland territories
(N S 16, 27 pp). Delft: Publication of the ITC-Unesco Centre for Integrated Studies.
von Humboldt, A. (1845). Kosmos (Vol. 1, pp. 64–65). Stuttgart & Tübingen.
Wardenga U. (2006). German geographical thought and the development of L€ anderkunde
(pp. 127–147). Inforgeo, 18/19, Lisboa, Edições Colibri.
Wiens, J. H., Moss, M. R., Turner, M. G., & Mladenoff, D. J. (Eds.). (2007). Foundation papers in
landscape ecology. New York: Columbia University Press.
Wu, J. (2013). Landscape sustainability science: Ecosystem services and human well-being in
changing landscapes. Landscape Ecology, 28, 999–1023.
Zonneveld, I. S.. (1995). Land ecology (199 pp). Amsterdam: SPB Academic Publishing bv.
Zonneveld, I. S. (2000). Count your blessings? Twenty-five years of landscape ecology. In J. Klijn
& W. Vos (Eds.), From landscape ecology to landscape science (pp. 30–42). Wageningen:
Kluwer Academic Publ.
Zonneveld, J. I. S. (1980). Het gat in de geografie. K.N.A.G Tijdschrift XIV, 2, 93–98.
Chapter 3
The Multiple Meanings of Landscape

Abstract The word landscape has multiple meanings in common language. It

refers to a tract of land as well to its visual appearance. It used to denote a territory
as well as historical and geographical regions and refers to an area shaped in a
characteristic way by the people living in it. Landscape as scenery had an important
influence on imagination and artistic presentation. Subtle differences exist between
languages, which sometimes make translations confusing. Adjectives are used to
describe more precisely the meaning in a given context. The distinction between
natural and cultural landscapes has been important in the development of different
approaches. Many other specifications, such as rural and urban landscape, country-
side and picturesque landscape, add to the complexity. Formal definitions were
made by conventions, such as the cultural landscapes in the UNESCO World
Heritage Convention and the definition given by the European Landscape Conven-
tion. In particular, the latter initiated rethinking landscape policy and stimulated
transdisciplinary approaches to landscape in an international perspective.

Keywords Etymology • Definition • Multilingual • European landscape

convention • UNESCO cultural landscapes

3.1 Introduction

This chapter deals with the etymology of the word ‘landscape’ and the complex
linguistic meanings that emerged from it. It originated in the Dutch-Germanic-
Scandinavian languages and subtle shifts in the meaning occurred when it became
introduced in other languages. It has multiple meanings and the meaning that is
used depends on the context the word is used in. To clarify this, adjectives are used.
In the international context of research and policy, the need for a common and clear
definition became necessary. Two important of these formal definitions will be
discussed in detail: the cultural landscapes in the framework of the UNESCO World
Heritage Convention, and the European Landscapes Convention by the Council
of Europe.

© Springer Science+Business Media B.V. 2017 35

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_3
36 3 The Multiple Meanings of Landscape

3.2 Origins and Etymology of the Word Landscape

The origin of the word landscape comes from Germanic languages. One of the
oldest references in the Dutch language dates from the early thirteenth century,
where ‘lantscap’ (‘lantscep’, ‘landschap’) refers to a land region or a specific
environment. It is related to the word ‘land’, meaning a particular territory, but its
suffix –scap or -scep refers to land reclamation and creation, as is also found in the
German ‘Landschaft’ – ‘schaffen’ ¼ to make. In the sixteenth century, the concept
is broadened and includes homeland and historical region or territory. The shift in
meaning from an ‘organised territory’ to ‘scenery’ is obvious and linked to a style of
Today, in the Dutch language the word bears following meanings:
• a (historical) region;
• in a geographical sense: part of the Earth’s surface that can be recognised as a
distinct entity;
• a painting representing a landscape (also: the format and shape of the painting);
• part of the countryside one sees from a viewpoint, i.e. the scenery.
Kenneth Olwig (1996, 2002, 2004) studied how the word landscape with its
multiple meanings was used during history since the Renaissance. He used the
meanings as given in the Merriam-Webster dictionary of 1995:
Landscape. Etymology: Dutch landschap, from land + -schap –ship. Date: 1598.
1 a: a picture representing a view of natural inland scenery; b: the art of depicting such
2 a: the landforms of a region in the aggregate; b: a portion of territory that can be
viewed at one time from one place; c: a particular area of activity: scene (political
3: obsolete: vista, prospect.

Olwig (1996) argued that landscape “need not be understood as being either
territory or scenery; it can also be conceived as a nexus of community, justice,
nature, and environmental equity, a contested territory that is as pertinent today as it
was when the term entered the modern English language at the end of the sixteenth
Essentially, landscape indicates a particular part of the land and its representa-
tion. Interesting is definition 2c, which also sets landscape as the scene of action.
Thus, landscape became also an expression of human ideas, thoughts, beliefs and
feelings. Luginbühl (2012) discusses the intimate relationship between society and
landscape, and the landscape as tangible reality and its representation. He follows
the theory of Augustin Berque (1982) that two types of societies exist: one without
the notion of landscape (sociétés sans paysage) and one with the notion of land-
scape (sociétés a paysage). ‘Societies without landscape’ have only a pure utilitar-
ian or symbolic relation with the land and are incapable of contemplating the
landscape. ‘Societies with landscapes’ contemplate the land resulting in landscapes
with proper names and in various artistic expressions and representations.
3.2 Origins and Etymology of the Word Landscape 37

Clearly, ‘land’ and ‘landscape’ are two different concepts. When ‘land’ refers
more to soil, terrain and territory, ‘landscape’ includes also the meaning of
‘organised land’, which is characteristic for the people who made it and live
there, a representation of a scene of action and activity. Landscape expresses also
the (visual) manifestation of the territorial identity.
The earliest realistic representations of landscape date from the fifteenth century,
in particular in Renaissance painting (Vos 2000) and emphasise the visual character
and scenery. Howard (1991) demonstrated how important a scientific study of
landscape paintings could be. Han L€orzing (2001) cites a comment Michelangelo
had given on the new style of painting introduced by Flemish painters coming to
Italy looking for inspiration and work and whose art he considered an inferior form
of painting (see also Chap. 11). Characteristic for these Flemish painters is the
astonishing detailed, almost photographic representation of landscape and nature.
On probably one of the first oil paintings, the Ghent Altarpiece or Adoration of the
Mystic Lamb of the brothers Van Eyck (completed around 1432), seventy-five
species of plants could be identified. This way of documentary realistic painting
became an important tool for documenting the new landscapes, and new flora and
fauna and people explored the Age of Discovery. However, although the elements
are realistic, the landscape represented is fiction and a composite of realistic looking
fragments of landscapes. Yves Luginbühl (2012) draws the attention to the large
number of common people represented, labouring in a rural landscape. He con-
siders these paintings also as the representation of a social utopia. These paintings
show often the land or a composite of the territories belonging to the commissioner.
In this sense, these landscape paintings are very similar to the estate maps. This is
well illustrated by The Hay Harvest of Pieter Breugel (Fig. 3.1).
The understanding of the rules of perspective and the development of cartogra-
phy also allowed a reverse way of representation: projecting paintings on the terrain
to create new artistic landscapes, which made some say regarding English garden
architecture that it was just ‘planting paintings’ (see also Chap. 11). The success of
the new trend of painting landscapes promoted the meaning of landscape as scenery
internationally. During the seventeenth century when Holland became a prosper-
ous, economic world power, many Dutch words were introduced in other lan-
guages, ‘landschap’ being one of these. The Dutch paintings representing
landscapes became a trendy fashion. When William III of Orange-Nassau became
king of England and during the Glorious Revolution of 1688, the word ‘landscape’
was introduced in English but mainly in the sense of a scenic representation and less
in the meaning of territory.
The Dutch suffix –schap is spelt differently in other languages and minor
differences in meaning can be noted. In Dutch, –schap derives etymologically
from schep or schop, meaning ‘spade’ or ‘shovel’. It refers to the clearing of the
land to make arable land. A similar meaning is found in the German Landschaft, as
schaffen means to make, to create. In Dutch and Flemish several geographical
regions have proper names ‘Land of. . ..’ Such as Land van Nete en Aa, Zeeland,
38 3 The Multiple Meanings of Landscape

Fig. 3.1 The Hay Harvest of Pieter Breugel (1565). The fictive (‘European’) landscape
represented here is a composite of detailed elements, realistically painted. We recognise isolated
farmsteads (1), a green village (2), the village windmill on a ridge, marking the border of the
territory (3), a walled town (4), a fortress or cloister on a rock (5). The landscape contains many
woodlots and hedgerows and the way is shown how trees are pollarded (6). The land use is mainly
hay land, which is questionable as hay lands mainly occurred on wet alluvial soils (as in 8) and
were not common on higher terrain and sloping ground as depicted in the front. At the foot of the
rock systematic land reclamation with enclosed fields can be seen (9). The local flora is shown in
front. (Oil painting on wood, 114  158 cm, Národnı́ Galery, Prague)

The English suffix –scape comes from the Old English –scipe, having the same
meaning as –ship, pointing at a partnership or relationship as in citizenship:
belonging to the citizens. So landscape refers to the characteristic image of the
land belonging to a given community. English is the only language where landscape
is also a verb: ‘to landscape’ and ‘landscaping’ points clearly to the making the
landscape and scenery in particular or improving nature. In British English, the
meaning of landscape as scenery has become prominent and the meaning as
(historical) territory or land vanished. Instead, the names of the counties are used
and to distinguish between the rural and the urban landscape, countryside and
townscape are used.
In Scandinavian languages, the term is based on the Old-Norse landskap. The
suffix –skap, skab refers also to territory and landskab was introduced to replace
land to denote territories with a political autonomy since the fifteenth century in
Denmark and the eighteenth century in Sweden (Olwig 1996). Landscape as a
territorial unit is still used in several regions such Friesland and Schlewig-Holstein
3.3 Subtleties of Language – Landscape Versus Land 39

(Cosgrove 2004) and recently, landskab has been re-introduced in Sweden as a

territorial planning unit referring to historical territories.
Jones (2005) and Olwig (2005a, b, 2013) discuss the shifting relation between
the landscape concept and (customary) law. The (oldest) territorial meaning of
landscape refers to a rather fuzzy-bordered space where a local community lives
and where customary rights organise the relations between people and the assem-
bled ‘things’ that form the landscape (Olwig 2013). The creation of centralised
states and privatising the (common) land transformed ancient territories into
administered regions with sharp formal borders and customs became written
laws. The complexity of the landscape also made compartmentalised different
landscape issues in different bureaucratic policy sectors (Olwig 2013).
In Roman languages (French, Italian, Spanish, etc.) the emphasis of the meaning
is completely different. The French word paysage contains clearly the concept of
land, pays, derived from the Latin pagus. The French geographer Paul Claval
(2004) sees the landscape as a social phenomenon representing the political or
social identity of a community. Nowadays many (historical) regions and landscapes
in France are still referred to as ‘Pays de. . .’ (‘land of. . .’). Nevertheless, pays
differs from the concept of land that in French is called terre, more referring to
terrain and soil. Related concepts terroir and territoire denote a tract of land with a
specific management. The term terroir refers to the agricultural land of a village
community. Also, the word campagne (champagne) is used to denote the rural
countryside. Territoire is similar with a region as a spatial, administrative or
political unit as in one of the initial meanings of ‘landschap’. Paysage also refers
to the scenery, the appearance of the land (compare with the French word for face:
visage) and also in French it means a genre in painting.

3.3 Subtleties of Language – Landscape Versus Land

In common language, the word landscape has multiple meanings, which also vary
among languages (Fig. 3.2, Table 3.1). Most people experience their physical
environment as a material and a perceivable reality full of meaning and so land-
scape is often linked to their personal experience and values. Thus, the landscape is
not only a space filled with all kinds of things, but has as a whole of existential
meanings, deeply rooted in the culture and history and reflecting the identity of the
land. In some languages, landscape may also refer to a social territory, which is
sometimes reflected in ancient administrative divisions.
A particular problem is given by the subtle differences in meaning of landscape-
related concepts in the many languages used in Europe, impeding understanding in
international co-operation. For example, the German ‘Landschaft’ has a focus on
the territorial meaning, while the English ‘landscape’ mainly refers to the scenery.
In old Nordic tradition ‘landscape laws’ were landscape specific frames for regional
land use regulation. An interesting subtle difference is found between American
and British English. The American concept of ‘landscape’ was introduced from the
40 3 The Multiple Meanings of Landscape

Fig. 3.2 Key meanings of the word ‘landscape’ in selected European languages and their

Table 3.1 The word landscape and related terms in selected European languages
Concept Dutch/Flemish German English French Spanish
Scenery Landschap Landschaft Landscape Paysage Paisaje
Land surface, Land, terrein Land Land, Terrain Terreno
terrain terrain
Earth Aarde Erde Earth Terre, sol Tierra
Soil Grond, bodem Grund, Boden Soil Sol Suelo
Spatial unit, area Streek, gebied (Erd)Gegend Region Région, Region,
(Greek chore) territoire territorio
Content of an area Landschap Landschaft Land, Territoire, Territorio,
(Greek topio) landscape pays pais
Rural, agrarian land Platteland, Flur Countryside Campagne, Campo
or region landelijk gebied terroir
Bordered, Territorium Raum, Gegend Territory Territoire, Territorio
delimited area terroir
Spatial unit (Old: landschap) (Old: Landscape Pays, Paisaje
streek, regio Landschaft), territoire
Administrative or Land, Land(schaft) Land Pays Pais
political unit (landschap)
3.3 Subtleties of Language – Landscape Versus Land 41

German Landschaft in the sense of territory homeland of a community and not from
the English landscape as scenery (Cosgrove 2004).
The French ‘paysage’ clearly refers to the characteristic appearance of a ‘pays’,
a characteristic region of land with deep historical roots. A similar meaning is found
in all Roman languages: Portuguese paisagem, Spanish paisaje, Catalan paisatge,
Italian paesaggio and Romanian peisaj. Also similar is the Russian and Bulgarian
пейзаж, pronounced as pyeı˘zazh. However, the term used in the Russian version of
the European Landscape Convention is Ландшафт, phonetically the same as the
German Landschaft.
Subtle differences exist also in Slavic languages where the concept landscape
roots in ‘krajina’, which is closely linked to ‘land’: in Slovak and Czech krajobraz,
in Polish kraštovaizdis (Lithuanian). The Slovenian pokrajina combines meanings
of soil and land (cf. Slovak for soil is pôda).
Also, the Estonian maastik refers to land (maa), as the Finnish maisema. The
Greek word for landscape (‘topío’ τoπιo) referring to scenery and derives from
‘topos’ meaning ‘place’, while ‘chora’ (χω  ρα) is used for region, land, realm and
There is no word for landscape in Arabic (Makhzoumi 2002) and it is also absent
in Turkish. The Western way of conceiving landscape as an aesthetical view of the
countryside and its meaning as a territorial unit does not exist in cultures of the
Middle East. Makhzoumi (2002) explains this by the extreme contrast between a
hostile desert environment and the human made agrarian and urban landscapes
where people live. Therefore, the focus of aesthetic appreciation is on enclosed,
ordered, constructed cultural environment, with qualities as comfort, security and
Zev Naveh and Arthur Lieberman (1994) point to the affinity between the
Hebrew word for landscape, noff, and the word yafe, which means beauty, and
thus refers to the scenery.
The difference between land and landscape becomes even visible in the Chinese
kanji-scripture where the signs for land and soil are similar as are the ones for
landscape and scenery and beauty (Fig. 3.3). See also Chap. 11.
Summarizing, in common language, the word landscape has multiple meanings
and subtle differences exist between ‘landscape’ and related terms in different
languages (Fig. 3.2, Tables 3.1, 3.2). According to the focus one makes, different
perspectives of research and actions are possible. Searching the ‘exact’ meaning of
the word and its ‘scientific’ definition dominated the early start of landscape
research (Naveh and Liebermann 1994; Zonneveld 1995; Olwig 1996; Muir
1999; Claval 2004; Antrop 2005). Landscape does not only refer to a complex
phenomenon that can be described and analysed using objective scientific methods,
it also refers to a subjective observation and experience and thus has a perceptive,
aesthetical, artistic and existential meaning (Lowenthal 1975; Cosgrove and Dan-
iels 1988). It is not surprising that the approaches to landscape are very broad and
not always clearly defined. Most interest groups dealing with the same territory of
land see different landscapes. The meaning of word landscape shifts by the context
it is used in and by the background of the users.
42 3 The Multiple Meanings of Landscape

Fig. 3.3 The Chinese

kanji-signs express ideas
and concepts (The signs for
land and soil are similar, but
different from landscape
that is more similar to the
signs for beautiful and

Table 3.2 The word landscape in the different linguistic versions of the European Landscape
Albanian peizazh French paysage Portuguese paisagem
Catalan paisatge Galician paisaxe Romanian peisajul
Croatian krajobraz German Landschaft Russian Ландшафт
Czech krajina Greek Τoπίo Serbian predeo
Danish landskap Hungarian t
aj Slovak krajina
Dutch landschap Italian paesaggio Slovenian krajini
English landscape Latvian ainava Spanish paisaje
Estonian maastik Norwegian landskap Swedish landskap
Finnish maiseman Polish krajobraz Turkish peyzaj

3.4 Landscape with Adjectives

The multiple connotations of the word landscape are also reflected in the use of
many adjectives to clarify the type of landscape one is speaking about. Landscape
can be experienced in very different ways and it is wise to include the perspective of
the observer in its definition: ‘landscape seen as. . .’ (Howard 2011). Luginbühl
(2012) considers the landscape a tangible reality perceivable by individuals and
characterised by two dimensions: a material one and an immaterial dimension. The
material dimension is inert, biological and social. The immaterial dimension
includes affection and aesthetics, sensory and symbolic. Denis Cosgrove (2003)
discussed two basic and diverging approaches to landscape, which he called the
ecological and the semiotic discourse. The ecological discourse focuses on the
complex interactions of natural processes that shape and characterize the land. The
semiotic discourse emphasis more on the way cultural processes shape the world.
In European-Christian tradition, a sharp distinction has been made between
‘natural’ and ‘cultural’ landscapes, as well as between ‘rural’ and ‘urban’ land-
scapes. Many other specifications were created as well, also metaphorically, such as
3.4 Landscape with Adjectives 43

in the expression ‘political landscape’. A variety of concepts associated with

landscape make the picture even more complex, such as ‘countryside‘, ‘campagne’,
‘region’ and ‘terroir’.

3.4.1 Natural and Cultural Landscape

The German geographer Friedrich Ratzel has been credited introducing the term
Kulturlandschaft about 1890, but earlier Joseph Wimmer has used it extensively in
his Historische Landschaftkunde and even earlier Carl Ritter used the term
Culturlandschaft in Die Erdkunde (Mathewson 2011, p.140). Ratzel saw the
Kulturlandschaft as the part of the initial Naturlandschaft that was transformed
by human activity. Both terms were mainly popular amongst German geographers
and are illustrative for the deterministic approach in geography.
Carl Sauer and other geographers criticized environmental determinism, which
lead to the development of environmental possibilism to explain cultural develop-
ment. Carl Sauer (1925) also refers to this German distinction between natural and
cultural landscape, but interprets it as two facets necessary to understand the whole
. . . geography is based on the reality of the union of physical background and facts of
human culture. The content of landscape is therefore found in the physical qualities of area
that are significant to man and in the forms of his use of the area, in facts of physical
background and facts of human culture. (Sauer 1925, p. 325)

In his “morphologic method of synthesis” to study landscape, both facets are

studied in a generic way considering specific factors and forms, which Sauer
represented in following diagram (Sauer 1925, p. 337, 343) (Fig. 3.4).

Fig. 3.4 The morphologic

method of synthesis (After
C. Sauer 1925)
44 3 The Multiple Meanings of Landscape

Clearly, a hierarchical relation between two facets is given, which Sauer sum-
marizes as “The cultural landscape is fashioned from a natural landscape by a
culture group. Culture is the agent, the natural area is the medium, the cultural
landscape is the result” (Sauer 1925, p. 343). But following precision is important:
“The cultural landscape is the geographic area in the final meaning (Chore). It
forms are all the works of man that characterize the landscape. Under this defini-
tion, we are not concerned in geography with the energy, customs, or beliefs of man
but with man’s record upon the landscape.” (Sauer 1925, p.342). It means that the
analysis is limited to the material, visible landscape.
The concepts of cultural and natural landscape raised many problems. Richard
Hartshorne criticised both terms as well as the concept of landscape itself as a key
subject of research in geography. He found the term landscape too confusing and
redundant with the concept region (Hartshorne 1939, 1961). Michael Jones (2003)
discusses many problems that emerged by using both concepts when dealing with
one spatial unit, causing a fundamental split in the holistic unity of the landscape.
Gradually, in the academic world, the formal distinction between natural and
cultural landscape vanished with the awareness that all over the world the environ-
ment is influenced by human activities and unaffected, really pristine landscapes do
not exist anymore. The variation of the world’s landscapes is a continuum on a
gradient scale of the intensity of human’s impact. Also, activities as ‘helping’
nature through conservation and making ‘new’ nature makes the use of the term
natural landscape even more confusing. Hence, new concepts as wilderness,
rewilding and green infrastructure were introduced.
When concepts as natural and cultural landscape have become obsolete in
scientific research, it is not the case in policy. For example, the first formal
definition of landscape was given in 1992 when ‘cultural landscape’ was added as
a new category on UNESCO’s World Heritage List (UNESCO 1992). Also, in
1995, the Council of Europe accepted a ‘Recommendation on the Integrated
Conservation of Cultural Landscapes Areas as Part of Landscape Policies’. The
introduction of the term ‘cultural landscape’ instead of simply ‘landscape’ reflects
also a more political strategic reasoning. Already in 1959, the International Union
for the Conservation of Nature (IUCN) was asked by UNESCO to prepare a list of
Nature Parks and reserves to be protected and again in 1972 to provide technical
evaluations and monitoring of protected areas. IUCN had also developed a man-
agement categories system, adopted worldwide, including Category V – protected
landscape/seascape. Although it is recognised that these areas have a “distinct
character in regard to their ecological, biological, cultural or scenic value” and
“allow a higher level of interaction with surrounding communities who are able to
contribute to the areas management and engage with the natural and cultural
heritage it embodies through a sustainable outlook”, the focus of the management
plan is mainly on nature conservation in the scope of preserving biodiversity. The
introduction of the category of ‘cultural landscape’ in the World Heritage Conven-
tion seems to correct this bias and to enhance both tangible and intangible heritage
values. Using the specification ‘cultural’ avoids interference in the different policy
3.4 Landscape with Adjectives 45

3.4.2 Rural and Urban Landscape

In old texts landscape is also used to refer to the rural land as opposed to the towns
and cities. The initial Latin meaning also associates ‘urban’ with privileged and
civilised (civilian rights). Consequently, everything that is rural is seen as retarded
and subjugated. This distinction also differentiates between two lifestyles and
visions upon the environment. The rural areas sometimes refer to the agricultural
land, omitting, however, forests and wasteland. Nevertheless, they contain villages
and towns, so ‘the rural’ also refers to the community. In the English tradition, this
is referred to as the countryside, the French equivalent is la campagne.
In policy terms, rural areas are often used as synonym of countryside, combining
an economic and social component. For example, the Rural Development Programs
2007–2013 of the EU defined three objectives as (1) improving the competitiveness
of agriculture and forestry, (2) improving the environment and the countryside, and
(3) improving the quality of life in rural areas and encouraging diversification of
economic activity (European Commission 2008). The EU considers the rural areas
a “vital part of its physical make-up and its identity” and defines more than 91% of
its territory as “rural”, where more than 56% of the EU’s population is living. Also,
these rural areas contain “the EU’s fantastic range of striking and beautiful land-
scapes are among the things that give it its character”. Nevertheless, rural areas are
considered problematic, with agriculture and forestry lacking competitiveness on a
global scale, with lower income for the population and less developed services. On
the other hand, the positive qualities are lyrically described as the countryside [. . .]
gives us essential raw materials. Its value as a place of beauty, rest and recreation –
when we look after it – is self-evident. It acts as our lungs, and is therefore a
battleground for the fight against climate change. And many people are attracted by
the idea of living and/or working there, provided that they have access to adequate
services and infrastructure.” And the task of the rural development policy is “all
about meeting the challenges faced by our rural areas, and unlocking their poten-
tial” (European Commission 2008). In many countries, rural areas are planned and
managed from economic (agricultural) and social perspectives and landscape is
seen as only one asset amongst others.
At the end of the nineteenth century and early twentieth century, geographers
studied the rural landscape from a structural, functional and historical perspective
(Claval 2004). They considered it as a specific form of cultural landscape, often
focused on agrarian landscapes and settlement types and patterns. Towns and cities
were often excluded from these studies and demanded a specific approach at a more
detailed scale. Rural landscapes were seen as expression of a long history of land
organisation by humans, recognisable as successive layers of adaption according to
changing social needs and technology. Actual landscapes are conceived as a
palimpsest to be read in a retrogressive way (Claval 2004).
The general urban sprawl and suburbanisation of the countryside, in particular
after the Second World War, blurred the neat difference between urban and rural.
New complex, highly dynamical, fragmented and multifunctional landscapes
46 3 The Multiple Meanings of Landscape

emerged and are denoted as peri-urban, suburban and rurban landscapes (Antrop
2000) or also the ‘new rural’ (Gulinck 2004). Besides morphological urbanisation,
processes of functional urbanisation change local communities even in remote rural
places (Van Eetvelde and Antrop 2004). The urbanised rural landscapes are embed-
ded in urban networks or urban fields. Simultaneously, vast areas outside these
urban fields, located in the periphery, are affected by marginalisation and land
abandonment, which made David Lowenthal (1997) refer to these areas as the ‘rural
Also ‘the urban’ needed new definitions. Highly dynamical urban areas cannot
be understood by just looking at the morphological urban areas; also functional
relations and processes have to be included. Concepts of Functional Urban Regions
(Cheshire 1995), Functional Urban Areas (Antikainen 2005; OECD 2012; Peeters
2011) and Urban Morphological Zones (EEA 2002) were introduced.

3.4.3 Ordinary and Spectacular Landscapes

The protection of natural monuments and landscapes focused first on spectacular

and unique landscapes, such as the Grand Canyon, but also palaces and gardens
such as Versailles. Their uniqueness and exceptional qualities were considered
being of ‘national’ or ‘universal’ importance and are mainly based on aesthetic
arguments and symbolic and historical meaning. Consequently, the protected areas
are rather limited in extent and atypical for the character of the wider landscape they
are located in. Most of the landscape is not heavily protected, but often a distinction
is made between ‘valuable’ and ‘ordinary’ landscapes. Paul Groth (Groth and
Bressi 1997) used the term ‘ordinary’ landscapes for all new landscapes that lack
a pronounced identity, but are heterogeneous and chaotic, and are the results from
modern, not concerted processes and pragmatic, short-term solutions. The ‘valu-
able’ landscapes correspond to traditional rural areas, to cultural landscapes with a
pronounced character and ecologically important areas. The historical geographer
Richard Muir (2000) noted that ironically most of the traditional rural landscapes
we appreciate today, are not the result of planning and designing, such as it was
with gardens and parks, but resulted from the labour of peasants and local commu-
nities for whom surviving was the most important motive for their actions.

3.4.4 Landscape and the Beautiful, the Sublime,

the Picturesque and the Pictorial

The aesthetic aspects of the landscape arose esoterically discussions about beauty
and related concepts such as the sublime, the picturesque and the pictorial (Bell
1999) (see also Chap. 6). These concepts were particularly important in the design
3.4 Landscape with Adjectives 47

of estate parks and landscapes, as well as in establishing national parks and

monuments. In the eighteenth century, philosophers studied the beautiful and its
relationship between the object observed and the observer. Although the aesthetic
sensation is essentially placed in the eye’s of the beholder, also aesthetic qualities in
the object, the landscape, were important to understand how they could initiate such
feelings and how this knowledge could be applied in design. Keys to attractiveness
and ‘grades of beautiful’ needed to be defined.
The Beautiful is one of the archetypal Forms (εἶδoς) or Ideas (ἰδεα) Plato uses in
his Theory of Forms. The Forms are the essential reality of all things in itself and
cannot be approached by sense perception but by pure thought. Something is
perceived beautiful – according to Plato – because of the presence (parousia) of
the Beautiful is in it, or it is sharing in (methexis) the Beautiful. Aristotle criticised
the Theory of Forms and associates the Beautiful with things well-made and moral
virtue. This is based on the meanings of the word to kalon, which is used in Greek
and is often translated as beauty: (1) things well-made and (2) actions well-done
(hence, the expression ‘that was a beautiful thing you did’). Consequently, beauty
belongs to sense perception and in relation to landscape to the harmony and
relations between its components and shapes, whether made by nature or people.
Edmund Burke (1757) introduced concept of the sublime in a treatise on
aesthetics A Philosophical Enquiry into the Origin of Our Ideas of the Sublime
and Beautiful. Burke argued that the sublime and the beautiful are clearly distinct.
He also focused on the physiological effects caused by the sensation of the sublime,
which he described as a dual emotion of fear and attraction, as “delightful horror”.
Immanuel Kant was inspired by Burke’s writings and developed the concept in an
early treatise on aesthetics Beobachtungen u€ber das Gef€ uhl des Sch€ onen und
Erhabenen (Observations on the Feeling of the Beautiful and Sublime) (Kant
1764) and later in his main work Kritik der Urteilskraft (Critique of Judgment)
(Kant 1790). Kant considers the beautiful and the sublime as two different feelings;
“the sublime moves, the beautiful charms” (“Das Erhabene r€ uhrt, das Sch€ one
reiz”, p. 47). First, he distinguished three kinds of the sublime: he called the
‘terrifying’ (Schreckhaft-Erhabene), the ‘noble’ (Edle) and the ‘splendid sublime’
The sublime is experienced when the feelings we have observing a landscape are
overwhelming and call timelessness and infinity (Schama 1995). These feelings are
often associated with the grandeur of the untouched, wild nature. Compared to the
beautiful, which can be small, the sublime always refers to the greatness of a
(spectacular) landscape and is related to scale and magnitude. Classical examples
are views of the Himalayas and Grand Canyon. Many Romantic paintings represent
sublime landscapes, which are often dramatic mountain peaks, dark forests or sea
The picturesque refers to ideal aesthetic qualities of the landscape as in a
painting, in particular as represented in the work of the seventeenth century
landscape painter Claude Lorrain (see also Chap. 11). The term was derived from
the Italian pittoresco, meaning, ‘in the manner of a painter’. The picturesque was
introduced by William Gilpin in his Essay on Prints (Gilpin 1768), where he
defined picturesque as “. . .a term expressive of that peculiar kind of beauty,
48 3 The Multiple Meanings of Landscape

which is agreeable in a picture”. The picturesque is situated between the aesthetic

ideals of the beautiful and the sublime. Gilpin made during his travels sketches and
notes of the landscapes he saw and formulated ‘principles of picturesque beauty’,
such as composition and texture. For Gilpin, the composition incorporates several
elements: a dark “foreground” with a “front screen“or “side screens”, a brighter
middle “distance”, and less distinctly even further “distance”. The composition also
unifies the whole. Texture could be “rough”, “varied”, or “broken”, but without
obvious straight lines. He considered nature good in producing textures and colours,
but less capable of creating perfect compositions, where the help of an artist was
needed. He also preferred low viewpoints as these emphasise the “sublime” and
landmarks, such as ruins, which add “consequence” to the scene.
Gilpin made the picturesque popular, in particular to the pleasure of tourists
looking for beautiful and sublime landscapes. Also, many English landowners
redesigned their gardens and parks based on Gilpin’s ideas. The picturesque style
in English gardening was part of the emerging Romanticism in the eighteenth
century, along with revived medievalism and Celticism. The style emphasizes
natural forms in the design (Turner 1998) and was exemplified by landscape
gardeners as Lancelot ‘Capability’ Brown (see also Chap. 11).
The pictorial landscape is a similar concept, but more recent. It emerged with
the art of photography in the late nineteenth century. ‘Image hunters’ searched
scenes of a landscape (and its inhabitants) that were considered being particularly
beautiful, nice or typical, iconic, traditional or even exotic. These, are the scenes
depicted on postcards and photographs (Fig. 3.5).

Fig. 3.5 Example of the nineteenth century-early 20th picturesque photographs by ‘image
hunters’ in remote rural areas in Europe: a family on the Island of Groix (Brittany) using dried
dung as combustible (
3.4 Landscape with Adjectives 49

3.4.5 Ephemeral and Seasonal Landscapes

The way landscape is manifested to the observer is highly dependent on short

events and cyclic changes caused by the changing daylight, atmospheric conditions,
the weather, seasons and the phenology of the vegetation (see also Chap. 6).
Brassley (1998) introduced the concept of ephemeral landscapes. Duration and
periodical events are variables in describing ephemeral landscapes. For example,
Brassley considered woodcutting ephemeral. Sometimes, also moving objects,
animals and people that are seen in the landscape are considered ephemeral.
Michael Jones makes the distinction between ephemeral landscapes and seasonal
landscapes (Jones 2007). Palang et al. (2007) refer to seasonal landscapes. Already
in 1929, Johannes Gran€o attempted to describe and map ephemeral changes in the
landscape around his summer house in Valosaari through the seasons (Gran€o and
Paasi 1997). His maps show changing colours, sounds, movement, etc. (Fig. 3.6).
Seasons define lifestyles (Dodgshon and Olsson 2007; Kizos 2007). Ephemeral
phenomena and seasons are also important factors in landscape perception and

Fig. 3.6 Example of J. Gran€o’s mapping of ephemeral phenomena in the proximity of his
summerhouse on Valosaari: invariable and variable colours. (1) permanent colour (coniferous
forest, rock faces, walls), (2) secondary permanent colour (fences, hayricks), (3) colour change
twice a year (vegetation covered by snow, deciduous forest, roofs), (4) colour changing more often
(water reflecting surroundings and covered by vegetation at times) (Gran€ o 1929)
50 3 The Multiple Meanings of Landscape

Fig. 3.7 Soil and crop marks show the former creeks before the polder was created (De Moeren,
Belgium, photo J. Semey 1988)

experiencing and are components in defining aesthetic attractiveness Stobbelaar

and Hendriks (2007). For archaeologists, ephemeral soil and crop marks are
valuable indicators for detecting sites (Fig. 3.7).

3.5 Formal Definitions

Formal definitions are based upon conventions and engage the parties who signed
it. Concerning landscape two formal definitions have an international realm.

3.5.1 Cultural Landscapes in the UNESCO World Heritage


In 1992 ‘cultural landscapes’ are introduced as a new category in the UNESCO

World Heritage Convention. They are described as to “represent the ‘combined
works of nature and of man’ designated in Article 1 of the Convention. They are
illustrative of the evolution of human society and settlement over time, under the
influence of the physical constraints and/or opportunities presented by their natural
3.5 Formal Definitions 51

environment and of successive social, economic and cultural forces, both external
and internal.” (UNESCO 1992). Three main categories are recognised:
1. designed landscapes have been created intentionally by man, such as gardens
and parkland landscapes. They are constructed for aesthetic (and sometimes
political) reasons and are often associated with monumental ensembles.
2. organically evolved landscapes are the result of and have developed from the
interactive process between a specific culture and in response to its natural
They fall into two sub-categories:
(a) relict (or fossil) landscapes are the ones that still show characteristic mate-
rial features resulting from the processes that made them but came to an end;
(b) continuing landscapes are the ones that are sustained by a still active
traditional way of life in the contemporary society;
3. associative cultural landscapes refer symbolically to powerful religious, artistic
or cultural associations of the natural element rather than material cultural
Categories (1) and (3) are clearly ‘special’ landscapes which are often considered
spectacular or sublime. Category (2) deals with traditional agrarian and pastoral
landscapes, which constitute the main part of characteristic landscape diversity in
the countryside. Important problems related to this category are already recognised
in the two subcategories. When lifestyle changes and does not sustain anymore the
landscape it created over centuries, the living landscape dies and becomes a relict.
As lifestyles are now changing profoundly, many landscapes of this category had to
be moved to the list of endangered world heritage. Obviously, protection and
conservation of cultural landscapes is not easy (Palang and Fry 2003), and maybe
pure conservation is not a sustainable option. The description of the categories also
shows that the UNESCO World Heritage is not dealing with the ordinary, every-day
landscapes. If these have heritage values, these are rarely considered “outstanding”
and “universal”. Looking at the World Heritage properties designated as ‘cultural
landscape’ one can see that many are associated with natural monuments and nature
parks, which often give a sublime or majestic setting to remaining cultural struc-
tures. The real motive to use the terminology ‘cultural landscape’ lies in the need to
differentiate between (sometimes competing) policy domains that might overlap, as
for example the subject landscape does between UNESCO and IUCN (Fig. 3.8).

3.5.2 The European Landscape Convention Backgrounds and History

The European Landscape Convention of the Council of Europe (ELC for short),
also referred to as the Convention of Firenze, was opened for signature on October
52 3 The Multiple Meanings of Landscape

Fig. 3.8 (a) Uluru (Ayers Rock) in central Australia is listed on the UNESCO World Heritage List
as an associative cultural landscape because of its religious significance for the Aboriginal
community. (b) It became a touristic attraction, gathering daily hundreds of tourists in the bush
to admire the sublime spectacle of the red glow on the rock at sunrise and sunset (Photos M. Antrop

20th, 2000 and entered into force after 10 member states ratified the Convention,
which was already on March 1st, 2004. At the end of 2015 the Convention was
ratified by 38 of the 47 member states, and two more signed. The aims of the
Convention are the promotion of landscape protection, management and planning
and to organise European co-operation on landscape issues (Article 3) (Council of
Europe 2000).
3.5 Formal Definitions 53

The great merit of this Convention is that it initiated many programmes for
studying the landscape in most European countries as never before. This is remark-
ably as the Convention has not a legal basis such as an EU-directive and no financial
means are provided. The Council of Europe was founded in 1949 as a post-war
political organisation to enforce the human rights and to promote the unity and
co-operation between the 46 European member states, and has mainly an ethical
authority. The ELC had a long preparation and was inspired by the Dobrˇíš Assess-
ment (Stanners and Bourdeau 1995) as indicated in the Preamble. The Convention
also introduced a series of formal definitions ‘for the purposes of the Convention’,
as well as a series of recommendations, which give a common and international
basis for action. The definition is not only a consensus between the ministers of the
members of the Council of Europe, but is also supported by positive recommenda-
tions of the committees on diversity and landscape (CO-DBP) and cultural heritage
(CC-PAT). Definitions and Scope

Although the definition of landscape in the ELC is quoted most often, article
1 formulates several related definitions that are as important. These are:
1. “Landscape is defined as an area, as perceived by people, whose character is the
result of the action and interaction of natural and/or human factors;
2. Landscape policy means an expression by the competent public authorities of
general principles, strategies and guidelines that permit the taking of specific
measures aimed at the protection, management and planning of landscapes;
3. Landscape quality objective means, for a specific landscape, the formulation by
the competent public authorities of the aspirations of the public with regard to
the landscape features of their surroundings;
4. Landscape protection means actions to conserve and maintain the significant or
characteristic features of a landscape, justified by its heritage value derived from
its natural configuration and/or from human activity;
5. Landscape management means action, from a perspective of sustainable devel-
opment, to ensure the regular upkeep of a landscape, so as to guide and
harmonise changes which are brought about by social, economic and environ-
mental processes;
6. Landscape planning means strong forward-looking action to enhance, restore or
create landscapes.”
The definition of landscape is a consensus and contains most of the etymological
meanings of the original word. It refers to a territory or land unit, to the perception
people have of it but broader than the scenic aspects alone. The character refers to
its holistic quality and also defines the identity and uniqueness. Finally, the inter-
action between natural processes and human activities makes the landscape a
dynamic, always changing phenomenon. The perspective is clearly human-centred,
54 3 The Multiple Meanings of Landscape

which makes it more particular than the organism-centred focus as used in land-
scape ecology (Fahrig 2005; Mac Nally 2005).
It is noteworthy to look also at the somewhat different formulation of the
definition of landscape in the Explanatory Report of the ELC:
‘Landscape’ is defined as a zone or area as perceived by local people or visitors, whose
visual features and character are the result of the action of natural and/or cultural (that is,
human) factors. This definition reflects the idea that landscapes evolve through time, as a
result of being acted upon by natural forces and human beings. It also underlines that a
landscape forms a whole, whose natural and cultural components are taken together, not
separately. (Point 30;

Two groups of observers are explicitly mentioned, ‘local people’ and ‘visitors’,
which correspond also to the ‘insiders’ and ‘outsiders’ (Selman 2004). The visual
aspect is emphasised and the character is the expression of the ‘whole’ landscape
represents. Also, the dynamical evolution and history of the landscape is given
more explicitly, and clearly landscape is seen as a holistic entity.
All other definitions in the Convention refer to actions humans should undertake
regarding the landscape. The emphasizes lies in following points:
• sustainable development: ‘to harmonise changes caused by social, economic and
environmental processes’
• participation of the public: the aspirations of the public’
• the role of the authorities: ‘competent public authorities’
• defining qualities and values: ‘significant or characteristic features’, ‘heritage
• various actions: protection (‘to conserve and maintain’), management (‘upkeep’,
harmonise’), planning (‘enhance, restore or create landscapes’).
None of the definitions indicate how these goals could be achieved. Articles 4–6
describe the measures that should be undertaken and these clarify the background of
these definitions. Article 4 defines the responsibilities of the member states when
implementing the ELC in their national and regional policies. The focus is clearly
on local self-government and ‘respecting the principle of subsidiarity’. This is
explicated in Article 5 describing the general measures each party should under-
take. Here the following societal meaning of landscape is given as basis: “to
recognise landscapes in law as an essential component of people’s surroundings,
an expression of the diversity of their shared cultural and natural heritage, and a
foundation of their identity” (Art. 5.a). Again, the importance of participation is
stressed: “to establish procedures for the participation of the general public, local
and regional authorities, and other parties with an interest in the definition and
implementation of the landscape policies”. (Art. 5.c) Finally, the meaning of
landscape is not restricted to a sector approach but as an integrator: “to integrate
landscape into its regional and town planning policies and in its cultural, environ-
mental, agricultural, social and economic policies, as well as in any other policies
with possible direct or indirect impact on landscape.” (Art. 5.d).
The specific measures (Art.6) include awareness raising of the values of land-
scapes, multidisciplinary training and education, as well as improving the
3.5 Formal Definitions 55

knowledge of landscapes through programmes of identifying and analysing land-

scapes, monitoring changes, assigning values and defining landscape quality objec-
tives. Again, the participation of the population in all these matters is stressed.
Article 2 defines the scope of the ELC, which encompasses the entire territory of
the member states and “covers natural, rural, urban and peri-urban areas. It includes
land, inland water and marine areas.” The important difference with the UNESCO
Cultural Landscapes and many legally protected landscapes is the concern for all
landscapes: the outstanding ones as well as the everyday, ordinary or degraded
landscapes. This is certainly the most distinctive and original of the ELC, which the
Explanatory Reports justifies as follows:
The Convention’s original feature is that it applies to ordinary landscapes no less than to
outstanding ones, since all decisively influence the quality of Europeans’ surroundings.
Thus everyday, outstanding and damaged landscapes all come within its scope. This
comprehensive coverage is justified for the following reasons: every landscape forms the
setting for the lives of the population concerned; urban and rural landscapes interlock in
complex ways; most Europeans live in towns and cities (large or small), the quality of
whose landscapes greatly affects their lives; and rural landscapes occupy an important
place in the European consciousness. It is also justified by the profound changes which
European landscapes, particularly peri-urban ones are now undergoing. (Point 43; http://

3.5.3 Shifting Landscape Research Since the Coming

of Formal Definitions

The growing interest in Europe for the landscape since the ELC is unequivocal.
Although the value of the traditional, natural and rural landscapes as heritage and
their meaning for quality of life was already recognised in law in many countries,
their designation was restricted to rather small areas considered highly valuable.
There was no comprehensive policy for all landscapes. One of the first effects of the
ELC on landscape research resulted from Article 6. “Special measures/Identifica-
tion and assessment”. This led to the making of new landscape inventories and
characterization, often resulting in ‘landscape atlases’ or ‘landscape catalogues’.
Two types can be recognized: (i) the GIS-based atlases giving a searchable collec-
tion of thematic map layers and often web-based, and (ii) more monographic
descriptions, well illustrated with maps and iconographic material, also referred
to as ‘landscape biographies’. Most of the atlases refer to the regional and national
scales, although some small-scale pan-European classifications were made as well.
A second important shift in landscape research comes from the emphasis the
ELC puts on the importance of all types of landscape for the public (‘as perceived
by people’, ‘the public’s aspirations’, etc.). This stimulated research in landscape
perception and preference as well as processes of participation. So far, this research
showed the difficulty and complexity in defining ‘the public’ and its aspirations.
Also, the information needed in participatory planning processes involving many
different stakeholders with different interests, demands a more appropriate
56 3 The Multiple Meanings of Landscape

translation of scientific knowledge allowing easy and clear communication, some-

thing many researchers have difficulties with (Jones and Stenseke 2011).
Another effect of the ELC is the emergence of new networks dealing with the
landscape, obviously with a specific European scope. Landscape Europe and Land-
scape Tomorrow are examples. Most of them started between 2003 and 2006 but
many of these networks became inactive when the funding programmes finished.
UNISCAPE (2008), CIVILSCAPE (2008), RECEP-ENECL (2006–2015) were
founded to encourage and support respectively universities, non-governmental
organisations and local and regional authorities in the implementation of the
European Landscape Convention. Other networks were set up in relation with a
European project funded by the European Commission (VOLANTE, HERCULES,
CHeriScape, among others). In older associations, a new European focus can be
noticed as well: EFLA emerged within IFLA, forming IFLA Europe, and IALE-
Europe was created within IALE. Most of these networks aim to pool interdisci-
plinary expertise and to develop partnerships. They focus on specific problems and
situations in Europe and offer applied research for planning and managing land-
scapes in a more holistic scope and in the perspective of sustainability. In addition,
they often add education and training both at international and local scale.
Today, many policy levels, sectors, interest groups and scientific disciplines are
involved in the landscape, making it a complex multi-layered business, with inter-
and transdisciplinary processes that sometimes interact, sometimes compete and
still too rarely give consistent results. In this complex ‘policy landscape’ the real
landscape is often the only integrating concept. In general, landscape research – in
particular in Europe – became more applied, more society oriented and less
theoretical and academic. Landscape studies diversified with varying depth and
quality, ranging from rigorous scientific analysis to almost pseudo-scientific papers
aimed at a broad public. More and more scientific disciplines borrow methods and
techniques from others, especially when they offer ‘innovation’ in their domain,
even when applied in a more amateurish way.
However, the unmistakable shifts that occurred in landscape research after 2000
cannot all be related to the ELC. Other reasons are found in the landscape changes,
which became unprecedented devastating and happen in a still accelerating pace.
Methods to study and monitor these changes need to be fast, comprehensive and
reliable. Solving specific, acute problems and strict deadlines dictate this kind of
more applied and policy-oriented research. Consequently, academic research and
applied landscape research are diverging. The academic merit system enforces this
process as well. Local and specific problem solving is seldom innovative on
methodological and theoretical aspects, and thus less suited to be published in
international, peer-reviewed journals. Similarly, commissioners of landscape stud-
ies are not interested in theoretical and abstract issues. They ask for practical
reports, which are often kept confidential as long as the participation procedures
are running and no political decisions are taken. With this shifting focus in research
goals, the funding sources shift as well. The mission of universities is pure research
with an international realm and producing PhDs. Administrations are rather reluc-
tant in funding doctoral research and prefer commissioning practical projects to
agencies, private companies and NGO’s, where also landscape experts are found.
References 57

3.6 Elements for an Operational Definition of Landscape

From the above discussed definitions, following elements emerge as constants in

defining the landscape. Landscape:
• is seen as a spatial entity, having a variable extent and scale, and has territorial
• is that what is perceived and experienced by humans,
• is composed of many very different elements and components that interact and
are hierarchically structured,
• has a spatial organisation and management that is largely influenced by humans,
• is dynamic and changes are an inherent property of it.
All these properties make the landscape a dynamic and holistic phenomenon,
hierarchically structured and scale dependent (Antrop 2000, 2005, Naveh and
Lieberman 1994). This makes landscape fundamentally different from the concept
‘land’ (Zonneveld 1995), which is considered as a piece of terrain, bounded in space
and bordered and very often owned by someone or some institution. Land refers to
(private) property that can be used more or less freely by its owner. Consequently, it
is important to consider who has the power, the spatial competence to make changes
(Hägerstrand 2001). Landscape on the contrary, as a common heritage of a com-
munity (Antrop 2005, Claval 2004). It became an integrative concept in spatial
planning, sustainable development, environmental management and heritage pro-
tection (Palang et al. 2004; Tress et al. 2003; Antrop 2003; Fry 2003; Fairclough
and Rippon 2002; Haines-Young 2000), demanding a transdisciplinary approach
(Tress et al. 2003; Naveh 2000).


Antikainen, J. (2005). The concept of functional urban area – Findings of the ESPON project 1.1.1.
Informationen zur Raumentwicklung, Heft, 7, 447–452.
Antrop, M. (2000). Changing patterns in the urbanized countryside of Western Europe. Landscape
Ecology, 15(3), 257–270.
Antrop, M. (2003). Expectations of scientists towards interdisciplinary/transdisciplinary research.
In B. Tress, G. Tress, & G. Fry (Eds.), Interdisciplinary and transdisciplinary landscape
studies: Potential and limitations, Delta series 2 (pp. 44–54). Wageningen: Delta Program.
Antrop, M. (2005). Why landscapes of the past are important for the future. Landscape and Urban
Planning, 1–2, 21–34. (2001).
Bell, S., 1999. Landscape. Pattern, perception and process. Taylor & Francis Group London.,
Berque, A. (1982). Vivre l’espace au Japon. Paris: PUF.
Brassley, P. (1998). On the unrecognized significance of the ephermerical landscape. Landscape
Research, 23(2), 119–132.
Burke, E. (1757). A philosophical inquiry into the origin of our ideas of the sublime and beautiful,
Harvard classics, Vol.24, Part 2 (pp. 1909–1914). New York: P.F. Collier & Son.
58 3 The Multiple Meanings of Landscape

Cheshire, P. (1995). A new phase of urban development in Western Europe? The evidence for the
1980s. Urban Studies, 32(7), 1045–1063.
Claval, P. L. (2004). The languages of rural landscapes. In H. Palang, H. Sooväli, M. Antrop, &
S. Setten (Eds.), European rural landscapes: Persistence and change in a globalising envi-
ronment (pp. 11–40). Dordrecht: Kluwer Academic Publishers.
Cosgrove, D. (2003). Landscape: Ecology and semiosis. In H. Palang & G. Fry (Eds.), Landscape
interfaces. Cultural heritage in changing landscapes (pp. 15–20). Dordrecht: Kluwer Aca-
demic Publishers.
Cosgrove, D. (2004). Landscape and landschaft. Lecture at the “Spatial Turn in History” Sympo-
sium German historical institute. GHI Bulletin, 35, 57–71.
Cosgrove, D., & Daniels, S. (1988). The iconography of landscape. Cambridge: Cambridge
University Press.
Council of Europe. (2000). European landscape convention. Firenze, 20.10.2000 http://conven
Dodgshon, R. A., & Olsson, E. G. A. (2007). Seasonality in European mountain areas: A study in
human ecology. In H. Palang, H. Sooväli, & A. Printsmann (Eds.), Seasonal landscapes,
Landscape series (Vol. 7, pp. 85–101). Dordrecht: Springer.
EEA. (2002). Urban morphological zones 2000 (UMZ2000).
maps/data/urban-morphological-zones-2000-umz2000-f1v0. Consulted 24 Apr 2013.
European Commission. (2008).
Fahrig, L. (2005). When is a landscape perspective important? In J. A. Wiens & M. R. Moss (Eds.),
1999. Issues in landscape ecology, Studies in landscape ecology (pp. 3–10). Cambridge:
Cambridge University Press.
Fairclough, G., & Rippon, S. (Eds.) (2002). Europe’s cultural landscape: Archaeologists and the
management of change. Europae Archaeologiae Consilium, Occasional paper 2, Brussels.
Fry, G. (2003). From objects to landscapes in natural and cultural heritage management: a role for
landscape interfaces. In H. Palang & G. Fry (Eds.), Landscape Interfaces. Cultural heritage in
changing landscapes (pp. 15–20). Dordrecht: Kluwer Academic Publishers.
Gilpin, W. (1768). An essay on prints: Containing remarks upon the principles of picturesque
beauty; the different kinds of prints; and the characters of the most noted masters.
Gran€o, J. G. (1929). Reine Geographie. Eine methodologische Studie beleuchtet mit Beispielen
aus Finnland und Estland. Helsinki, Helsingfors, Acta Geographica, 2(2), 202 pp.
Gran€o, O., & Paasi, A. (Eds.). (1997). Pure geography. J.G. Gran€ o., translated by M. Hicks.
Baltimore/London: The John Hopkins University Press. 191 pp.
Groth, P., & Bressi, T. W. (Eds.). (1997). Understanding ordinary landscapes (272 pp). New
haven: Yale University Press.
Gulinck, H. (2004). Neo-rurality and multifucntional landscapes. In J. Brandt & H. Vejre (Eds.),
Multifunctional landscapes. Theory, values and history (Vol. Vol. I, pp. 63–74). Southampton:
WIT Press.
Hägerstrand, T. (2001). A look at the political geography of environmental management. In
A. Buttimer (Ed.), Sustainable landscapes and lifeways: Scale and appropriateness
(pp. 35–58). Sterling: Cork University Press.
Haines-Young, R. H. (2000). Sustainable development and sustainable landscapes: Defining a new
paradigm for landscape ecology. Fennia, 178(1), 7–14.
Hartshorne, R. (1939). The nature of geography. A critical survey of current thought in the light of
the past. Annals of the Association of American Geographers, 29(3–4), 173–658.
Hartshorne, R. (1961). The nature of geography. Reprinted with corrections from the Annals of the
Association of American Geographers, 504 pp.
Howard, P. J. (1991). Landscapes: The artists’ vision. London: Routledge.
Howard, P. J. (2011). An Introduction to landscape (309 p). Farnham: Ashgate.
Jones, M. (2003). Human geographical landscapes: J.G. Gran€ o’s approach to landscapes as
scientist and artist. In O. Gran€o (Ed.), Origin of landscape science. J.G. Gran€ o an a new
pure geography for a new state (pp. 71–98). Turku: The Turku University Foundation.
Jones, M. (2005). Law and landscape – Some historical-geographical studies from Northern
Europe. In M. Jones & T. Peil (Eds.), Landscape, law and justice (pp. 95–109). Odo: Novus.
References 59

Jones, M. (2007). Seasonality and landscape in Northern Europe: An introductory exploration. In

H. Palang, H. Sooväli, & A. Printsmann (Eds.), Seasonal landscapes, Landscape series (Vol.
7, pp. 17–60). Dordrecht: Springer.
Jones, M., & Stenseke, M. (Eds.). (2011). The European landscape convention. Challenges of
participation, Landscape series, Vol. 13 (326 pp). Dordrecht: Springer.
Kant, I. (1764). Beobachtungen u€ber das Gef€ uhl des Sch€onen und Erhabenen. Project Gutenberg.
Kant, I. (1790). Kritik der reinen Vernunft. Project Gutenberg.
Kizos, T. (2007). Island lifestyles in the Aegean Islands, Greece: Heaven in summer, hell in
winter? In: Jones, M., 2007. Seasonality and landscape in Northern Europe: An introductory
exploration. In H. Palang, H. Sooväli, & A. Printsmann (Eds.), Seasonal landscapes, Land-
scape series (Vol. 7, pp. 127–149). Dordrecht: Springer.
orzing, H. (2001). The nature of landscape. A personal quest (176 pp). Rotterdam:
010 Publishers.
Lowenthal, D. (1975). Past time present place: Landscape and memory. The Geographical Review,
65(1), 1–36.
Lowenthal, D. (1997). European landscape transformations: The rural residue. In P. Groth & T. W.
Bressi (Eds.), Understanding ordinary landscapes (Yale University Press ed., pp. 180–188).
New Haven.
Luginbühl, Y. (2012). La mise en scène du monde. Construction du paysage européen (p. 432).
Paris: CRNS Editions.
Mac Nally, R. (2005). Scale and an organism-centric focus for studying interspecific interactions
in landscapes. In J. A. Wiens & M. R. Moss (Eds.), 1999. Issues in landscape ecology, Studies
in landscape ecology (pp. 52–69). Cambridge: Cambridge University Press.
Makhzoumi, J. M. (2002). Landscape in the Middle East: An inquiry. Landscape Research, 27(3),
Mathewson, K. (2011). Landscape versus region – Part II. In J. Agnew & J. S. Duncan (Eds.), The
Wiley-Blackwell companion to human geography (pp. 130–145). Oxford: Wiley-Blackwell.
Muir, R. (1999). Approaches to landscape (310 pp). London: MacMillan Press.
Muir, R. (2000). The new reading the landscape. Fieldwork in landscape history (256 pp). Exeter:
University of Exeter Press.
Naveh, Z. (2000). What is holistic landscape ecology? A conceptual introduction. Landscape and
Urban Planning, 50, 7–26.
Naveh, Z., & Lieberman, A. (1994). Landscape ecology: Theory and application (360 pp).
New York: Springer Verlag.
OECD. (2012). Defining functional urban areas (FUA) for the OECD metropolitan database.
OECD, 9 pp.
Olwig, K. R. (1996). Recovering the substantive nature of landscape. Annals of the Association of
American Geographers, 86(4), 630–653.
Olwig, K. R. (2002). Landscape, nature and the body politic: From Britain’s renaissance to
America’s new world (299 pp). Madison: University of Wisconsin Press.
Olwig, K. (2004). A theoretical reflexive approach to the historical aspects of multifunctionality in
landscapes: Opposing views. In J. Brandt & H. Vejre (Eds.), Multifunctional landscapes.
Theory, values and history (Vol. I, pp. 115–136). Southampton: WIT Press.
Olwig, K. (2005a). Editorial: Law, polity and the changing meaning of landscape. Landscape
Research, 30(3), 293–298.
Olwig, K. (2005b). The Landscape of ‘customary’ law versus that of ‘natural’ law. Landscape
Research, 30(3), 299–320.
Olwig, K. (2013). The law of landscape and the landscape of law: The things that matter. In
P. Howard, I. Thompson, & E. Waterton (Eds.), The Routledge compagnion to landscape
studies (pp. 253–262). London: Routledge.
Palang, H., & Fry, G. (Eds.). (2003). Landscape interfaces. Cultural heritage in changing
landscapes (406 pp). Dordrecht: Kluwer Academic Publishers.
60 3 The Multiple Meanings of Landscape

Palang, H., Sooväli, H., Antrop, M., & Setten, S. (2004). European rural landscapes: Persistence
and change in a globalising environment (p. 482). Dordrecht: Kluwer Academic Publishers.
Palang, H., Sooväli, H., & Printsmann, A. (Eds.). (2007). Seasonal landscapes, Landscape series
(Vol. 7, p. 256). Dordrecht: Springer.
Peeters, D. (2011). Functional urban regions database. Technical Report, ESPON, 18 pp.
Sauer, C.O.. (1925). The morphology of landscape. University of California Publications in
Geography, 2; Reprinted in Leighly, J. (ed.), 1974. Land and life (pp. 315–350). Berkeley:
University of California Press.
Schama, S. (1995). Landscape and memory (p. 652). New York: A. Knopf.
Selman, P. (2004). Barriers and bridges to sustaining cultural landscapes. In R. H. G. Jongman
(Ed.), The new dimensions of the European landscape, Wageningen UR frontis series vol. 4
(pp. 93–102). Dordrecht: Springer.
Stanners, D., & Bourdeau, Ph. (eds). (1995). Europe’s Environment. The Dobris Assessment.
European Environment Agency, EC DG XI and Phare, Copenhagen, pp. 676.
Stobbelaar, D. J., & Hendriks, K.. (2007). Seasonality of agricultural landscapes: Reading time and
place by colours and schapes. In: Jones, M., 2007. Seasonality and landscape in Northern
Europe: An introductory exploration. In: Palang, H., Sooväli, H., Printsmann, A., (eds), 2007.
Seasonal Landscapes. Springer, Landscape Series, Vol.7, pp. 103–126.
Tress, G. (2003). Expectations of policy-makers, funding bodies and end-users towards interdis-
ciplinary and transdisciplinary research. In B. Tress, G. Tress, A. van der Valk, & G. Fry
(Eds.), Interdisciplinary and transdisciplinary landscape studies: Potential and limitations,
Delta series 2 (18 p). Wageningen: Delta Program.
Turner, T. (1998). Landscape planning and environmental impact design (425 pp). London: UCL
UNESCO. (1992).
Van Eetvelde, V., & Antrop, M. (2004). Analyzing structural and functional changes of traditional
landscapes – two examples from Southern France. Landscape and Urban Planning, 67(1–4),
Vos, W.. (2000). The history of European landscape painting. In: Klijn, J., Vos, W. (eds.), From
Landscape ecology to landscape science (pp. 81–96). Wageningen: Kluwer Academic
Zonneveld, I. S. (1995). Land ecology (199 pp). Amsterdam: SPB Academic Publishing bv.
Chapter 4
Approaches in Landscape Research

Abstract The complexity of landscape and its multiple meanings make that it is
conceptualized differently according to the approach followed: as a complex spatial
system of phenomena in interaction, as a scene or image that can be described using
rules of perception, and as an existential phenomenon with strong symbolic mean-
ings and values. The bird’s-eye perspective offers a synoptic and detailed view of
the visible aspects of the landscape with a dynamic scale by zooming in and out.
This vertical perspective is often indirect and distant from the landscape as in aerial
photography and cartography. The interior perspective is the one how most people
experience landscapes every day. The inner perspective creates mindscapes that
influence the way we value landscapes. The transcendental perspective sees land-
scape as holistic phenomenon and focuses on the meta-reality generated by the
composing parts, such as coherence and complexity. Scientific disciplines combine
in specific ways these different approaches. Following approaches are discussed:
geography, ecology, landscape ecology, history, historical ecology, archaeology,
environmental psychology and landscape architecture, as well as possibilities for
inter- and transdisciplinary research.

Keywords Ways of seeing • Viewpoint • Bird’s-eye perspective • Interior

perspective • Inner perspective • Discipline • Interdisciplinary • Transdisciplinary

4.1 Introduction

In this chapter, we discuss principles that form the basis for the different approaches
in landscape research. Two important aspects are the viewpoints taken by the
observer of the landscape (‘ways of seeing’ as Cosgrove (2002) called them) and
the framework of the scientific discipline involved or the context in which the
research is done.

© Springer Science+Business Media B.V. 2017 61

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_4
62 4 Approaches in Landscape Research

4.2 Ways of Seeing

Cosgrove (2002) said that “landscape denotes primarily geography as it is seen,

imaged and imagined” and that the evolution of landscape meanings also depended
on changing technologies in sensing and representing our environment. The viewer
of the landscape chooses a specific viewpoint and selects what to see and how.
The complexity of landscape and its multiple meanings make that it is concep-
tualized differently according to the approach followed. Landscape can be seen as:
• a complex spatial system of objects (elements) and continuous phenomena in
interaction. In this approach following concepts are used: structure, pattern,
functions, ecosystem, change, dynamics. The systems theory is the most impor-
tant paradigm in this approach. These concepts can be described, sometimes
measured and analysed using landscape metrics and indicators.
• a scene or image that can be described using rules of perception. Basic concepts
are: views, view-sheds, isovists, vistas, perspectives, but also concepts related to
preference such as aesthetics, openness, naturalness, disturbance, etc. Theories
of environmental perception and Gestalt-psychology are applied as well as
design principles.
• an existential phenomenon with strong symbolic meanings and values. Basic
concepts used in this context are: home(land), heritage, history, genius loci,
character, landmarks, social construct, narratives, etc. These are approaches of
arts, philosophy, humanistic geography and sociology.
According to the viewpoint of the observer, four perspectives can be recognized
Fig. 4.1):
• a viewpoint from above offers a bird’s-eye perspective, looking from ‘outside’ to
the landscape in a vertical or oblique way;
• a viewpoint ‘interior’ in the landscape offers mainly a horizontal perspective: the
way most people perceive and experience the landscape;
• an ‘inner’, mental perspective offers mental images (mindscapes) of the land-
scape and allows representations and visualisation of the landscape, such as
mental maps;
• a transcendent, abstract perspective: the landscape as a holistic meta-reality.

4.2.1 The Bird’s-Eye Perspective: Landscapes at a Distance

The perspective from above uses a real or virtual viewpoint distant from the actual
landscape. It is the bird’s-eye perspective as can be seen directly from a high
position and from the air. As an indirect observation, aerial photographs and
satellite imagery also give this perspective. It offers a synoptic and detailed view
4.2 Ways of Seeing 63

Fig. 4.1 Viewpoints and perspectives

of the visible aspects of the landscape with a dynamic scale by zooming in and out.
The scale defines the extent of the view and the degree of detail (the ‘grain’)
observed. The vertical perspective is commonly used in cartography. Maps show
a conceptualised representation of the landscape according to mapmaker’s rules.
This perspective clearly shows spatial patterns and the context of distinct
elements. Often also the hierarchical structure and composition are revealed as
well as the coherence and relations between the constituent parts. The information
content is usually very high and allows the formulation of hypotheses about
processes that are active in the landscape, about its history and the land use.
However, this perspective is literally distant, since the information used is
mainly visual and the observer has no direct contact with the landscape. Knowledge
about the landscape is derived from the interpretation of spatial patterns.
Humans instinctively looked for high viewpoints offering bird’s-eye perspec-
tives over the landscape. Mountaintops and towers gave an oblique perspective
over a vast area, allowing orientation and cognitive mapping. These were the
viewpoints of the early mapmakers. Since the development of photography, it is
also possible to register more objectively this synoptic view. In 1858, Gaspard Felix
Tournachon alias Nadar took the first aerial photograph from a balloon near Paris.
Aerial photography offered unexpected views of the landscape and revealed pat-
terns and features that were unknown until then. The benefit of the new technology
was obvious and soon the most diverse devices were created to bring cameras in the
air: balloons, kites, pigeons and airplanes. Aerial photographs taken from the
Western front during the First World War demonstrated the potential for military
64 4 Approaches in Landscape Research

Fig. 4.2 Aerial photograph of Boekhoute (Belgium) in 1944. Settlement pattern and field systems
are represented in detail. The inundation enhances the micro-relief and shows the different
elevation between the polders (Photo Aerial Air Force)

reconnaissance and map making (Stichelbaut and Chielens 2014). Thus, the new
technology developed rapidly, in particular during the wars (Fig. 4.2). Stereoscopic
photogrammetry became the basis for a new approach in cartography and photo-
interpretation using stereovision for military intelligence and scientific research.
Orthophotographs are geometrically rectified and became a new type of map,
commonly used as base layers in Geographical Information Systems (GIS)
(Fig. 4.3a).
Aerial photographs offer a detailed and synoptic and simultaneous view of many
landscape components such as land use, vegetation, settlements, field systems and
landform emphasising their relations and coherence. This made Carl Troll say:
“aerial photography is in a high degree landscape ecology” (Troll 1939). Aerial
imagery allows also the detection of phenomena that cannot be observed easily
4.2 Ways of Seeing 65

Fig. 4.3 Orthophotograph (a) and oblique aerial photo (b) of the Sint Pietersveld in Flanders
(Belgium). Elements and objects are more easily identified on oblique than on vertical photo-
graphs. Both show spatial patterns and coherence between the elements. However, oblique
photographs are distorted and loose detail with the decreasing scale towards the background,
making them less practical to make measurements and perform quantitative analyses. (Copyright
(a) Eurosense; (b) Ghent University and J. Semey 1989)
66 4 Approaches in Landscape Research

Fig. 4.4 Archaeological soil and crop marks on an oblique aerial photograph in Aartrijke
(Belgium). Soil and crop marks are ephemeral phenomena that can be detected from a bird’s-
eye perspective. They are caused by differences in soil moisture, soil depth and stoniness that
influence the growth of the vegetation. Also, micro-relief can be detected from this viewpoint in
particular when the sun elevation is low (Copyright Ghent University, photo 54.525 J. Semey

from a ground position, such as soil and crop marks, which are indicative for
archaeology and micro changes in soil conditions and drainage and indications of
pollution. Aerial photography became a basic tool in the study of landscape, in
physical, historical, rural and regional geography, in land evaluation, as well as in
archaeological prospecting.
For surveying, often large-scale oblique photographs are used offering great
detail (Fig. 4.3b). Their registration is very flexible and fast, allowing interactively
testing the best exposure conditions and using them for monitoring and making
repeat photography of specific features. Applications as these are particularly
important in archaeological surveying (Dassié 1978), as demonstrated by the
intensive surveying in Britain by O. Crawford (1960), in Germany by Irwin Scollar
(1975), in Picardie (France) by Roger Agache (1978) and many others. Archaeo-
logical prospecting using aerial photography proved very fruitful, as for example
demonstrated by the work of Jacques Semey who took more than 70,000 pictures in
20 years over Flanders, revealing more than 650 unknown archaeological sites
(Ampe et al. 1996; De Reu et al. 2010) (Fig. 4.4).
However, the extent covered by one oblique photograph is limited and distor-
tions of the geometry and illumination conditions are important. For mapping
purposes, systematic stereoscopic vertical photographs are preferred. These often
4.2 Ways of Seeing 67

range from scales between 1:10,000 and 1:50,000. The vertical perspective
demands some training in photo-interpretation to recognise objects accurately and
understand the features shown, certainly when special films, such as false colour
infrared are used. Stereovision enhances the holistic view of the landscape, espe-
cially microtopography, due to the vertical exaggeration.
With small-scale imagery from satellites, synoptic views are possible at regional
to even global scales. The first generation satellite images were made for intelli-
gence purposes during the Cold War. An interesting series were the stereoscopic
Corona photographs made by the U.S. Air Force for strategic reconnaissance from
1959 to 1972, covering the USSR, China, the Middle East and other strategic areas.
The program was declassified in 1995 and the imagery became available for
scientific and civil use, revealing its significance in landscape archaeological
surveying, as the high resolution (approx. 2 m), stereoscopic photographs show
areas which were not yet mapped in detail at that time.
From 1967 on electronic imagery from earth observation satellites became
available for civil and scientific use. The spatial resolution of the first generation
imagery was insufficient to recognise individual objects and landscape elements.
However, the digital format of raster images stimulated the development of multi-
spectral image classification and spatial filtering techniques as well as digital
mapping and the development of raster-GIS. Successive improvements in the
remote sensing technology from the 1980s on allowed producing images showing
landscape structures in more detail, and from the 1990s on satellite imagery could
compete with the resolution of aerial photographs (Jensen 2000). Simultaneously,
image processing and classification expanded into the world of personal desktop
computing and gradually cheap image coverage became available to all. Systems as
Google Earth and Bing Maps nowadays offer detailed bird’s-eye images from
different sources, scales and periods available to all and offer in combination with
digital terrain models also oblique 3D-views from any viewpoint possible.

4.2.2 The ‘Interior’ Perspective: Being in the Landscape –

Lookouts and Composite Landscapes

Standing in the landscape offers the observer a horizontal perspective and this is the
way most people experience landscapes every day. The position and movement of
the observer are important variables in understanding the views. Elevated positions
offer panoramic lookouts of the landscape in an oblique view (Fig. 4.5). On flat
terrain, amidst vegetation and buildings, only small parts of the landscape can be
seen and a mental representation of the whole landscape has to be constructed from
assembling views at different positions. This mental map allows orientation in
space and gradually allows understanding relationships and patterns that compose
the landscape (Fig. 4.6).
Fig. 4.5 Lookout view from the hill Rodeberg (Belgium). Hilltops and towers offer an overview
of the landscape revealing some spatial arrangement and elements that can be identified easily.
Topography, buildings and vegetation block the view and much of the landscape structure remains
hidden (Photo M. Antrop 2005)

Fig. 4.6 The observer, standing in the landscape on flat ground and surrounded by objects that
mask the view has to combine mentally views from different positions in order to understand the
spatial configuration and the relations between the elements (Drongen, Belgium, Open Street Map;
photos M. Antrop)
4.2 Ways of Seeing 69

Standing in the landscape, the observer experiences his surroundings with all
senses. Even if the visual perception dominates, the resulting image becomes
‘coloured’ by the other senses. In the landscape, it makes a difference when one
is looking at nice scenery with a bad smelling waste dump or noisy motorway
behind the observer or not. This is a holistic experience that is difficult to register
using photography. Hence the discussion about the bias in using landscape photo-
graphs as ‘objective’ registration of the landscape and as the sole stimuli in
preference studies.
Experiencing landscape directly by standing in and moving through it is the way
we come to evaluate it as ordinary, picturesque, spectacular or sublime. It is the
perspective of everyone, but also of the poet, painter, photographer and publicity
maker who use the landscape in their expressions.

4.2.3 The Inner Perspective: Mindscapes and Visualisations

The inner perspective projects mental representations and memories upon the
landscapes we observe and also influence the way we visualise landscapes in
representations such as drawings, paintings and models. Growing up, we develop
mental mapping for orienting ourselves in our environment and to help us
interpreting and understanding the surroundings. This is a vital survival skill,
which gradually develops our ‘mental map’ composed of several mindscapes
according to our experience and the landscapes we visited.
The reality is not the physical landscape we observe, but its mental interpreta-
tion. We tend to focus on what we already know, on what is familiar and on what is
important in terms of safety and prospect. Thus we focus on fixed landmarks for
orientation, identify characteristic elements, detect risks and disturbances, and look
for relations and coherence. Recognition and understanding depends on the legi-
bility of the landscape. Landscape reading is determined by observer’s properties
as education, social status, etc. These are studied in landscape experience and
preference research (Sevenant 2010).
Mindscapes represent also ideal landscapes, integrating our knowledge, memo-
ries and feelings associated with places and regions. They define the image we have
of our ‘homeland’ and ‘home’ or ‘domestic landscape’, as well as the genius loci
(spirit) of a place (Fig. 4.7).
The aesthetic, existential and symbolic properties and preferences were mainly
studied in spectacular and sublime landscapes and from an artistic perspective.
Applications are found in landscaping estate gardens and parks. Early studies of the
rural and ordinary landscape rarely focused on these aspects. Only since the second
half of the twentieth century they became studied in a humanistic and sociological
approach. New concepts were introduced as well such as (sense of) place and
placenessness, non-places (non-lieux) and quality of space. The relations between
language and linguistics, and between landscape and place became more important.
Kenneth Olwig (1996, 2002) and Denis Cosgrove (1984, 1993) demonstrated how the
70 4 Approaches in Landscape Research

Fig. 4.7 The spirit of place: water sources were always considered mysterious. Such vital places
were marked by placing chapels and special trees nearby (Stambruges, Belgium) (Photo M. Antrop
2005). Many of these trees are fetish-trees, still venerated as shown by the ex-votos

landscape, as a concept and a representation, was intentionally used politically in

shaping territories according to ideology and to develop national identities and
stimulate nationalism. An extreme example is given by the planning and landscap-
ing rules for creating ‘ideal German’ landscapes in Nazi-Germany.
Our mindscape also influences the way landscapes are represented in maps and
2D or 3D computer visualisations. People became familiar with cartographic
representations of water in shades of blue according to the depth, lowlands in
green, hilly uplands in a range of yellow-orange and mountains in brown and
white for the tops. Recognising landforms on hill-shaded relief maps with virtual
illumination from the northwest, according to mapmaker’s conventions, causes no

4.2.4 Landscape as Meta-Reality: The Transcendental


Landscape is holistic, meaning that the whole is more than the sum of the compos-
ing parts. How to grasp that ‘more than the sum’? That is what the transcendental
4.3 Disciplines 71

approach to landscape attempts to do. The focus is not on the composing parts, but
on the meta-reality they generate. One is looking for characteristics beyond the
visible and physical landscape. How can the coherence between the composing
elements be described and measured? What do landscape diversity and heteroge-
neity mean? To study these meta-properties, two approaches are possible: a
philosophical-psychological approach and a parametric-reductionist approach.
The philosophical-psychological approach is based on the Gestalt-theory and
hierarchical system thinking. Complex landscape entities that function autono-
mously in some degree are seen as the building blocks of the landscape. They are
referred to as holons (Naveh and Lieberman 1994, Antrop 2004), black boxes
(‘Pandora boxes’) (Zonneveld 2005) or ecodevices (Van Wirdum 1981). Although
they possess some freedom in their functioning, they interact with each other and
are embedded in a multi-scale hierarchical system. Zev Naveh called it the Total
Human Ecosystem or THE (Naveh 2000).
The second approach is parametric and reductionist. Meta-properties, such as
diversity and coherence, are formally defined, as well as the parameters to describe
and measure these. The analysis uses statistics, models and thematic mapping for
visualising the meta-properties. Therefore, new geostatistics, landscape metrics and
indicators were developed. Special tools, such as Fragstats, became very popular in
the analysis of spatial patterns (McGarigal and Marks 1995; Li and Wu 2007).

4.3 Disciplines

4.3.1 Geography and Historical Geography

Landscape was a core subject of study in geography during its early development as
an empirical science. Landscape was seen as the synthesis of the interaction
between the natural environment and human society and characterised by unique
geographical regions. It resulted in the study of land use zoning and vegetation
patterns, of agrarian systems and settlement patterns, of hydrographical and trans-
portation networks, etc. The spatial diversity was explained by the variation in
ecological and cultural factors and in a dynamical perspective covering the geo-
logical evolution and history. It implied integrating sciences as geology, soil
science, botany, hydrology and geomorphology as well as demography, anthropol-
ogy, economy, politics and history. As such, geography was interdisciplinary ‘avant
la lettre’.
Important methodological developments were initiated in geography and later
used by most disciplines involved in landscape studies. Important contributions
from geography were found in field surveying, cartography and map analysis in
various forms, air photo and image interpretation, early development of spatial
analysis, modelling and geostatistics, and the conceptualisation of geographical
72 4 Approaches in Landscape Research

information systems (GIS). Many of these are common tools in other disciplines,
which also enhanced continued the methodological development.
Historical geography studies the evolution of (mainly cultural) landscapes using
maps and plans, written sources and iconographic material. Also, structures, ele-
ments and place names (toponyms) in the contemporary landscapes that witness
from the past are studied (Van Slembrouck et al. 2005). Early studies focus mainly
on the agrarian landscape. The actual landscape is seen as a palimpsest – an analogy
introduced by O.G.S. Crawford – a sheet of vellum used over and over again for
writing texts, each time erasing the older ones, but leaving some fragments between
the new text (Turner 2013).
Historical geography uses two approaches to study landscapes. The first focuses
on the reconstruction of the landscape in a given period, the other focuses on
trajectories of change, also referred as landscape paths. A complete integrated
history of the landscape in a certain region results in a landscape biography
(Kolen 2005). Classifying and mapping the actual landscape according to its
historical dimension is achieved in a Historic Landscape Characterisation (HLC)
(Rippon 2004).
When the study starts from relicts in the actual landscape and gradually tries to
reconstruct past situations, a retrospective method is used (Rippon 2012). When all
information is used to reconstruct as complete as possible a given period, the
method is called retrogressive. When making a landscape biography, the recon-
struction of past landscapes and their genesis is also linked to the study of practices
and technology used in agriculture and forestry, and also the political, social and
economic context is taken into consideration.

4.3.2 Landscape Ecology

Ecological thinking in the study of the landscape existed already before ecology
was established as a discipline (Claval 2005). The concept of landscape has been
introduced in ecology rather late (Décamps and Décamps 2004). Initially, the
landscape was seen in ecology as one of the scale levels in the increasing com-
plexity of the organisation of ecosystems.
Basically, the concept of landscape necessitates an observer. Different observers
see and conceive different landscapes. This applies to humans, but also to all living
species. The landscape of a cow in a pasture is fundamentally different from that of
a bird. Landscape ecology took this organism-centred perspective as a basis for its
definition of landscape (Wiens 1976). Consequently, the landscape consists of a
heterogeneous mosaic of habitats that are functionally important to a given species.
The landscape is a spatial configuration of beneficial and hostile patches for the
organism. For an organism the landscape extends as its home range, and thus covers
different areas according to the mobility of the species concerned. This defines the
scale the landscape is studied at and consists of two parameters: the extent
4.3 Disciplines 73

(corresponding to the home range) and the grain or the degree of detail (resolution)
needed to describe all significant features.
Although these concepts are principally valid, when Homo sapiens sapiens is the
species studied, they become more complex. The extent of the human home range is
now global, his technology allows him to transform and more or less control his
environment. The historical development of the cultural landscape shows this
increasing complexity from a simple landscape ecological model to a global
Total Human Ecosystem (THE) (Naveh 2007). Anyhow, landscape ecological
principles are needed in the human perspective of the landscape. Michael Moss
(1999) formulated it as follows:
To me, landscape ecology is simply about the study of landscapes and of the need to derive
understanding about landscapes in order to enhance our abilities to manage them more
effectively. Landscape ecology is not the only field to focus on the landscape but it has
emerged in the last few decades because, quite clearly, existing approaches that sought to
address a whole range of landscape scale environmental issues were proving to be

Landscape ecology conceives the space of a landscape as a mosaic composed of

landscape elements, which configure a spatial structure, characterised by a series of
meta-properties such as diversity, heterogeneity and order. The patch-corridor-
matrix model is used essentially (Forman and Godron 1986). This conceptual
model initiated new theoretical and methodological approaches in the study of
landscapes. The spatial aspect became integrated in the classical systems theory
used in ecology. The central paradigm is that the spatial structure of the landscape
interacts continuously with ecological processes that shape it. Techniques of spatial
analysis and geostatistics from geography were used and developed, creating a
proper quantitative approach to the study of structural characteristics of the land-
scape by means of landscape metrics and landscape indicators.
Landscape ecology contributed to new insight and applications in nature con-
servation and landscape restoration and had an important impact on spatial and
environmental planning. Typical examples are the introduction of ecological net-
works and green infrastructure (Ahern 1995; Baudry and Merriam 1988). Paul
Opdam (2005) demonstrated how the island theory from ecology and the theory
on meta-populations became introduced in interdisciplinary spatial planning of
ecological networks. Landscape ecological concepts proved to be useful also in
the study of palaeo-landscapes and the development of cultural landscapes since the
Neolithic. Early human settlements can be seen as patches in a vast matrix of
wilderness. Contemporary landscapes often form an urban matrix with some
fragmented patches of agricultural land, woodland and nature, as well as different
types of corridors. The patch-matrix model even applies on landscape preference:
the smaller patches are regarded as valuable entities and receive more attention that
the vast matrix of ‘ordinary’ landscape around. Protection of the remaining open
rural space against ‘development’ uses arguments against fragmentation. Corridors
define our mobility and different superimposed networks of transportation corridors
74 4 Approaches in Landscape Research

4.3.3 Historical Ecology

Historical ecology studies past ecological conditions, processes and practices to

understand the occurrence and distribution of species, as well as human actions in
relation to the environment. Results indicate that many of now lost practices were
highly sustainable.
The distribution of species today also indicates past situations. The study of
seeds and pollen found in filled ditches and ponds and peat deposits enables the
dating and reconstruction of the vegetation types in past landscapes, as well as crops
used by humans. Understanding long-term processes such as grazing helps explain
the succession of landscape types (Vera 2000). Oliver Rackham (1990, 2004)
demonstrated the importance of studying ancient trees in the reconstruction of
past landscapes.

4.3.4 Archaeology

Since approximately two decades archaeologists developed a specialisation called

landscape archaeology or geo-archaeology (Aston and Rowley 1974; Turner 2013).
Branton (2009) speaks of historical archaeology as the archaeology of places. It is
the result from the input of approaches, theories and concepts of several disciplines
from natural sciences in archaeological research. A wide variety of methods and
techniques is used from soil science, geology, geomorphology, geophysical
prospecting, dating techniques, pollen analysis, aerial photography and remote
sensing, spatial analysis and GIS. This demands an interdisciplinary approach at
the scale of the landscape. Landscape archaeology focuses upon the reconstruction
of palaeo-landscapes and the relations that ancient civilisations developed regard-
ing the use of the natural resources in their environment. Landscape archaeology is
very similar to settlement archaeology and ecological archaeology, but focuses on
landscape modelling in a dynamic perspective.
Landscape archaeologists introduced concepts as time depth of the landscape,
landscape paths or trajectories. They also developed methods for Historic Land-
scape Characterisation (HLC) (Clark et al. 2004), which aim to integrate landscape
archaeology, historical geography and historical ecology for applications in heri-
tage protection and spatial planning. Also, they focus on the management of change
in the perspective of archaeological conservation (Fairclough and Rippon 2002).

4.3.5 Environmental Psychology

Since the second part of the twentieth century, psychologists showed a growing
interest in the relations between the environmental conditions and the development
4.3 Disciplines 75

of the human personality and behaviour. In the beginnings, the focus was on the
ways people perceived their surroundings, and how meanings and values were
formed. The environment proved to be important in understanding processes of
learning and behaviour. For example, following biophysical environmental factors
were identified to influence health conditions and stress: weather, noise, upheaval
and pollution. Also factors defining the social environmental proved to be impor-
tant: population density, accessibility, mobility, territoriality, defining public, pri-
vate and personal space, and finally aesthetics. All these are represented or reflected
in the landscape.
Theories and methods from social sciences were used to study the relations
between landscape properties, environmental factors and psychological and social
indicators. Methods consist of surveys, interviews and experiments. Common is the
use of photographs or video of a landscape and measuring the physiological
response. Applications of these studies are mainly found in physical (urban)
planning and design.

4.3.6 Landscape Architecture

Landscape architecture emerged from the garden architecture of palaces in close

relationship with arts, architecture and urban design. Landscape architecture devel-
oped outside academic sciences as a profession where the creativity of the designer
and the originality of the design prevailed (Bell 1999). The American Society of
Landscape Architects (ASLA), founded in 1899, is the oldest association to deal
with the landscape in a professional manner.
The importance of landscape architecture grew with the political impact of large
development projects and land reforms. As an expressive form of art in garden
design, landscape architecture became also important as an instrument in ideolog-
ical and political propaganda (Olwig 2002). The final shaping of new constructions
and infrastructure and their integration in the landscape has become a main task for
architects and landscape architects today.

4.3.7 Economics

Price (2013a) reminds that implicit applications of economics to landscape date

already from when land became a commodity as a natural resource. Also, when
landscapes were transformed and created for aesthetic purposes, such as in land-
scape gardening, economical considerations about costs were important. The for-
mal discipline of landscape economics derived from the growing demand of
landscape as a common aesthetic good by the broad public (Price 2013a). The
book Landscape Economics by Colin Price (1978) can been seen as the start of this
new research field.
76 4 Approaches in Landscape Research

The economic evaluation of landscape builds upon the practice of monetary

valuation of the environment and concepts as ecosystem functions and services and
natural capital (Costanza et al. 1997). The ecological complexity is translated into a
series of ecological functions (regulation, habitat production and information),
which provide goods and services that are valued by humans (de Groot et al.
2002). Antrop et al. (2013) proposed a framework to link landscape qualities,
functions, values to specific (multifunctional) land uses. Methods developed
for the valuation of natural and environmental goods and services were applied
on landscapes as well. However, besides instrumental values of material
and ecological components of the landscape, landscapes are characterized by
intangible values, such as aesthetic and cultural values. Aesthetic value in
particular is regarded a non-instrumental value (van der Heide and Heijman
2013). The difference between instrumental and non-instrumental values is
the basis of the discussion between objectivity and subjectivity in landscape
evaluation (Price 2013b).

4.4 Inter- and Transdisciplinary Approaches

Landscape encompasses most of the societal sectors, such as agriculture, forestry,

nature and heritage conservation, urban and spatial planning, recreation and
tourism. Special landscapes can be protected as monument, and land use is
controlled by environmental and planning legislation. Most landscape related
issues cover several sectors and demand research input from several disciplines.
No sector-oriented or singular disciplinary approach proved to be adequate and
efficient to deal with landscape issues. Stimulated by the Aarhus Convention,
adopted by the European Commission in 1998, the of democratic participation in
policy became important, and necessitated more input from the ‘public’, the
(local) population and (potential) users in all landscape matters. Therefore, a
transdisciplinary approach in landscape research is necessary (Naveh 2007) and
essential for the further development of disciplines (Wu and Hobbs 2007). Tress
et al. (2005a, b) analysed the integration of scientific disciplines and
non-academicians in landscape studies and proposed clear definitions for the
different approaches. In a (single) disciplinary approach the landscape is studied
from the specific problems and goals of one discipline. Other aspects of the
landscape are disregarded. In a multidisciplinary approach, several disciplines
study simultaneously the same landscape and have a common research theme, but
still keep their specific approaches. Different aspects of the landscape are studied,
but not yet integrated. The final report looks like a collection of chapters each
devoted to one discipline. In interdisciplinary research, a central common prob-
lem and research goal is studied simultaneously and interactively by several
disciplines. Each discipline only offers a contribution that is significant for the
common goal. The chapters in the final report will refer to partial aspects of the
problem and the steps to solve it. Interdisciplinary research implies that a
4.4 Inter- and Transdisciplinary Approaches 77

common language is developed and understood by all participating disciplines

and from the start communication and co-operation is essential to integrate all
new knowledge. However, interdisciplinary research only involves scientists or
experts. In transdisicplinary research also policy makers, administrators and
laypeople participate. Tress et al. (2005a) call inter- and transdisciplinary
research also integrative research (Fig. 4.8). The differences between the different
degrees of disciplinary integration should become clear in the work organization
and the report structure (Fig. 4.9).

Fig. 4.8 Level of

integration and participation
in landscape related
research (After Tress et al.

Fig. 4.9 Work organization and report structure in different degrees of integrated landscape
78 4 Approaches in Landscape Research


Agache, R. (1978). La Somme Pré-Romaine et Romaine (515 pp). Société des Antiquaires de
Picardie: Amiens.
Ahern, J. (1995). Greenways as a planning strategy. Landscape and Urban Planning, 33, 131–155.
Ampe, C., Bourgeois, J., Crombé, Ph., Fockedey, L., Langohr, R., Meganck, M., Semey, J., Van
Strydonck, M., & Verlaeckt, K., (1996). The circular view. Aerial photography and the
discovery of Bronze Age funary monuments in East-and West-Flanders (Belgium). Germania
74, 1. Halband, 45–94.
Antrop, M. (2004). Assessing multi-scale values and multifunctionality in landscapes. In J. Brandt
& H. Vejre (Eds.), Multifunctional landscapes, Theory, values and history (Vol. I,
pp. 165–180). Southampton: WIT Press.
Antrop, M., Sevenant, M., Tagliafierro, C., Van Eetvelde, V., & Witlox, F. (2013). Setting a
framework for valuing the multifunctional landscape and its multiple perceptions. In M. van
der Heide & W. J. M. Heijman (Eds.), The economic value of landscapes (pp. 23–52). London:
Routledge Studies in Ecological Economics.
Aston, M., & Rowley, T. (1974). Landscape archaeology: An introduction to fieldwork techniques
on post-roman landscapes. Vancouver: Newton Abbot.
Baudry, J., & Merriam, G. (1988). Connectivity and connectedness – functional versus structural
patterns in landscapes. In: K-F. Schreiber (Ed.), Connectivity in landscape ecology (pp. 93–96)
Proc 2nd IALE-seminoar, Münstersche Geogr Arb 29.
Bell, S. (1999). Landscape. Pattern, perception and process (344p). London: Taylor & Francis
Branton, N. (2009). Landscape approaches in historical archaeology: The archaeology of places.
In T. Majewski & D. Gaimster (Eds.), International handbook of historic archaeology.
New York: Springer.
Clark, J., Darlington, J., & Fairclough, G. (2004). Using historic landscape characterisation.
English Heritage & Lancashire County Council, 63 pp.
Claval, P. (2005). Reading the rural landscapes. Landscape and Urban Planning, 70(1–2), 9–19.
Cosgrove, D. (1984). Social formation and symbolic landscape. London: Croom Helm.
Cosgrove, D. (1993). Landscapes and myth, gods and humans. In B. Bender (Ed.), Landscape
politics and perspectives, Explorations in anthropology series (351 pp). Oxford: Berg
Cosgrove, D. (2002). Chapter 12: Landscape and the European sense of sight – Eyeing nature. In
K. Anderson, M. Domosh, S. Pile, & N. Thrift (Eds.), Handbook of cultural geography
(pp. 249–268). London: SAGE Publications.
Costanza, R., d’Arge, R., de Groot, R. S., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem,
S., O’Neill, R. V., Paruelo, J., Raskin, R. G., Sutton, P., & van den Belt, M. (1997). The value
of the world’s ecosystem services and natural capital. Nature, 387, 253–260.
Crawford, O. G. S. (1960). Archaeology in the field (280 pp). London: Phoenix House.
Dassié, J. (1978). Manuel d’archéologie aérienne. Paris: Technip.
de Groot, R. S., Wilson, M., & Boumans, R. (2002). A typology for the description, classification
and valuation of ecosystrem functions, goods and services. Ecological Economics, 41,
De Reu, J., Bats, M., Bourgeois, J., & Antrop, M. et al. (2010). Digitizing, inventorying, reviewing
and analyzing the “Bronze Age barrows database” of East and West Flanders (Belgium).
Lunula, Archaeologia protohistorica.
Décamps, H., & Décamps, O. (2004). Au printemps des paysages, E´cologie (232 pp). Paris: Buchet
et Chastel.
Fairclough, G., & Rippon, S. (Eds.) (2002). Europe’s cultural landscape: Archaeologists and the
management of change. Europae Archaeologiae Consilium, Occasional Paper 2, Brussels.
Forman, R., & Godron, M. (1986). Landscape ecology (619 pp). New York: Wiley.
References 79

Jensen, J. R. (2000). Remote sensing of the environment. An Earth resource perspective. Geo-
graphic information systems. Prentice-Hall: Saddle River, 544 pp.
Kolen, J. (2005). De biografie van het landschap. Drie essays over landschap, geschiedenis en
erfgoed. Amsterdam: Vrije Universiteit.
Li, H., & Wu, J. (2007). Landscape pattern analysis: Key issues and challenges. In J. Wu & R. J.
Hobbs (Eds.), Key topics in landscape ecology, Studies in landscape ecology (pp. 39–61).
Cambridge: Cambridge University Press.
McGarigal, K., & Marks, B. J. (1995). FRAGSTATS: Spatial pattern analysis program for
quantifying landscape structure. USDA For. Serv. Gen. Tech. Rep. PNW-351.
Moss, M. R. (1999). Fostering Academic and Institutional Activities in Landscape Ecology. In:
J. A. Wiens, & M. R. Moss (Eds.), Issues in landscape ecology. International Association for
Landscape Ecology, (pp. 138–144). Fifth World Congress, Snowmass Village, Colorado, USA.
Naveh, Z. (2000). What is holistic landscape ecology? A conceptual introduction. Landscape and
Urban Planning, 50, 7–26.
Naveh, Z. (2007). Transdisciplinary challenges in landscape ecology and restoration ecology – An
anthology, Landscape series, Vol.6, XVII (423 pp). Dordrecht: Springer.
Naveh, Z., & Lieberman, A. (1994). Landscape ecology: Theory and application (360 pp).
New York: Springer Verlag.
Olwig, K. R. (1996). Recovering the substantive nature of landscape. Annals of the Association of
American Geographers, 86(4), 630–653.
Olwig, K. R. (2002). Landscape, nature and the body politic: From Britain’s renaissance to
America’s new world (299 pp). Madison: University of Wisconsin Press.
Opdam, P. (2005). Ecosystem networks: A spatial concept for integrative research and planning of
landscapes. In: B. Tress, G. Tress, G. Fry, & P. Opdam (Eds.), From landscape research to
landscape planning: Aspects of integration, education and application. Wageningen UR
Frontis Series, 12 (pp. 51–65).
Price, C. (1978). Landscape economics. London: Macmillan.
Price, C. (2013a). Reaching the economics of landscape. In P. Howard, I. Thompson, &
E. Waterton (Eds.), The Routledge compagnion to landscape studies (pp. 308–321). London:
Price, C. (2013b). Subjectivity and objectivity in landscape evaluation: An old topic revisited. In
M. van der Heide & W. J. M. Heijman (Eds.), The economic value of landscapes, Routledge
studies in ecological economics (pp. 53–76). London: Routledge.
Rackham, O. (1990). Trees and woodland in the British landcsape. London: Dent.
Rackham, O. (2004). Pre-existing trees and woods in country-house parks. Landscape, 5(2),
Rippon, S. (2004). Historic landscape analysis. Deciphering the countryside, Practical handbooks
in archaeology n 16 (166 pp). Walmgate: British Council for Archaeology.
Rippon, S. (2012). Historic landscape analysis: Deciphering the countryside, Practical handbooks
in archaeology, No.16 (160 pp). York, CBA Publications.
Scollar, I. (1975). Transformation of extreme oblique aerial photographs to maps or plans by
conventional means or by computer. Aerial Reconnaissance for Archaeology.
Sevenant, M. (2010). Variation in landscape perception and preference. Experiences from case
studies in rural and urban landscapes observed by different groups of respondents. Doctoral
Dissertation. Ghent University, Department of Geography, Ghent. p. 480.
Stichelbaut, B., & Chielens, P. (2014). The great war seen from the air. In Flanders fields
(pp. 1914, 350 pp–1918). Yale University Press: New Haven.
Tress, B., Tress, G., & Fry, G. (2005a). Defining concepts and the process of knowledge
production in integrative research. In B. Tress, G. Tress, G. Fry, & P. Opdam (Eds.), From
landscape research to landscape planning: Aspects of integration, education and application,
Wageningen UR Frontis Series, n 12 (pp. 13–26). Dordrecht: Springer.
Tress, B., Tress, G., & Fry, G. (2005b). Ten steps to success in integrative research projects. In
B. Tress, G. Tress, G. Fry, & P. Opdam (Eds.), From landscape research to landscape
80 4 Approaches in Landscape Research

planning: Aspects of integration, education and application, Wageningen UR Frontis Series, n

12 (pp. 241–257). Dordrecht: Springer.
Troll, C. (1939). Luftbildforschung und Landeskundige Forschung. Erdkundliches Wissen.
Schriftenreihe für Forschung und Praxis, Heft 12, F. Wiesbaden: Steiner Verlag.
Turner, S. (2013). Landscape archeology. In P. Howard, I. Thompson, & E. Waterton (Eds.), The
Routledge compagnion to landscape studies (pp. 131–142). London: Routledge.
van der Heide, M., & Heijman, W. J. M. (Eds.). (2013). The economic value of landscapes,
Routledge studies in ecological economics (329 pp). London: Routledge.
Van Slembrouck, N., Lehouck, A., & Thoen, E. (2005). Past landscapes and present-day tech-
niques: Reconstructing submerged medieval landscapes in the western part of Sealand Flan-
ders. Landscape History, 27, 51–64.
Van Wirdum, G. (1981). Design for a land ecological survey of nature protection. In Tsjallingii, &
de Veer (Eds.), Perspectives in landscape ecology. Proceedings of the Intern (pp. 245–251).
Congress of the Netherlands Society for Landscape Ecology, Centre for Agricultural Publish-
ing and Documentation, Wageningen.
Vera, F. W. M. (2000). Grazing ecology and forest history (528 pp). Wallingford: CABI
Wiens, J. A. (1976). Population response to patchy environments. Annual Review of Ecological
Systems, 7, 81–129.
Wu, J., & Hobbs, R. J. (2007). Key topics in landscape ecology. Cambridge: Cambridge University
Zonneveld, I. S. (2005). The land unit as a black box: A Pandora’s box? In J. A. Wiens & M. R.
Moss (Eds.), 1999. Issues in landscape ecology, Studies in landscape ecology (pp. 331–345).
Cambridge: Cambridge University Press.
Chapter 5
Basic Concepts of a Complex Spatial System

Abstract The landscape is holistic, which is also referred to as ‘the whole is more
than the sum of its composing parts’, and is related to the German concept of the
Gestalt. Human perception also works by holistic Gestalt-principles and will be
discussed in a separate chapter. The holistic principle means that the structural
context of the composing elements defines their actual meaning in the whole and
the relationships between the elements. System-theoretical models for landscapes
introduced the concepts of holons and ecodevices as hierarchically structured
building blocks of the landscape. Consequentially, context and scale are important
factors in studying the landscape. Scale has different meanings according to the
context it is used in and thus confusion is possible. Scale defines the hierarchical
structure and the way features of the landscape can be represented on maps. To
simplify data collection and to reduce the complexity of the landscape composition
and configuration, landscapes are often decomposed in thematic layers. The com-
bination of the thematic properties defines the landscape type that characterizes an
area. The transition between landscape types and regions can result in crisp or fuzzy
borders. Sometimes the transition zones form gradients and ecotones, which if large
enough become landscape units themselves.

Keywords Holism • Gestalt • Scale • Heterogeneity • Palimpsest • Pattern and

process • Connectivity

5.1 Introduction

The landscape as a complex spatial system is characterised by some specific

concepts. Holism is the most basic of them and gives the fundament to the
hierarchical system of landscape and its composition and configuration by holons.
Landscape is a spatial system that is scale dependent. Also scale is a complex
concept with multiple meanings and defines landscape heterogeneity and diversity,
as well as the meaning of the borders between landscape components. The distinc-
tion between components that consist of discrete objects and the ones that are
continuous phenomena is fundamental. To reduce the complexity, landscape can be
conceived as consisting of different thematic layers that each should be studied by
proper methods.

© Springer Science+Business Media B.V. 2017 81

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_5
82 5 Basic Concepts of a Complex Spatial System

Holism forms also the basis of landscape character, identity and diversity
through its relation to Gestalt-principles, which are fundamental to our perception
of the landscape, and which will be discussed more in details in the next chapter.

5.2 Holism

The expression “Landschaft ist der Totalcharakter einer Erdgegend” (‘landscape is

the total character of a region’) was attributed (but not proven) to Alexander von
Humboldt (Hard 1970; Zonneveld 1995; Küster 2008). The Totalcharakter fits well
in the Gestalt concept and holism. Although Alexander von Humboldt did not use
the concepts holism and Gestalt and refers to landscape mostly as scenery, he was
the first to demonstrate that nature forms a whole interacting system and that ‘the
whole is more than its composing parts’ (von Humboldt 1807; Wulf 2015).
Holism is a philosophical principle that was introduced by the South African
statesman Jan Smuts in his book Holism and Evolution (Smuts 1926), defining
holism as the “tendency in nature to form wholes that are greater than the sum of the
parts through creative evolution.” He refers to the hierarchical organisation and
structure of the cosmos, as it also shown in the evolution theory (Zonneveld 1995).
The holistic principle is commonly described as ‘the whole is more than the sum of
its composing parts’. The German concept is Gestalt (Fig. 5.1). Gestalt-psychology
studies also the human perception and learning processes, which both are signifi-
cant in understanding how humans perceive the landscape (see Chap. 6). Generally,
we first discern wholes, as forms or patterns without details, to which we assign
immediately a meaning or eventually identify. It is what happens when we see a
person from a distance and his figure, posture, way of walking etc. allows us to
identify him. Only during successive observations, details become apparent. When
no immediate identification is possible, we may feel uneasy, uncertain. Then the
brain will attempt to find alternative meanings to reduce that uncertainty and this
process continues until a satisfactory meaning or identification is obtained. This
may cause ambiguous interpretations, which also show our ability to switch alter-
natively and mentally foreground and background to find other solutions.
Perceiving the landscape as a Gestalt means that our ‘natural’ experience of
landscape is holistic. This is obvious in artistic representations, in studies on
landscape aesthetics, perception and preference, as well as when considering
landscape as a social construct. In landscape architecture, planning and heritage
conservation, often the term ensemble is used to denote holistic entities. These
aspects will be elaborated more deeply in following chapters.
However, holism in the sense that ‘the whole is more than the sum of the
composing parts’ poses a paradox. In principle, it is not possible to reconstruct
the ‘whole’ from a detailed analysis of all its parts and the ‘more than’ will always
be missing. Also, it is unlikely that everything can be studied to approach the
‘whole’ as complete as possible. This was the main criticism from the exact
sciences and the argument to reject the holistic idea and to promote instead a
5.2 Holism 83

Fig. 5.1 Example of a

Gestalt. Our mind always
attempts to give some
meaning to the observed
pattern, which may lead to
the interpretation of
different realities

reductionist and experimental approach. As a consequence, the approach to land-

scape analysis in natural sciences became reductionist, while in social sciences it
remained more perception-oriented and phenomenological. Anyhow, both
approaches studied different aspects of the landscape, and none the ‘whole’.
Theoretically, holism remains an important principle, which regained signifi-
cance with the development of systems theory in ecology. System-theoretical
models for landscapes were formulated, such as the one by Jan Zonneveld
(1985), who sees the landscape as an organised open ecosystem of interdependent
components (Fig. 5.2).
An important concept is a holon. Zev Naveh and Arthur Liebermann (1994)
developed this concept in the landscape context, fitting it in a multi-scale hierar-
chical structure of the Total Human Ecosystem (THE) (Fig. 5.3). Holons are seen as
subsystems having a certain degree of freedom in functioning, thus are more or less
autonomous. Each holon can consist of holons of a lower level and can be
embedded in holons higher in the hierarchy.
Holism and holons can easily be understood when referring to the human body.
The body is also composed of interacting subsystems working more or less auton-
omously, such as the digestive system, the cardiovascular system, the movement
system, each of the senses, etc. Each of them can be studied using special methods
but is improbable that merging all this knowledge will result in a complete
understanding of a person. Similarly, each individual fits as a holon in larger social
systems, as family, community, culture and nation.
The concept of a hierarchically organised system of holons helps to overcome
the holistic paradox. To understand a holon in a comprehensive way, it is not
necessary trying to analyse and understand everything, but only the significant
context of the holon at the appropriate scale. Thus, scale and context are essential
variables to set the conditions for the landscape analysis (Fig. 5.4). In the example
of the human body, one could say that the ophthalmologist will look at the eye and
84 5 Basic Concepts of a Complex Spatial System

Fig. 5.2 The landscape as an open system. The circles represent the components of the landscape.
Primary components are the substrate (S), which consists of the geology and landform, the climate
(C), and humans (M). These components get external input by the energy of the sun (E1), the heat
of the Earth (E2) and knowledge and information (E3). The primary components determine the
functioning of the secondary components: plants (P), animals (A), soil (S), air (O) and water (W).
All components are mutually interdependent. They can be grouped in three spheres with increasing
complexity: the abiotic sphere, the biosphere and the noetic sphere (Model of J. Zonneveld 1985)

Fig. 5.3 The Total Human

Ecosystem (THE):
application of the holon
concept in the hierarchy in
ecology (After Naveh and
Liebermann 1994)

the whole context that relates to its functioning, but it is unlikely that he needs
information about the digestive system to make an adequate diagnosis.
Understanding how landscapes function and how they developed, demands a
multiscale approach by which the overall complexity is reduced at the appropriate
level. It is what Richard Dawkins (1996) calls “hierarchical reductionism”. Forman
and Godron (1986) refer to the approach as “shuttle analysis”, zooming in from
space to the smallest element in the landscape, thus revealing stepwise the details
necessary to understand.
5.2 Holism 85

Fig. 5.4 Hierarchy of holons: holons are open systems that have a certain freedom and autonomy;
they can be part of larger holons and can consist of smaller holons. The whole forms a hierarchical
structure. Each holon has its proper scale and context, which become essential variables to study
the holon in a comprehensive way. Scales and letters refer to the entities in the example of the
Grande Brière (Fig. 5.5)

Ies Zonneveld (1995) considered the land units, defined in land evaluation and
landscape classification, as holons and regarded these as complex, hierarchical
wholes. At the initial stage of a study, these can be considered as ‘black boxes’,
meaning they still are ‘opaque’ to us as far as internal processes are concerned.
Nevertheless, they allow us to describe and order rapidly the complex variation in
the landscape. Our ‘natural’ ability to recognise, name and classified such Gestalt-
entities “follows millennia-old wisdom, derived from common practices of
pre-technological land users like hunters, farmers, and herdsmen who invented
this principle at the dawn of humanity’s struggle for life in the landscape. A
major testimony to this is the wealth of information represented by the ecologically
inspired land toponyms.” (Zonneveld 1995).
The following example of a multiscale landscape analysis of the Grande Brière
marshland (Western France) demonstrates the principle of hierarchical reduction-
ism (Fig. 5.5). The analysis uses three scales (Fig. 5.4) and zooms in from the small
reconnaissance scale showing the whole region, to the scale of individual objects
and elements. Scale 1 shows two main holons: the swamp and the islands with
settlements. Scale 2 zooms in on one of the settlements revealing its internal
structure and composition. Scale 3 looks at the elements that characterise the
landscape. The typical elements and selected village are used to construct the
settlement model for the area. An adequate understanding of the whole landscape
system can be achieved without having to study all settlements in detail.
The second example analysis one of the landscape types on the island Lanzarote
and illustrates the holon concept as an ecodevice, which was used to create a unique
cultural landscape (Antrop 2006). It also demonstrates that not only ecological
knowledge is necessary to understand the creation of this landscape, but also the
societal, political and economic context is essential.
86 5 Basic Concepts of a Complex Spatial System

Fig. 5.5 The hierarchical structure of landscape holons is particularly apparent in aerial photo-
graphs as in this example of the Grande Brière (France). (1) Scale 1: in the vast swamp covered
with reeds, sandy outcrops form islands suitable for settling. The Île de Fédrun is a holon that can
be used as a model for all settlements in the region. (2) Scale 2: the village Fédrun forms an outer
ring on the edge of the island. Fields occur at the highest central part (x). The outer fringe ( y)
consists of gardens and orchards stretching to surrounding water channels, forming a natural
windscreen. Scale 3: each house has its own pier and the water channels are the main transport
ways in the marshland (a). Traditionally the villagers subsisted from eel fishing and reed cutting
(b, c). The low, traditional loam houses have reed roofs and the typical architecture (d) contributes
to the local identity of the place. ((1) Google Earth, photos (2–5) M. Antrop 2006))
5.2 Holism 87

Lanzarote is an active volcanic island of the Canaries (Spain) and recent lava and
ash fields cover most of the land. The climate is subtropical with almost no rainfall
and a continuous, dry north-eastern trade wind. Consequently, the island has almost
no tree cover and agriculture is difficult. La Geria is a wine-producing valley
between volcanic cones and is covered by black volcanic ashes (tefra) and lapilli.
Somehow, people did not leave the unfriendly environment and transformed this
pristine natural landscape in the eighteenth century into the main production area of
the sweet malvasia wine. Several, enchaining events determined this history.
From 1730 to 1736 a series of violent volcanic eruptions produced 32 new
volcanoes and lava covered a quarter of the island’s surface, including the most
fertile soils and eleven villages (Borisch 2007). People began to give up hope and
many migrated. Fearing that an abandoned island could become lost for the crown,
king Philipp V issued a decree forbidding leaving the island on penalty of death.
The ‘stay or die’ rule forced the farmers to find solutions to regain arable land. The
initial solution was to dig pits in the volcanic ashes to recover the fertile soil. Soon
they discovered that the mixture of volcanic ashes with the original sandy soil
improved its fertility and that plants in the pits grew better as they were protected
from the wind. This technique of making artificial soil is called enarenado artificial
and soon became popular over the whole island.
The choice for the sweet wine can also be explained by the historical context.
Sweet wines, such as the Portuguese Port and Madeira, were very appreciated in
Britain that was then a political world power. During the Napoleonic Wars
(1806–1814) and the Continental Blockade, the Canaries were a free trade zone.
Therefore, the farmers of Lanzarote found it profitable to fill in the market and
started to produce malvasia wine transforming the volcanic landscape of La Geria
into a vineyard. Thus, they dug ten thousands of small pits in the ashes to reach the
soil beneath and planted in each one vine. To protect the plant from the drying wind,
they built open stone walls along the north-eastern edge of the pit. The open
construction acted as a windbreak, reducing the wind speed and avoiding turbu-
lence in the pit. The whole system is oriented to capture the solar energy most
effectively. During the day, the black volcanic soil is heated intensively, but cools
rapidly after sunset. The volcanic lapilli are highly hygroscopic and absorb mois-
ture from the air during the night, which is collected by the plant at the bottom by
gravity. Each pit is an artificial ecodevice and repeating it thousands of times over
the whole area created a unique, sustainable cultural landscape with a pronounced
This technique of enarenado artificial, making artificial soil, also known as
‘lithic mulching agriculture’, was so successful that it was applied all over the
island in various forms and with a multifunctional use of the fields having different
crops. The unique combination of nature and sustainable agriculture was one of the
factors to designate Lanzarote as a World Biosphere Reserve (Fig. 5.6).
Van Wirdum (1981) applied the holon concept on the water regulation in the
Dutch polder system but called it an ecodevice (Fig. 5.7). It is a functional
interpretation of holons and joins the more recent concept of ecosystem services
(see also Chap. 4).
88 5 Basic Concepts of a Complex Spatial System

Fig. 5.6 The wine region La Geria on the island of Lanzarote. (1) La Geria forms a valley between
volcanic cones and is covered by black volcanic ashes (tefra) and lapilli. (2) To recover arable land
after the volcanic eruption, farmers dug ten thousands of small pits (a) in the ashes (d ) to reach the
soil beneath and planted in each one vine (b). (3) To protect the plant from the drying trade winds,
they built open stone walls along the north-eastern edge of the pit (c and 4). These windbreaks
reduce the wind speed and avoid turbulence in the pit. The whole system is oriented so the solar
energy is captured most effectively. During the day, the black volcanic soil is heated intensively
but cools rapidly after sunset. The volcanic lapilli is highly hygroscopic and absorb moisture from
the air during the night, which is collected by the plant at the bottom by gravity. This technique of
enarenado artificial is nowadays applied all over the island in various forms and with a
multifunctional use of the fields having different crops (4) (Photographs M. Antrop 2009)

Fig. 5.7 The ecodevice is a holon model of an ecosystem controlled by a series of functions. It is
an open subsystem and has an input (source) and output (sink). The ecodevice is controlled by for
functions: supply, resistance, disposal and retention. The balance between these functions keeps
the ecodevice between critical thresholds (minimum, maximum; carrying capacity) that define its
survival (After van Wirdum 1981)
5.3 Scale and Heterogeneity 89

5.3 Scale and Heterogeneity

5.3.1 A Source of Conceptual Confusion

The word scale acquired different meanings according to the discipline and context
it is used in. In geography and cartography, the scale is the ratio between the
represented length on a map or photo and the real length on the terrain. Thus, a
small scale map (i.e. 1: 100,000) has a large extent covering a large area, but shows
little detail. A large scale map (i.e. 1: 5000) has a small extent and covers a small
area, but shows many details at their accurate position and with high precision.
Other disciplines, in particular in planning and policy use scale in the opposite
meaning. A large-scale project means an extensive one, covering possibly a large
area. To represent it as a whole, a small-scale map will be used. Here scale means
the geographical scope. Similarly, small-scale projects refer to projects with a small
extent. Confusion can be avoided by speaking of a fine or detailed scale and a coarse
Clearly the scale concept consists of two components: (1) the degree of detail of
the representation, and (2) the extent of the representation. This concerns the spatial
scale as well as the temporal scale and scale of organisation and management
(Agarwal et al. 2002) (Table 5.1).

5.3.2 The Spatial Scale

The cartographic meaning of scale dates from the printed maps and is basically
static and categorical. With digital mapping and new visualisation techniques, scale
became dynamic and continuous. Zooming in and out became a standard procedure
in exploring landscapes as represented on maps and imagery. The limiting factor is
the resolution. In landscape research in general, and in landscape ecology in
particular, scale properties became important explanatory variables in the analysis
of landscape patterns and processes. In landscape ecology, the concept of scale is
also species dependent (Wiens and Milne 1989).
The spatial scale is defined by the spatial resolution or grain, and the extent or
area covered for the analysis. The grain does not necessary correspond to the
resolution of the documents used; it is the smallest unit of observation which is
chosen for the intended analysis. An aerial photograph may have a spatial

Table 5.1 Dimensions of the scale concept

Application domain Degree of detail Size-extent
Spatial (geometric) Resolution, grain Extent
Temporal Time interval Duration
Organisation and planning Actor, ‘agent’ Domain
90 5 Basic Concepts of a Complex Spatial System

resolution of less than 1 m, the smallest observation unit can be 1 ha or even a field.
However, the grain cannot be finer and more detailed than the spatial resolution of
the base document. For analogue maps and vector maps a grain corresponds to the
Minimum Mapping Unit (MMU). For example, for CORINE Land Cover, the
MMU is 25 ha.
The scale becomes a variable that determines the observation of the landscape
and this has important consequences. Many properties of the landscape are scale-
dependent, such as diversity, heterogeneity and the correlation between the

5.3.3 The Temporal/Time Scale

Studying temporal changes demands observations on successive times that can be

compared. The interval between observations is equivalent to the grain. Two
editions of maps or aerial photographs representing the situation at a 10 year
interval for example, show the global and cumulated changes between the year Y
and Y + 10 but not the duration of the individual changes, nor changes that can be
reversed. Typical examples are deforestation-reforestation within the time interval.
The duration is equivalent to the extent and encompasses the time period of
investigation, for example from 1700 to 2012 (see also Chap. 7).

5.3.4 The Organisation and Planning Scale

The lowest level of decision-making lies with the individual. The complexity of the
decision-making increases with the number of actors involved. Different levels of
organisation in the process of decision-making can be recognised and actors are
sometimes referred to as agents. The domain of competence of an agent varies
spatially and temporally. In most countries, at least three levels can be recognised:
the national, the regional and local level or scale. Often additional levels exist as
inter-communal and inter-regional co-operations, some which are specific for the
management of landscapes that stretch over several administrative units. Federal
states often have a federal level. Above this lies the international level and also here
some hierarchy and differences in competences can be found (see also Chap. 12).

5.3.5 Landscape Heterogeneity Is Scale Dependent

When looking at the landscape from a distance, the field of view, degree of detail
and heterogeneity depend on the distance of the observation, thus upon the scale.
When the observation distance increases, the field of view and the extent of the
5.4 Discrete Objects and Continuous Phenomena 91

landscape viewed increases, while the degree of detail decrease. Heterogeneity

changes when zooming in and out. Forman and Godron (1986) studied this using a
technique they called the ‘shuttle analysis’, and made the distinction between
landscapes with micro and macro heterogeneity.
Landscape heterogeneity is fundamental in understanding the interaction
between landscape structure and ecological processes and human activities. It is
related to concepts such as fragmentation, complexity, diversity, coherence and
order. It influences biodiversity (Fahrig et al. 2011; Katayama et al. 2014) and
landscape ecosystem services (Turner et al. 2013). Landscape diversity is consid-
ered a distinctive feature of the identity or regions (Stanners and Bourdeau 1995).
Many methods were developed to quantify spatial heterogeneity (Li and Reynolds
1995; Garrigues et al. 2006; Mander et al. 2010) (see Chap. 8).

5.4 Discrete Objects and Continuous Phenomena

A landscape consists of discrete objects, such as buildings, trees, and of continuous

phenomena, such as land form and soils. Discrete objects are often referred to as
landscape elements, while the continuous features are called components. They
show a great variety and have many functions. Consequently, they can be coded,
modelled and mapped in several ways (see also Chap. 8).
Discrete spatial observations can result also from sampling continuous phenom-
ena, which can then be modelled into continuous geographical surfaces (Unwin
1981) or fields (Longley et al. 2001), representing one or several variables. Geo-
graphical surfaces are often visualized as isopleth maps. Digital elevation models
(DEM) or digital terrain models (DTM) have become common representations of
the topographical surface.
Spatial sampling is scale dependent. For instance, the distance between the
observations (lag) is important to understand the spatial autocorrelation between
the measurements. Spatial autocorrelation influences landscape heterogeneity
(Forman and Godron 1986; Burel and Baudry 2003) and the coherence and frag-
mentation of landscape patterns (Mander et al. 2010).
Discrete values can also be assigned to spatial units represented by polygons on a
map. This technique is used in the construction of choropleth maps. Often admin-
istrative units such as municipalities or census tracts are used, but also geometric
patterns such as quadrants and grids with a regular tessellation of squares can be
used. For example, the percentages of different land use categories can be assigned
to each square kilometre grid, to be used in defining the landscape type for each cell.
92 5 Basic Concepts of a Complex Spatial System

5.5 Landscape in Layers

Although we perceive the landscape in a holistic way, we also instantly discern

differences in its composition. Differences in shape, size and colour characterise
different kinds of features. Intuitively these are grouped in themes, which receive
different degrees of attention according to the intention and background of the
observer. This way of ordering mainly reduces the complexity.
Geographers adopted a method to decompose the landscape in superimposing
layers of different kinds of features. Each of these can be mapped and described
more easily using the most appropriate data and methods. This resulted in the
parametric approach in landscape evaluation and in the thematic map overlaying
in GIS-analysis (see also Chap. 10).
The features in the different landscape layers are often studied by different
disciplines. For example, landform is the domain of geologists and geomorpholo-
gists, land cover the one of botanists and ecologists, and human settlements belong
to the domain of historical geographers. The landscape-in-layers perspective stim-
ulates highly specialised, reductionist approaches and causes loss of synthesis and
holistic perspective (Fig. 5.8).

5.6 The Map Is Not the Landscape, Nor Is Its


The Polish-American scientist and philosopher Alfred Korzybski (1933) noted that
many people do confuse maps with territories, hence his expression “the map is not
the territory”. He used this to illustrate that an abstraction derived from something
tangible, or a reaction to it, is not the thing itself. The same holds for maps and
landscapes, and for all kind of aerial images of landscapes as well. People tend to
confuse models of reality with reality itself. Maps are tools for inventory, analysis
and communication, and the result of the mapmaker’s intention, vision and choices.
Mark Monmonier (2005) called it “selective truth”. Even when the mapmaker has
good intentions, the result can be wrong. Monmonier (2005) formulated it like
“watch out for the well-intended mapmaker who doesn’t understand cartographic
principles yet blindly trusts the equally naive software developer determined to give
the buyer an immediate success experience— default settings are some of the worst
offenders.” Maps have been used also to manipulate and deceive people (Harley
1988). Monmonier (2005) confessed he was inspired by Darrell Huff’s How to Lie
with Statistics (Huff 1954) when writing his How to Lie with Maps (1991, 1996).
Both books should be compulsory literature for landscape researchers reaching out
for techniques of mathematics, statistics and cartography. Olwig (2004) used the
famous painting of the Belgian surrealist René Magritte of a pipe with the subtitle
“Ceci n’est pas une pipe” (“This is not a pipe”) to express the same idea in relation
to all pictorial landscape representations.
5.6 The Map Is Not the Landscape, Nor Is Its Representation 93

Fig. 5.8 Landscape in layers: (0) the landscape synthesized in a block diagram, (1) substrate layer
(geology, soils), (2) landform layer (relief, topography), (3) land cover layer, (4) network layer
(roads, river, etc.), (5) settlement layer (Presented as such, no relations between layers are shown)

There are several reasons why reading the landscape from maps can be tricky.
First, there is geometry, scale in particular. The geometry of maps is the result of a
cartographic projection at a predefined scale, which defines the codified
94 5 Basic Concepts of a Complex Spatial System

representation of features, using cartographic rules and semiology (Bertin 1983).

The older historical maps and cadastre plans are based on a local reference system
and seldom fit in a general cartographic projection. Hence, their geometry is
distorted and their coverage fragmentary. With the advances in triangulation
since the second half of the sixteenth century, maps became an instrument to define,
administer and control territories and people, and as such an instrument in
constructing space (De Keyzer et al. 2014). Moreover, landowners commissioned
maps for attesting their possessions and for tax purposes, and only features were
mapped that fitted that purpose. Often, features not fitting in the profit of the
landowner were added or omitted, such as disputed land. It is only with the coming
of the national cartographic surveys in the nineteenth century that homogeneous
and geometrically accurate coverage for the whole of a country became available.
Nevertheless, geodetic reference systems still vary between regions and countries,
and also between successive map editions. Since the Second World War map
making is based on aerial photogrammetry and for each edition of a map exists a
simultaneous aerial photo cover, offering a holistic and not interpreted view on
the land.
People not trained in cartography are rarely aware that objects can only be
located accurately and represented in their true size and shape on maps with scale
not smaller than 1: 20,000. To keep the map readable, generalization according to
different rules is applied on smaller scales, causing displacement, selecting and
simplifying objects, smoothing shapes and the use of symbols. (see Chap. 8) Many
landscape features on maps are delineated as objects. In reality, some features have
crisp borders, and others have fuzzy gradients to neighboring features.
With the coming of GIS, maps became more dynamic, multilayered data sets ‘on
demand’. Geometry became universal, scales are not fixed any more, and nor are
semiology and visualization properties (Kraak and Ormeling 2010). Map creation is
now available to the creativity of everyone.

5.7 Borders, Fuzziness, Gradients and Ecotones

Defining landscape entities and the delimitation of regions are old problems in
geography. As spatial units, landscapes and regions do seldom have crisp borders,
rather forms fuzzy transition zones with the adjacent units. An important part of
Gran€o’s work (Gran€o 1922, 1935, 1952) was devoted to solve the problems related
to mapping core landscape types and their transitional borders using cartographic
overlay techniques (see Chap. 10). Recognizing the transition zone as a separate
landscape unit is a matter of its width and the threshold set by the researcher
considering the mapping scale used.
In landscape ecology, edges (ecotones), border transitions and gradients
(ecoclines) receive special attention (Naveh and Lieberman 1994; Forman 1995;
Arnot et al. 2007; McGarigal and Cushman 2005). Fuzzy set theory and fractals
have been used in studying borders in the landscape, with applications in landscape
5.8 Interaction Between Spatial Patterns and Processes 95

ecology (Farina 2012), landscape change (Syrbe 1997; Leyk and Zimmermann
2007), landscape archaeology (Mink et al. 2009; Ďuračiová et al. 2013), in percep-
tion studies (Hägerhall et al. 2004), and in landscape design (Bell 2004).

5.8 Interaction Between Spatial Patterns and Processes

Landscapes are structured by spatial patterns of diverse discrete elements and

variations of continuous features. Similar elements can spatially be clustered or
distributed in a random or regular manner. Patterns of different features can relate
to each other in a functional way and may show spatial association, i.e. they
covariate in space. We perceive and experience this coherence as order. When no
relationships between the different elements can be recognised, or no clear structure
can be identified, we experience the landscape as chaotic. Order stands for func-
tional coherence and processes ruled by causal dependency. Chaos stands for
randomness. Both refer to the degree information that is present to allow us to
understand landscape structures and their functioning (see also Chap. 8).
The importance of the analysis of patterns, spatial association and covariation
between landscape features is based on a fundamental paradigm in landscape
ecology, i.e. the dynamical interaction between spatial structures and the ecological
functioning, as expressed by Forman and Godron (1986):
An endless feedback loop:
Past functioning has produced today’s structure;
Today’s structure produces today’s functioning;
Today’s functioning will produce future structure.

Basically this means that the actual spatial structure controls the actual dynamics
in the landscape, which simultaneously transform the structure into a better-adapted
new one. This also means that a structure that is not functional anymore will
gradually become void and vanish. This process can be observed in all traditional
agricultural and pastoral landscapes where the practices that shaped the landscape
and defined its identity and character became lost. Applied to spatial planning, this
paradigm means that there are two options to change the actual situation into a
planned one: taking structural or functional measures, or both. An example is given
in traffic control and safety: to reduce accidents and speeding one can take struc-
tural measures such as making roundabouts, or take measures that affect the
functioning, such as setting speed limits and raising fines.
The study of spatial landscape patterns and their relationship to ecological
processes and functioning of the landscape became the core business of landscape
ecology (Turner et al. 2003; Wu and Hobbs 2002, Wiens and Moss 1999, Gardner
et al. 1990; Turner 1989, Turner et al. 2001). A great variety of methods and
techniques were developed to study this complex relationship (Burel and Baudry
2003; Turner et al. 2001; Klopatek and Gardner 1999; McGarigal and Marks 1995;
Turner and Gardner 1990; Farina 1998).
96 5 Basic Concepts of a Complex Spatial System

5.9 Connectivity and Connectedness

The fundamental holistic nature of landscape is well expressed in first law of

geography according to Tobler (1970): “Everything is related to everything else,
but near things are more related than distant things.” Relations often mean connec-
tions. Hence, the concepts connectivity and -to a lesser extend- connectedness are
widely used in landscape research in diverse contexts and several methods have
been developed to describe and measure the degree of connectivity (see Chap. 8).
Both concepts became popular in conservation biology and landscape ecology,
using the patch-corridor-matrix-mosaic model to study spatial landscape patterns.
A wide range of literature is available on this subject (Forman and Godron 1986;
Antrop 1988; Baudry and Merriam 1988; Turner and Gardner 1990; Metzger and
Décamps 1997; Burel and Baudry 2003). Merriam (1984) introduced the concept of
landscape connectivity, which he defined as the degree to which the landscape
facilitates or impedes movement of species among resource patches. Baudry and
Merriam (1988) use the term connectedness to refer to the fact that two adjacent
patches of the same type are spatially joined, and use the term connectivity for the
possible movement of an individual of a given species between patches, whether or
not they are spatially connected. However, both concepts were often used without
clear distinction. Therefore, to remove the ambiguity, the concepts of structural and
functional connectivity are used instead (Vogt et al. 2009). Structural connectivity
refers to the physical arrangement of landscape elements, i.e. determines connect-
edness. Functional connectivity refers to the species-specific movement potential
through a landscape. A spatial analysis of the structural connectivity is used to
assess or predict the functional connectivity (Goodwin and Fahrig 2002). However,
measures of structural connectivity often have no link to movement behaviour
(Metzger and Décamps 1997).
However, both concepts are also used in other contexts of landscape analysis.
Networks analysis developed from mathematics and topology and became popular
in geography for the study of hydrographical networks, settlement patterns, in
transportation networks and in the analysis of the visual landscape. In landscape
design and architecture, these concepts refer to visual relations between spaces and
objects. Similar uses are found in landscape archaeology.
In all cases, both concepts have different meanings and sometimes the defini-
tions are vague and generic. Often, in domains without a formal definition of
connectivity, the word is as a synonym for connectedness, while in other disciplines
the difference is important. In modern computer science and information technol-
ogy, both concepts are basic. Here, the common noun form is connectivity, but the
common adjectival form is connected.
Connectivity is a basic concept of graph theory and important in the study of
network flow problems, e.g. in hydrographical networks and in some cases also in
transportation networks.
5.11 Reading a Palimpsest 97

Another example of connectivity is found in regular tessellations. Here, the

connectivity describes the number of neighbours accessible from a central tile.
This form is used in space-time geography (see for example Christaller 1933) and
diffusion modelling (see for example Hägerstrand 1967). Examples are found in the
study of territorial patterns of settlements and also in spatial planning for optimiz-
ing service zones and accessibility (see also Chap. 10).

5.10 Multifunctionality

A consequence of the holistic nature of landscape is the multifunctional potential of

the land. This is most clearly expressed by the multiple ecosystem services of the
landscape, in complex land use forms and in rights in using the land. Although the
term ‘multifunctionality’ was not used as such, the principle was present in many
traditional land use systems, such as many agro-pastoral systems, and created
landscapes with a very distinct character. Examples are many of the Mediterranean
landscapes, such as the montado and dehesa, the cultura promiscuita (Pinto-Correia
and Vos 2004), chesnut-forest landscapes (Vos and Stortelder 1992), and wooded
meadows (Emanuelsson 2009). In general, this kind of multifunctionality is often
considered being positive (Vos and Meekes 1999) and the knowledge of managing
these landscapes belongs to our heritage (Austad 2000). Nevertheless, many of
these traditional landscapes have been threatened by modern developments and are
rapidly vanishing and so is the knowledge that maintained them (Pinto-Correia
1993; Vos and Stortelder 1992; Vos 1993).
Hence, the concepts ‘multifunctionality’ and ‘mulifunctional landscapes’ as
such gained a renewed attention around the beginning of the new millennium, as
demonstrated by the international conference on the matter held in 2000 in
Roskilde, Denmark (Brandt and Vejre 2004). Multifunctional landscapes can be
seen as one of many strategies towards a sustainable development at the landscape
level (Brandt and Vejre 2004; Haines-Young and Potschin 2004). Different types of
multifunctionality can be recognised based on spatial and temporal criteria (Brandt
and Vejre 2004) and scale (Antrop 2004) (see also Chap. 12).

5.11 Reading a Palimpsest

The landscape has been compared to a palimpsest, a manuscript on expensive

parchment that has been scraped off several times so that it can be re-used. It refers
to successive time layers in the development of the landscape, where older ones are
only vaguely and partially visible compared to the present ones. Understanding the
landscape is like reading such an ancient manuscript, written in fonts and a
language different from the one we use today. Deciphering the manuscript demands
98 5 Basic Concepts of a Complex Spatial System

careful and systematic reading and the method consists of reading different layers in
succession. There are four main layers in reading the landscape:
• a scene offering an experience.
• a natural, physical system that forms the substrate of the land
• a cultural system with places and territories and land use
• a history that remains in successive, incomplete layers
The first layer is the scenery, which can ‘read’ as a work of art, as a painting,
although it is more like experiencing a theatre play. Perceiving and experiencing the
landscape will be discussed in Chap. 6.
Following layers demand a more systematic, ‘scientific’ reading.
The natural, physical system the substrate that gave opportunities and restric-
tions to humans to live on the land and shape it into a landscape. It carries the
cultural layer. Both will be discussed in detail in Chap. 9.
All these layers are essentially dynamic and transform in different ways, speed
and scale influenced by a series of equally dynamic driving forces. Successive time
layers make the palimpsest of the unique history of each landscape unit. This will be
treated in Chap. 7.


Agarwal, C., Green, G. M, Grove, J. M., Evans, T. P., & Schweik, C. M. (2002). A review and
assessment of land-use change models: Dynamics of space, time, and human choice. Gen.Tech.
Rep. NE-297. Newtown Square, PA:U.S. Radnor: Department of Agriculture, Forest Service,
Northeastern Research Station, 67 pp.
Antrop, M. (1988). Invisible connectivity in rural landscapes. In K.-F. Schreiber (Ed.), Proceed-
ings of the Second International Seminar of IALE, connectivity in landscape ecology
(pp. 57–62). Münster: Münsterische Geographische Arbeiten 29.
Antrop, M. (2004). Assessing multi-scale values and multifunctionality in landscapes. In J. Brandt
& H. Vejre (Eds.), Multifunctional landscapes, Theory, values and history (Vol. I,
pp. 165–180). Southampton: WIT Press.
Antrop, M. (2006). From holistic landscape synthesis to transdisciplinary landscape management.
In B. Tress, G. Tress, G. Fry, & P. Opdam (Eds.), From landscape research to landscape
planning: aspects of integration, education and application, Wageningen UR frontis series, N
12 (pp. 27–50). Dordrecht: Springer.
Arnot, C., Fisher, P., 2007. Mapping the ecotone with fuzzy sets. In: Morris A. Kokhan S. (eds.),
Geographic uncertainty in environmental security, Springer, Dordrecht, 19–32.
Austad, I. (2000). The future of traditional agriculture landscapes: retaining desirable qualities. In
J. Klijn & W. Vos (Eds.), From landscape ecology to landscape science, WLO (pp. 43–56).
Wageningen: Kluwer Academic Publishers.
Baudry, J., & Merriam, H. G.. (1988). Connectivity and connectedness: Functional versus struc-
tural patterns in landscapes. In K. F. Schreiber (Ed.) Connectivity in landscape ecology
(pp. 23–28). Proceedings of the 2nd International Association for Landscape Ecology.
Münstersche Geographische Arbeiten, 29.
Bell, S. (2004). Elements of visual design in the landscape. London: Spon Press.
Bertin, J. (1983). Semiology of graphics. Madison: University of Wisconsin Press.
References 99

Borisch, W. (2007). When the volcanoes. . . “Notes about the occurrences between the years 1730
and 1736” written by the Priest of Yaiza, Don Andrés Lorenzo Curbelo (32 pp). Lanzarote:
Editorial Yaiza S.L.
Brandt, J., & Vejre, H. (2004). In J. Brandt & H. Vejre (Eds.), Multifunctional landscapes. Theory,
values and history (Vol. I, 3–32 pp). Southampton: WIT Press.
Burel, F., & Baudry, J. (2003). Landscape ecology: Concepts, methods and applications (362 pp).
Enfield/New Hampshire: Science Publishers, Inc..
Christaller, W. (1933). Die zentralen Orte in S€ uddeutschland. Jena: Fischer.
Dawkins, R. (1996). The blind watchmaker. London: Norton.
De Keyzer, M., Jongepier, I., & Soens, T. (2014). Consuming maps and producing space.
Explaining regional variations in the reception and agency of mapmaking in the Low Countries
during the medieval and early modern periods. Continuity and Change, 29, 209–240.
Ďuračiová, R., Lieskovský, T. Stopková, E., & Kročková, K. (2013). The benefit of fuzzy logic to
protection of cultural and historical heritage. GIS Ostrava 2013 - Geoinformatics for City
Emanuelsson, U. (2009). The rural landscapes of Europe. How man has shaped European nature.
The Swedish Research Council Formas, p. 383.
Fahrig, L., Baudry, J., Brotons, L., Burel, F. G., Crist, T. O., Fuller, R. J., Sirami, C., Siriwardena,
G. M., & Martin, J.-L. (2011). Functional landscape heterogeneity and animal biodiversity in
agricultural landscapes. Ecology Letters, 24, 101–112.
Farina, A. (1998). Principles and methods in landscape ecology. London: Chapman & Hall.
235 pp.
Farina, A. (2012). Landscape ecology in action. Dordrecht: Springer. 317 pp.
Forman, R. T. T. (1995). Land mosaics: The ecology of landscapes and regions. Cambridge:
Cambridge University Press.
Forman, R., & Godron, M. (1986). Landscape ecology (619 pp). New York: Wiley.
Gardner, R. H., & O’Neill, R. V. (1990). Pattern, process and predictability: The use of neutral
models for landscape analysis. In M. G. Turner & R. H. Gardner (Eds.), Quantitative methods
in landscape ecology (pp. 289–307). New York: Springer-Verlag.
Garrigues, S., Allard, D., Baret, F., & Weiss, M. (2006). Quantifying spatial heterogeneity at the
landscape scale using variogram models. Remote Sensing of Environment, 103, 81–96.
Goodwin, B. J., & Fahrig, L. (2002). How does landscape structure influence landscape connec-
tivity? Oikos, 99, 552–570.
Gran€o, J. G. (1922). Eesti maastikulised u€ksused, Loodus I, 2, pp. 105–123; 4, pp. 193–214; 5, pp.
Gran€o, J. G. (1935). Geographische Ganzheiten. Petermanns Mitteilungen, 81, 295–301.
Gran€o, J. G. (1952). Régions géographiques et une méthode pour les délimiter. Compte Rendu du
XVIè Congrès International de Géographie, Lisbonne 1949, 11 pp.
Hägerhall, C., Purcella, T., & Taylor, R. (2004). Fractal dimension of landscape silhouette outlines
as a predictor of landscape preference. Journal of Environmental Psychology, 4, 247–255.
Hägerstrand, T. (1967). Innovation diffusion as a spatial process (A. Pred, Trans.). Chicago:
University of Chicago Press.
Haines-Young, R., & Potschin, M. (2004). Valuing and assessing of multifunctional landscapes:
An approach based upon the natural capital concept. In J. Brandt & H. Vejre (Eds.),
Multifunctional landscapes. Theory, values and history (Vol. Vol. I, 181–192 pp). Southamp-
ton: WIT Press.
Hard, G. (1970). Der ‘Totalcharakter der Landschaft’. Re-Interpretation einiger Textstellen bei
Alexander von Humboldt. Erdkundliches Wissen, 49–71.
Harley, J. B. (1988). Maps, knowledge, and power. In D. Cosgrove & S. Daniels (Eds.), The
iconography of landscape (pp. 277–312). Cambridge: Cambridge University Press.
Huff, D. (1954). How to lie with statistics. New York: Norton.
Katayama, N., Amano, T., Naoe, S., Yamakita, T., Komatsu, I., S-i, T., et al. (2014). Landscape
heterogeneity–biodiversity relationship: effect of range size. PloS One, 9(3), e93359.
100 5 Basic Concepts of a Complex Spatial System

Klopatek, J. M., & Gardner, R. H. (Eds.). (1999). Landscape Ecological Analysis. Issues and
Applications (400 pp). New York: Springer-Verlag.
Korzybski, A. (1933). Science and sanity: An introduction to non-aristotelian systems and general
semantics, supplement III: A non-aristotelian system and its necessity for rigour in mathemat-
ics and physics (pp. 747–761). New York: Institute of General Semantics.
Kraak, M.-J., & Ormeling, F. (2010). Cartography: Visualization of geospatial data (p. 198).
Harlow: Prentice Hall/Pearson Education Limited.
Küster, H. (2008). Kulturlandschaften: Analyse und Planung (p. 180). Frankfurt: Peter Lang.
Leyk, S., & Zimmermann, N. E. (2007). Improving land change detection based on uncertain
survey maps using fuzzy sets. Landscape Ecology, 22, 257–272.
Li, H., & Reynolds, J. F. (1995). On definition and quantification of heterogeneity. Oikos, 73(2),
Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rind, D. W. (2001). Geographic information
systems and science. New York: Wiley, & Sons.
Mander, Ü., Uuemaa, E., Roosaare, J., Aunap, R., & Antrop, M. (2010). Coherence and fragmen-
tation of landscape patterns as characterized by correlograms: A case study of Estonia.
Landscape and Urban Planning, 94(1), 31–37.
McGarigal, K., & Cushman, S. A. (2005). The gradient concept of landscape structure. In J. A.
Wiens & M. R. Moss (Eds.), 1999. Issues in landscape ecology, Studies in landscape ecology
(pp. 112–119). Cambridge: Cambridge University Press.
McGarigal, K., & Marks, B. J. (1995). FRAGSTATS: Spatial pattern analysis program for
quantifying landscape structure. USDA For. Serv. Gen. Tech. Rep. PNW-351.
Merriam, G. (1984). Connectivity: A fundamental ecological characteristic of landscape pattern.
In J. Brandt & P. Agger (Eds.), Proceedings of the 1st international seminar on methodology in
landscape ecological research and planning (pp. 5–15). Denmark: Roskilde University.
Metzger, J.-P., & Décamps, H. (1997). The structural connectivity threshold: an hypothesis in
conservation biology at the landscape scale. Acta Oecologica, 18(1), 1–12.
Mink, P. B., Ripy, J., Bailey, K., & Grossardt, T. H. (2009). Predictive Archaeological Modeling
using GIS- Based Fuzzy Set Estimation: A Case Study in Woodford County, Kentucky.
Kentucky Transportation Center Faculty and Researcher Publications. Paper 12. http://
Monmonier, M. (2005). Lying with maps. Statistical Science, 20(3), 215–222.
Naveh, Z., & Lieberman, A. (1994). Landscape ecology: Theory and application (360 pp).
New York: Springer Verlag.
Olwig, K. R. (2004). “This is not a landscape”: Circulating reference and land shaping. In
H. Palang, H. Sooväli, M. Antrop, & S. Setten (Eds.), European rural landscapes: Persistence
and change in a globalising environment (pp. 41–66). Dordrecht: Kluwer Academic
Pinto-Correia, T. (1993). Threatened landscape in Alentejo, Portugal: The ‘montado’ and other
‘agro-silvo-pastoral’ systems. Landscape and Urban Planning, 24(1–4), 43–48.
Pinto-Correia, T., & Vos, W. (2004). Multifunctionality in mediterranean landscapes – Past and
future. In R. Jongman (Ed.), The new dimensions of the European landscape, EU Frontis
Series. Wageningen: Springer.
Smuts, J. C. (1926). Holism and evolution. New York: The Macmillan Company.
Stanners, D., & Bourdeau, P. (Eds.). (1995). Europe’s environment. The Dobris assessment, EC
DG XI and Phare (p. 676). Copenhagen: European Environment Agency.
Syrbe, R.-U. (1997). Landscape evaluation of heterogeneous areas using fuzzy sets. Cybergeo:
European Journal of Geography [online], Dossiers, 10ème Colloque Européen de Géographie
Théorique et Quantitative, Rostock, Allemagne, 6–11 septembre 1997, document 40, http://
Tobler, W. (1970). A computer movie simulating urban growth in the Detroit region. Economic
Geography, 46(2), 234–240.
References 101

Turner, M. G. (1989). Landscape ecology: The effect of pattern on process. Annual review of
Ecology and Systematics, 20,171–197. In J. H. Wiens, M. R. Moss, M. G. Turner, & D. J.
Mladenoff (Eds.), 2007. Foundation papers in landscape ecology (pp. 545–571). New York:
Columbia University Press.
Turner, M. G., & Gardner, R. H. (Eds.). (1990). Quantitative methods in landscape ecology.
New York: Springer-Verlag.
Turner, M. G., Gardner, R. H., & O’Neill, R. V. (2001). Landscape ecology in theory and practice:
Pattern and process. New York: Springer-Verlag.
Turner, N. J., Davidson-Hunt, I. J., & O’Flaherty, M. (2003). Living on the edge: Ecological and
cultural edges as sources of diversity for social–ecological resilience. Human Ecology, 31(3),
Turner, M. G., Donato, D. C., & Romme, W. H. (2013). Consequences of spatial heterogeneity for
ecosystem services in changing forest landscapes: Priorities for future research. Landscape
Ecology, 28(6), 1081–1097.
Unwin, D. (1981). Introductory spatial analysis (212pp). London: Methuen.
Van Wirdum G., 1981, Design for a land ecological survey of nature protection. In: Tsjallingii &
de Veer (eds.), Perspectives in landscape ecology. Proceedings of the Intern (pp. 245–251).
Wageningen: Congress of the Netherlands Society for Landscape Ecology, Centre for Agri-
cultural Publishing and Documentation
Vogt, P., Ferrari, J. R., Lookingbill, T. R., Gardner, R. H., Riitters, K. H., & Ostapowicz,
K. (2009). Mapping functional connectivity. Ecological Indicators, 9, 64–71.
von Humboldt, A. (1807). Ansichten der nature. eBook-Originalausgabe, eClassica, Aurabooks.
Vos, W. (1993). Recent landscape transformation in the Tuscan Apennines caused by changing
land use. Landscape and Urban Planning, 24(1–4), 63–68.
Vos, W., & Meekes, H. (1999). Trends in European cultural landscape development: perspectives
for a sustainable future. Landscape and Urban Planning, 46(1–3), 3–14.
Vos, W., & Stortelder, A. H. F. (1992). Vanishing Tuscan landscapes, landscape ecology of a sub-
mediterranean-montane area (Solano basin, Tuscany, Italy). Wageningen: Pudoc.
Wiens, J. A., & Milne, B. T. (1989). Scaling of “landscape” in landscape ecology, or, landscape
ecology from a beetle’s perspective. Landscape Ecology, 3, 87–96.
Wiens, J. A., & Moss, M. R. (eds.) (1999). Issues in landscape ecology. International Association
for Landscape Ecology, Fifth World Congress, Snowmass Village, Colorado, USA.
Wu, J., & Hobbs, R. (2002). Key issues and research priorities in landscape ecology: An
idiosyncratic synthesis. Landscape Ecology, 17(4), 355–365.
Wulf, A. (2015). The invention of nature – Alexander von Humboldt’s new world (p. 473).
New York: Alfred A. Knopf.
Zonneveld, J. I. S. (1985). Levend land. De geografie van het Nederlandse landschap. Scheltema
& Holkema/Utrecht/Antwerpen: Bohn.
Zonneveld, I. S. (1995). Land ecology (199 pp). Amsterdam: SPB Academic Publishing bv.
Chapter 6
Sensing and Experiencing the Landscape

Abstract The concept landscape implies the perception by a human observer.

People sense and experience their environment holistically, using all senses simul-
taneously. Gestalt-principles apply here. Various disciplines study landscape per-
ception and the mental information processing resulting in mindscapes and adapted
behaviour. Several theories have been formulated to explain this process using
biological, cultural and individual factors. Perception is often restricted to the visual
landscape. Landscape experience refers to the whole arousal resulting from sensing
the landscape. Landscape preference focuses upon the assessment we make of this
experience. The properties and conditions of human vision help to understand how
we analyse the landscape scenery and define basic concepts for all kinds of
visualisations, as in painting, photography and computer modelling. Landscape
experience research follows either an objectivist paradigm, aiming to identify
physical landscape properties that can be related to preferences, or a subjectivist
paradigm, focusing on the psychological and sociological response. The first one is
a landscape centred approach, the second one focuses on the observer and his social
and cultural background.

Keywords Perception • Visual landscape • Gestalt • Landscape preference

6.1 Introduction

Our knowledge about the landscape comes through our senses and the experiences
we get from this perception. Our senses and mind work in a holistic manner and our
perception is filtered and coloured by our previous experiences and our knowledge.
Thus, the landscape is also a subjective reality and relative to each observer
individually. Nevertheless, sensing the environment and the mental processing of
information and learning are more or less similar to all humans. It is possible to
study these processes in an objective way and thus gain a better understanding of
the landscape and our relation to it.
This chapter deals with landscape perception, experience and preferences as
different phases in this holistic information processing. First, ‘technical’ and phys-
iological properties of our vision, in particular, are discussed, as our vision is the
dominating sensing of the landscape. The technical properties of vision help

© Springer Science+Business Media B.V. 2017 103

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_6
104 6 Sensing and Experiencing the Landscape

understand how our perception actually works and form the basis of specific
concepts and definitions used in the study of the visual landscape. Also, Gestalt-
principles, which control our sensing, are discussed as they help understand the
interpretation and meaning we give to complex patterns and eventually why
confusion may occur. They also make the link to the holistic conception of the
landscape. Finally, the psychological and cognitive aspects are summarised. Dif-
ferent methods are presented and links are made with ‘objective’ tools of sensing
and visualising the landscape, such as photography.

6.2 Landscape Perception, Experience and Preference

The fundamental distinction between the concepts land and landscape is that
landscape includes the perception by a human observer and the representation he
or she makes of it. People sense and experience their environment holistically,
using all senses simultaneously. Gestalt-principles apply here. We are good at
processing complex patterns and in dealing with a lot of extremely diverse infor-
mation. All information we gather through our senses is immediately processed by
our mind and compared to what we know already. This process is simultaneously a
process of learning. Besides the landscape out-there, we have it also in our mind, as
a mental map or mindscape. The concept landscape unifies the material reality in
the world with a subjective mental representation we make of it and the knowledge
that is essential to orient ourselves and behave in our environment. This means that
different observers looking at the same tract of land see and experience different
landscapes. In this sense landscape is a relativistic phenomenon.
There are many ways of seeing the landscape, which is expressed in many words
we use to specify. As Denis Cosgrove clearly discussed: “A glance is different from
a stare, a sight is different from a vision. In considering the active use of the sense of
sight most languages make a fundamental distinction between seeing and looking
(. . .). The former suggests the passive and physical act of registering the external
world by eye; the latter implies an intentional directing of the eyes towards an
object of interest. In English, viewing implies a more sustained and disinterested
use of the sense of sight; while witnessing suggests that the experience of seeing is
being recorded with the intention of its verification or subsequent communication.
Gazing entails a sustained act of seeing in which emotion is stirred in some way,
while staring holds a similar meaning but conveys a sense of query or judgement on
the part of the starer. (. . .) As the dual meaning of ‘I see’ indicates connections
between seeing and cognition are similarly complex. ‘Insight’ captures the human
capacity to ‘see’ more than it immediately visible to the eye (. . .). ‘Vision’ is at once
the physiological function and an imaginative capacity in which non-material
phenomena are somehow witnessed.” (Cosgrove 2002).
Sensation and experience of the environment define the concept landscape itself.
The concept landscape changed each time technology of perception changed, such
as with the introduction of photography, and new modes of representation were
6.2 Landscape Perception, Experience and Preference 105

introduced, such as constructing perspectives, mapping and computer visualisations

(Cosgrove 2002).
In his book Reine Geographie (Gran€o 1929; English translation Pure Geography
1997), Johannes Gran€o integrates the human way of sensing in a method of
mapping the perceived environment, as he calls it. He even uses this to define the
landscape, making the fundamental distinction between Fernsicht (or landscape)
and Nahsicht (proximity). He defines landscape as “a visible distant environment,
or distant field of vision” (Gran€o and Paasi 1997) and proximity as “that part of the
environment that is perceivable with all the senses and is situated between the
observer and the landscape” (Gran€o and Paasi 1997). He admits that the boundary
between the proximity and the landscape cannot be determined accurately and that
the distance to it varies and a ‘transitional zone’ can be recognised.
Various disciplines study landscape perception and the mental information
processing resulting in mindscapes and adapted behaviour. Environmental psychol-
ogy, human geography, sociology and landscape architecture have specific
approaches to this understanding, but share, explicitly or implicitly, three core
assumptions (Jacobs 2006): (1) the way people perceive landscapes is influenced,
but not determined by physical landscape attributes, (2) a complex mental process
of information reception and processing mediates between the physical landscape
and the mindscape, and (3) the factors that influence this process can be divided into
biological, cultural and individual factors (Bourassa 1990, 1991).
How we sense, experience and understand the landscape is a complex matter.
Various concepts are used and their definitions vary according to the context they
are used in, i.e. in landscape perception, experience and preference. Perception is
often restricted to the sensing of visual information and focuses upon the landscape
scenery, the visual landscape. Visual perception dominates by 80% the other senses
and probably the visual information was most important for our survival. Hence, the
emphasis of the research lies in the visual landscape and the scenery. Second,
comes sound. The Canadian composer and environmentalist and founder of the
World Soundscape Project, R. Murray Schafer (1977) introduced the concept
soundscape. It denotes the acoustic environment that can be experienced by humans
and acoustic ecology studies the relationship between living beings and their
environment (Wrightson 2000).
The visual information is the key and information from other senses is linked to
it, as are also memories, affections and facts. This makes us remembering smells or
sounds when looking at a landscape or even its representation in a painting or
photograph. Hence, a research field of landscape visualisation developed, with
applications in environmental impact studies and design research (Nijhuis et al.
These associations create expectations and set a framework for valuing what we
see. To address all these aspects, the terminologies landscape experience and
landscape preference are used as well. Landscape experience is the broadest
concept and refers to the whole arousal resulting from sensing the landscape.
Landscape preference focuses upon the assessment we make of this experience in
terms of like-dislike, beautiful or not, useful of not. The distinction between
106 6 Sensing and Experiencing the Landscape

objective aspects of perception and subjective psychological aspects is extremely

difficult to make. In a large extent, our knowledge steers and controls what we see
and how we see it and this makes landscape experience subjective and individual.
However, one can assume that these processes are largely similar between
individuals of the same species and some of the physiological and psychological
aspects can be studied in a scientific objective way (van Campen 1993; McConnell
1989). Hence, the idea emerged to link physical properties of the landscape to the
perceptive response, which leads to different paradigms and models in landscape
preference research. Three approaches can be distinguished:
• the focus is on the ‘mechanical’, ‘technical’ and physiological aspects of per-
ception. An example is the use of eye-tracking;
• the focus is on the mental information processing. This is related to the study of
learning processes and mental representations in mental maps or cognitive maps;
• the focus is on the preference in relation to the psychological, social and cultural
properties of the observers.
Although landscape perception and experience are highly subjective and diffi-
cult to link to scenic indicators in the landscape, a lot of research has been done in
this field (Fry et al. 2009; Ode 2005) and new methods have been introduced such as
eye-tracking (Dupont et al. 2014). Many practical questions in landscape planning
and design aim to use this relationship objective perception and subjective experi-
ence to create specific preferences by appropriate choice and arrangement land-
scape elements and their scenic properties. Gestalt-principles were used – not
always knowingly – in garden and landscape design (Antrop 2007) (see
Chap. 11). Research includes defining indicators of the visual landscape for land-
scape preference (Ode et al. 2009; Dramstad et al. 2006), the assessment of visual
landscape character, e.g. in landscape archaeology (Fry et al. 2004), defining
intrinsic landscape values (Antrop 2012), landscape aesthetics and the assessment
of beauty (Bourassa 1991; Sevenant and Antrop 2009), analysing the visual land-
scape and modelling its visualization (Nijhuis et al. 2011; Nijhuis 2015).
The research of landscape experience is very complex as the environmental
context and perceptible realm are interacting. This places the aesthetic experience
in an ecological relationship (Gobster et al. 2007). Landscape properties and
physiological and psychological and cultural factors are interacting and influence
behaviour and actions (Sevenant and Antrop 2010a, b) and landscape architecture
(Nijhuis 2011).

6.3 Seeing the Landscape

6.3.1 The Human Vision

Visual perception dominates the information we sense. Physiological properties of

our vision define in a high degree ‘technical’ or ‘mechanical’ settings of our
6.3 Seeing the Landscape 107

Fig. 6.1 Field of vision of human eyes (Binocular vision gives an overlap of the field of vision of
both eyes with a span of approximately 130 . Here, we have the best resolution and a coloured
view, while at the periphery the details become blurred and colours disappear)

perception. Many of these can be studied empirically and used to define quantitative
thresholds in defining views and vistas. For example, a semi-panoramic view can be
defined formally as a view from an immobile position and head with a horizontal
angle of at least 180 with no obstacles at a distance closer than 1200 m. These
numerical values correspond to the average horizontal view angle with two eyes
and the critical viewing distance.
The average field of view of a human eye is 95 away from the nose, 60 toward
the nose, 75 downward and 60 upward, which makes an almost 180-degree
forward-facing field of view. Thus, a quick glance of our eyes corresponds to a
wide-angle view that, however, is only sharp and coloured in the centre, and
progressively becoming less sharp and less colourful towards the periphery (Fig. 6.1).
Our eyes move constantly (saccades) exploring a scene and dynamically adjust
exposure and focus on objects at a variety of distances. Looking around encom-
passes a broader angle of view (Fig. 6.2). The eyes trace contours and fix on objects
that are conspicuous in the scene according to their size, shape and colour (Fig. 6.3).
Our vision works very differently from a still camera. It is rather like a video
camera, panning the scene, focusing on some objects and compiling snapshots to
construct a 3D mental image of the whole that is seeming without distortion. No
camera lens can duplicate what the eyes see in reality, in terms of magnification, the
field of focus and the general angle of view including the out-of-focus periphery.
The eye appears to combine a narrowly focused angle of view as a long telephoto
lens, the magnification of a standard lens, and the general angle of view,
corresponding to a wide-angle lens, but out-of-focus at the edges.
The perspective we see objects depends on the focal length of the lens and upon
the viewing distance. The theoretical viewing distance depends on the eye-height
(h in meter), and was calculated by Gran€o (1929) as follows:

Viewing distance ðin meter Þ ¼ 3:827√h

When standing on a butte of 10 m and with an eye-height of 1.5 m we see

theoretically 12.1 km far.
At a viewing distance of more than 1000 m, we see all objects with a view angle
of less than 5 , which does not allow to see differences in size. At approximately
1200 m, most people have no stereo-vision anymore (Meienberg 1966; Middleton
108 6 Sensing and Experiencing the Landscape

Fig. 6.2 The human field of vision with the head in a static position. Horizontal: EB eye-base,
VLRE view limit right eye, VLLE view limit left eye, MVR monocular view right, MVL
monocular view left, ER eye rotation angle, SBV stereoscopic binocular vision, CV colour vision,
SR symbol recognition, WR word recognition, LS Line of sight. Vertical: VFOV vertical field of
view, LSS line of sight standing, LSs line of sight sitting (After Panero and Zelnik (1979))

1968; Van der Ham et al. 1970) (Fig. 6.6). Small elements, such as individual trees
in a forest edge or architectural elements of a building cannot be discerned
anymore. This critical viewing distance depends on the visual acuity expressed as
resolution and in particular our stereoscopic resolving power. Gran€o (1929) used
unknowingly this critical viewing distance in his differentiation between the con-
cepts ‘proximity’ (Nahsicht) and ‘landscape’ (Fernsicht) (see Chap. 3).
The eye is capable of responding to an enormous range of light intensity
exceeding a variation of over 10 billion-fold and using a non-linear (logarithmic)
response to brightness. The mode of vision and the sensitivity of the eyes change
with the illumination level and wavelength (Figs. 6.4, 6.5 and 6.6). The difference
in illumination between an object and its surroundings defines its contrast. A
minimal contrast is necessary to discern an object and this varies according to the
intensity level. The incremental threshold of noticeable light intensity change (ΔI)
has a constant relationship with the intensity level I, i.e. ΔI/I ¼ constant. Table 6.1
summarises the most important properties of the human vision. Note that the colour
vision and acuity of our eyes shifts with the illumination conditions of the land-
scape, acting as a filter upon our perception.

6.3.2 Concepts and Definitions

To describe and study the human perception of the landscape, several concepts have
been defined in a formal way:
6.3 Seeing the Landscape 109

Fig. 6.3 Eye-tracking follows the fixation points of both eyes (1) and measures the duration of the
fixation as well as the movements (saccades) from one point to the next (2). A heat map (3) gives
the synthesis showing the areas that receive most of the attention (Dupont et al. 2014)

• Average eye-height: is used in the calculation of viewsheds in GIS. Default

values are often 1.5 m and 1.7 m.
• Viewpoint: the location the observer is standing.
• Viewing distance: distance between the observer and an obstacle (object,
screen), to a layer in the scene or to the skyline.
• Critical viewing distance: the distance where contours of objects become fuzzy,
small details disappear in the texture of the background. Corresponds to the
limits of stereovision. Approximately 1200 m.
• Viewing direction: the direction one is looking to; is not necessarily horizontal.
Consists of two components: the inclination (up-down) and the azimuth (heading
from the North for example).
• View angle: the view expressed as an angle (horizontal/vertical, in degrees or arc
degrees) seen at a fixed position without moving eyes or head. When using
camera lenses view angles may differ from the human vision and determine the
110 6 Sensing and Experiencing the Landscape

Fig. 6.4 Luminance and visual acuity of the human eye (after different sources)

Fig. 6.5 Spectral response of the human eye (after different sources)

distortions of the perspective. A normal lens corresponds approximately with the

focus angle of the human view and has a 43 mm equivalent focal length.
• Viewing circle: an observation from a fixed position with an angle of 360 .
• Viewing sector: views formed by successive partially overlapping layers.
• Field of view: (often horizontal) angle of view from a fixed position.
• Scene: representation of the landscape as projected on a plane (as on a photo-
graph or painting).
• Depth layer: an imaginary plane containing all elements at a similar distance
from the observer and seen at the same scale and in the same degree of detail.
The view is often composed of successive, partially overlapping layers as the
flats in a theatre. Three important layers are referred to as foreground, centre and
• Horizon: the horizontal plane in the scene at eye-height. It defines the distortions
of the perspective one see. It is different from the skyline.
• Skyline: the silhouette of the landscape as seen against the sky. The ultimate
layer in a view.
6.3 Seeing the Landscape 111

Fig. 6.6 The critical viewing distance. The size of objects seen in a projected image decreases
with the distance to the observer. In the zone of the critical viewing distance, height differences
become very small and stereo-vision stops. The objects we see merge together in the background

• Vista: a focal view, often framed by a foreground or directed towards some

interesting feature or landmark. The field of view often corresponds to the focus
angle of the human eye.
• Eye-catcher: an object that strongly contrasts with its surroundings by its size,
shape, colour or material.
• Landmark: an element, often an eye-catcher as well, used to position and orient
oneself in the landscape. This can be an object or feature with important
symbolic value.
With the development of computer-based mapping and analysis of the visual
landscapes, and techniques for visualisation and rendering, several new concepts
have been introduced. However, it is useful here to note that these caused some-
times a shift in meaning with the definitions used in perception studies.

6.3.3 Landscape as a Scene: Depth Layers, Viewing Sectors

and Skyline

We can conceive a landscape scene as a 2D image projected on a plane perpendic-

ular to the viewing direction (Fig. 6.7). Different rules of perspective apply,
illustrated in Figs. 6.8, 6.9 (see also Chap. 11).
112 6 Sensing and Experiencing the Landscape

Table 6.1 Properties of human vision

Horizontal view angle
- with one eye ca. 94 –104
- with two eyes ca. 180
Stereoscopic focus span ca. 125
Horizontal range colour vision ca. 30 –60
Vertical view angle ca. 120
Vertical range colour vision ca. 70
Angular resolution (in arc seconds “)
- Monoscopic - with strong illumination (the photopic vision) 6000
- with weak illumination (the scotopic vision) 12000
(corresponds to 10 lines/mm at a distance of 20 cm)
- Stereoscopic 1000
Critical viewing distance ca. 1200 m
Theoretical viewing distance at eye height of 1.5 m 4.7 km
Minimal perceivable contrast - with weak illuminationa 5%
- with strong illumination 2%
Spectral sensitivity (wavelength range visible light) 400 nm–700 nm
Max. sensitivity - with a photopic vision: in green wavelength band 555 nm
- with a scotopic vision: in blue wavelength band 507 nm
Number of colours (photopic vision) > 16 million
Colour model used to describe colours HSIb
Maximal colour sensitivity of the photoreceptor cells: the rods 500 nm (blue-green)
the cones 550 nm (green-yellow)
Contrast is the relative difference of light reflection between an object and its background or two
adjacent objects
Hue (colour), Saturation, Intensity. Monitor screens use the RGB colour model, printing uses the
CMYK model

The contour of the landscape with the sky forms the skyline (Fig. 6.10). Horizon
and skyline fall together only in perfectly horizontal and flat landscapes with an
observation at ground level. The shape of the skyline is typical for certain landscape
types, such as an urban landscape, a polder landscape or a mountain landscape
(Fig. 6.14). The fractal dimension of the skyline has been used to differentiate
between landscape types (Hägerhäll et al. 2004).
Partially overlapping landforms and silhouettes of constructions and screens at
different depth layers segment the scene into viewing sectors or sub-scenes
(Fig. 6.11). When analysing the scene as a whole, these sub-scenes give the first
steps in the differentiation of the holistic Gestalt into parts that are analysed more in
detail in the process of understanding the whole pattern.
The degree of detail we see decreases with the distance from the observer. This
causes the apparent grouping of all elements at a similar distance into scenic layers.
In most cases three depth layers can be recognised: the foreground, the middle and
the background (Fig. 6.12). The shift in the coarseness and density of the texture in
the image enhances the depth layers we see (Figs. 6.13 and 6.14).
6.3 Seeing the Landscape 113

Fig. 6.7 The 2D landscape perspective: we perceive the landscape as projected on a vertical plane
that intersects the horizontal line of sight at eye-height (O) at c, which is the centre of the
perspective. The horizontal line through at eye-level defines the horizon (H) and the horizontal
plane at the level of the observer forms the ground line (G). The silhouette of the projected
landscape forms the skyline (S) and partially overlapping successive scenes (1–5) make the
viewing depth layers. The vertical image size of objects decreases with the viewing distance and
also the differences between the heights of objects decreases. The height of the house (2) at a
viewing distance B has a projected image height b, which in this case is the same as the image
height (a) of the mountain viewed at distance A. At a viewing distance of more than 1000 m we see
all objects with a view angle of less than 5 which does not allow to see differences in size
(M. Antrop 2007)

6.3.4 Photographic Perspectives

Photography allows a fast registration of landscapes and is often used as a scientific

description tool. However, photography rarely represents the landscape in a neutral
way. As already discussed, our vision works very differently from a still camera. A
camera gives a static image with geometric and illumination properties depending
on the lenses and settings of the camera used.
The distance to the subject and the angle of view define the perspective of the
photograph. Primarily it is the distance from the camera to the subject that causes
the perspective distortion (Fig. 6.15). A camera using different focal lengths
shooting the same subject from the same position and distance will produce
photographs with the same perspective. However, changing the distance to the
subject using the same lens will change the relative sizes and spatial relationships
among the different objects represented in the photograph (Fig. 6.16). The angle of
114 6 Sensing and Experiencing the Landscape

Fig. 6.8 Lines that are parallel to the projection plane are seen as horizontal and vertical lines.
Parallel lines that are oblique to the projection plane converge in a vanishing point (VP) that
defines the horizon (H). This is called a one-point linear perspective (M. Antrop 2007)

Fig. 6.9 When the eyes move vertically to fixate points of interest at different elevations, multiple
horizons are defined. The successive vanishing points (VP1 and VP2) each have a one-point
perspective. This creates an illusion of foreshortening (M. Antrop 2007)
6.3 Seeing the Landscape 115

Fig. 6.10 A photograph perspective. The optical centre of the camera C defines the horizon H. In
this case, the horizon does not correspond to any tangible landscape feature such as the skyline or
the texture thresholds that define the depth layers. However, it is close to the focus of interest (the
eye-catcher E). Photographers often use a composition following the 1/3 rule. In this case, it
enhances the three main depth layers (Photo of Widecombe, M. Antrop 2004)

Fig. 6.11 Depth layers, viewing sectors and eye-catcher (View from Launceston Castle, UK,
photo M. Antrop 2003)

view of the lens is dependent on the focal length in relation to the sensor size. It
defines the size the subject is represented (magnification) and the span of the
A ‘normal lens’ or ‘standard lens’ has a perspective that corresponds more or less
that of the naked human eye, i.e. the relative sizes and distances between object in
116 6 Sensing and Experiencing the Landscape

Fig. 6.12 Depth layers: the viewing distance determines the detail and image size. Objects at a
similar viewing distance are grouped into one depth layer. Depth layers differ in the texture density
of the detail. Generally at least three depth layers are formed: foreground, middle and background.
Background starts often at the critical viewing distance when details cannot be discerned anymore
(View from Belvédère, Han-sur-Lesse, Belgium, photo M. Antrop 2006)

Fig. 6.13 Elements that are uniformly distributed in space and looked at in a horizontal perspec-
tive show an increasing density with viewing distance. Texture density seems to change with
discontinuous steps, which define depth layers (After Gibson 1950)

the image corresponds more or less what the naked eye sees. A ‘normal lens’ has an
angle of view of about 43 , which corresponds to a focal length equal to the image
diagonal. For the 35 mm film format (image dimensions 2436mm) and a lens with
6.3 Seeing the Landscape 117

Fig. 6.14 The silhouette of all superimposed objects and depth layers with the sky forms the
skyline. Natural features such as vegetation give a skittish and random skyline (A); the skyline of
town and cities is more geometrical and often characteristic (B). Differences between both have
been assessed with the fractal dimension of the edge. (Photo M. Antrop 2005)

Fig. 6.15 The photographic perspective depends on the angle of view (focal length of the lens)
and the distance to the objects

Fig. 6.16 Photographic perspectives: (a) series taken from the same position using lenses with
different focal length; (b) series taken with a fixed focal length, but with varying distance to the
subject, causing more important changes in the proportional sizes between the elements. (Skylines
of Ghent, Belgium, photos M. Antrop 2013)
118 6 Sensing and Experiencing the Landscape

a diagonal of 43 mm, the commonly used normal lens has a focal length of 50 mm,
but values between 40 and 58 mm are considered normal. Digital cameras have
sensors of various sizes and a ‘normal’ lens is defined as an equivalent focal length
(efl) to the 35 mm film format. Focal lengths that are shorter than the normal lens are
called wide-angle lenses, which have a wider angle of view, and long-focus lenses,
such as a telephoto lens, have a narrower angle of view. For zoom lenses focal
lengths vary continuously and perspective distortions accordingly.
Finally, the photographic perspective depends also on the view direction of the
optical axis. The horizon is not always position in the centre of the picture. A
horizon situated higher the photo centre gives a ‘bird’s eye’ perspective, and one
situated lower a ‘frog’ perspective. In these cases, three vanishing points exist.
The aperture of the lens determines the depth-of-field (DOF) and is set by the
photographer. It creates an impression of depth in the image and enhances the
contrast between an object and the background and foreground. The depth of field
in a photograph also differs from what is seen with the naked eye.

6.3.5 Panoramas, Vistas, Eye-Catchers and Landmarks

Vistas are focal views, often corresponding to a sub-scene with a field of view
corresponding to the focus angle of approximately 40 . Sometimes, smaller view-
ing angles are used to define vistas, in particular in garden design and landscape
architecture. The landscape architect René Pechère (1995) used a 22 -rule and gave
many examples of its use in making vistas in garden design. This viewing angle also
corresponds to the angle formed by the span seen at a stretched arm (a span is the
distance measured by a human hand, from the tip of the thumb to the tip of the little
finger) (Fig. 6.17).

Fig. 6.17 The span of a spread hand with a stretched arm was used to measure an angle of
approximately 22 (Photo M. Antrop 2006)
6.4 Conditions of Perception 119

Fig. 6.18 Land art is used to enhance eye-catchers and may create new landmarks as this example
of a painted water tower unique for the municipality of Bierbeek (Belgium) (Photo M. Antrop

Picturesque views, as shown in postcards, often correspond to typical vistas. The

locations from where these can be seen are often marked in the landscape as
(panoramic) viewpoints.
An eye-catcher is an object in the landscape that clearly contrasts in size, shape,
colour or material from the common ones in the surroundings. They are often tall
elements and have a distinct identity. Some are used as landmarks or beacons that
help to orient people in the landscape (Fig. 6.18). In this case, they have a proper
identity (name) and belong to the collective knowledge.

6.4 Conditions of Perception

6.4.1 Standing Where?

The viewpoint also has properties that influence how we observe the landscape. The
position can be dominant in the landscape or can be dominated by the landscape
(Fig. 6.19). Mountain and hilltops are typical examples of dominating viewpoints.
The strategic meaning of dominant viewpoints is well illustrated by the site of many
settlements, showing that visual control over the surrounding territory was impor-
tant. However, high viewpoints are not the only places that offer a wide panoramic
view on the landscape. This is also the case for low lying positions completely
dominated by the surrounding relief, but offering long viewing distances, such as in
valleys and basins.
120 6 Sensing and Experiencing the Landscape

Fig. 6.19 The viewpoint of

the observer can dominate
the landscape or can be
dominated by it. The
vertical viewing angle to the
skyline can describe this

Table 6.2 Movements in the landscape

Movement of the observer
• Fixed position Head, eyes
• Moving position Head, eyes, trajectory
Fixed objects with movement Example: wind turbines
Moving objects
• Single Sun, Moon, a ship, animals, people
• Groups, ‘streaming’ Clouds, water, cars on a motorway

6.4.2 Movement

The scenery changes continuously when the observer moves in the landscape
(Table 6.2). All the scenes are mentally processed to construct a global image of
the landscape, allowing us to see ‘behind the obstacles’ and know what is there.
Gradually we construct a mental map of the area.
Besides the movement of the observer, many other movements animate the
landscape: the streaming of water, cars on a motorway, moving people and animals,
the rotation of wind turbines, the movement of trees and clouds in the wind. They
all attract the attention and often become a (temporary) eye-catcher. The effect
depends on the speed and continuity of the movement, and the viewing distance.
Noise may enhance the effect. Regular and slow motions are experienced as restful.
The speed of moving is an important factor how we perceive the landscape.
When driving for example, the faster we go, the further we focus ahead and the less
we see the landscape around (Table 6.3, Fig. 6.20). Antonson et al. (2014) studied
how different types of landmark objects in the landscape affect drivers. They found
that human perception of a landscape can be explained by Appleton’s evolutionary–
biological theory (Appleton 1975).
6.4 Conditions of Perception 121

Table 6.3 Proportion of the landscape, road and sky, and the focal point of car drivers moving at
different speed
Speed Landscape Road Sky Focus point at
60 km/u 83% 8% 10% 200 m
100 km/u 17% 28% 55% 600 m
After Neuray (1982)

Fig. 6.20 The focal view of

the eye in de direction of
movement narrows with
increasing speed (After
Neuray 1982)

6.4.3 Atmospheric Perspective and Skylight

The atmospheric perspective or aerial perspective results from the experience that
distant objects usually look fuzzy, hazy and bluish. The association between
haziness and the bluish colour allows us to assess the distance of objects. The
scattering of the light passing through the atmosphere causes this effect. Longer
wavelengths (red) are scattered more than shorter ones (blue), resulting in a spectral
shift of the light toward blue. This effect increases with the distance and the purity
of the air, and is affected by water vapour and smoke (Middleton 1968). Skylight
gives a diffuse lighting of all shadowed objects and allows us to see the landscape
when no direct sunlight is available. Skylight is bluish compared to direct sunlight.
Scattered skylight, which is bluish, is added to the reflected colours as a veil. It
causes a decrease of the colour saturation and the contrast between objects and their
122 6 Sensing and Experiencing the Landscape

6.4.4 Time and Lighting Conditions

The combination of the radiation of the sun and the sky gives the natural lighting. It
varies continuously and affects the shadows and colours in the landscape (Lynch
and Livingston 2001).
Colour temperature defines how we see the colours of the objects (Table 6.4). It
is measured in degrees Kelvin and varies from white-blue of a clear blue sky
towards more yellow-red as given by candlelight. The elevation and the direction
of the sun, the time of the day and the atmospheric conditions together define the
colour temperature.
In photography, the ‘golden hour’ (sometimes known as magic hour, especially
in cinematography) is the period around sunset and sunrise when the sun’s altitude
is approximately 10 –12 .
When the sun is near the horizon, sunlight travels a longer distance through the
atmosphere, reducing the intensity of the direct light. More blue light is scattered,
and the sunlight appears more reddish. The illumination of the landscape comes
more from indirect skylight (Lynch and Livington 2001). In addition, the sun’s low
angle with the horizon produces longer shadows. Photographers call this period
shortly after sunrise or before sunset, the ‘golden hour’. Daylight is redder, con-
trasts are softer, shadows are less dark, and highlights are less likely to be
The time between sunrise and 10 am, and between 3 pm and sunset gives a low
angle lighting with low colour temperature which creates on a sunny day long
shadows, high contrast and depth in the landscape scene (Table 6.5).

Table 6.4 Some colour Direct sunlight (‘golden hour’) 4500–2200

temperatures (in Kelvin) Direct sunlight (high sun) 5400–5800
Blue sky 12,000–27,000
Hazy sky 7200–8400
Clouded sky 6700–7000
Average light in shadows 8000
Candlelight 2100

Table 6.5 Sun’s elevation, ratio sun/sky radiation and illumination

Sun’s elevation Ratio solar/sky radiation Total illumination
15 1,8/1 22.000 lux
25 2,6/1 40.000 lux
30 3/1 50.000 lux
45 4,5/1 77.000 lux
60 6/1 97.000 lux
S. Schneider (1972)
6.5 Gestalt-Principles and Perception 123

The psychological impression of atmosphere or ambience we experience in the

landscape is the result of the holistic sensation combining illumination, tempera-
ture, movement, sounds, etc.

6.5 Gestalt-Principles and Perception

6.5.1 Visual

The visual Gestalt-principles were introduced between the two World Wars, first by
the Czech psychologist Wertheimer and later elaborated by the Germans K€ohler
(1929) and Koffka (1935). The ‘law’ of Pr€ agnanz expresses the fundamental
principle of Gestalt-perception. The German concept Pr€ anganz could be translated
as pithiness and refers to what is brief and full of meaning. The principle says that
we organise our experience in a manner that is simple by making it regular, orderly
and symmetric. This is achieved in different ways and Gestalt psychologists
formulated different ‘Gestalt-laws’, which are rather rules or principles describing
how we treat complex patterns or scenes we perceive. However, it should be noted
that these ‘Gestalt-laws’ do not explain what is happening and why. This is the basic
criticism of cognitive psychology and neuroscience on the Gestalt-theory of per-
ception, which is considered redundant.
Gestalt-principles help understand the interpretation and meaning we give to
complex patterns and eventually why confusion may occur. We perceive patterns as
a Gestalt, as a whole, which we try to structure and to organise into meaningful
entities using Gestalt-principles. These can easily be illustrated in a series of
figures, often used as eye-tricks and some of the mechanisms involved can be
revealed by optical illusions (Figs. 6.21, 6.22 and 6.23). Figure 6.21 shows:
(1) A Gestalt: the pattern of dark patches shows a horse rider. Each of the patches
is meaningless. Together and with the ‘empty’ space between, they make a
meaningful figure. The empty space is as essential information as the patches.
We also construct the landscape Gestalt and its character from fragments.
(2) Gestalt-principle of closure: we tend to enclose apparent shapes to form fictive
spaces. In garden design, this is used to create separate gardens and parts
(3) Gestalt-principle of context and semiotics: characters are signs that transmit
information. The middle sign is not a normal letter and receives its meaning
from the context it is placed in. The context makes that the meaning of the
word is transmitted correctly even with distorted signs.
(4) Gestalt-principle of reversible perspective or Gestalt-switch: we try to find
meanings in the patterns we observe. Therefore, all possible combinations are
made between the parts until we recognise a meaningful figure. As a conse-
quence, we see the transparent cube ‘from the front’ and ‘from the back’. The
124 6 Sensing and Experiencing the Landscape

Fig. 6.21 Gestalt-

principles illustrated by
optical illusions
(explanations see text)

Fig. 6.22 Tall vertical objects are eye-catchers in the landscape. However, their height is difficult
to estimate (Photo M. Antrop 2003)
6.5 Gestalt-Principles and Perception 125

Fig. 6.23 The sky of a thunderstorm: simultaneous contrast enhances the differences between
intensity and saturation of the colours (Photo M. Antrop 2005)

Gestalt-principles of symmetry and a ‘good geometric Gestalt’ make that we

interpret the left figure easier as a hexagon than a cube.
(5) Gestalt-principles of similarity and proximity. Similar elements are grouped
into new entities. The closer they are, the stronger the grouping is. Elements
that differ in shape, size, texture or colour are seen as disturbing elements or
(6) Gestalt-switch: when there is ambiguity between foreground and background,
one pattern can be seen alternately as different figures. Ambiguous figures
offer multiple meaningful interpretations. Reversing what is presumed to be
foreground, allows switching between the different solutions. Often the initial
choice of foreground and background is based upon past experience and thus
relies on the subjective knowledge of the observer.
(7) Gestalt-principles of closure, context and simultaneous contrast acting
together: we see black circles with a missing sector, and the sectors suggest
a (transparent) triangle lying on the circles. The grey colour of the triangle
seems lighter the background, although there is no difference.
(8) Gestalt-principle of context: the context influences the perception of size. The
central circle in both figures has the same size (circles of Tichener).
(9) Influence of context on directions (illusion of Hering): the two horizontal lines
are straight and parallel. The crossing radial lines distort reality so we them as
curved lines. The radial lines correspond to our daily experience of
126 6 Sensing and Experiencing the Landscape

perspective formed by converging lines. We see them as ‘background’ com-

pared to the thicker lines on the ‘foreground’.
(10) Experiment of Müller-Lyer: the dominance of the vertical dimension. The
vertical lines have the same length as the horizontal ones. Nevertheless, we see
them longer. The vertical dimension is the only one that remains constant
when we move through space. Verticals allow orientation and thus we see
them as more important.

6.5.2 Soundscapes

Gran€ o (1929) used the settings of his summer house at Valosaari to explore the
description and mapping all the sensory characteristics in the proximity such as
openness and obstructiveness of the view, colours and variation of the colours
through the seasons, effects the temperature and the wind, auditory and olfactory
phenomena, light conditions and mobile elements (Fig. 6.24).
The sound is the second most important sense in experiencing the landscape. The
acoustic environment is mainly studied in the context of environmental impact
assessment. The acoustic ecology tries to define soundscapes, designate tranquillity
areas and assess the noise disturbances and their effects. Tranquillity areas are often
defined as areas with an extent of for example at least one square kilometre, where
sounds made by human activities do not disturb natural sounds made by the wind,
water and wildlife.
Raymond Murray Schafer introduced the concept soundscape to describe the
characteristic combination of sounds that arises from an immersive environment. It
is a mixture of all kinds of natural sounds (wind, water, animals, spaces) and sounds
created by humans. He also used the term soundmark, in an analogy of landmark
that is a sound, which is unique to an area. In his book The Tuning of the World, he
wrote “Once a Soundmark has been identified, it deserves to be protected, for
soundmarks make the acoustic life of a community unique” (Schafer 1993). The
study of soundscape is the subject of acoustic ecology.
The sound is a sequence of waves of pressure that propagates through compress-
ible media such as air or water at a specific speed and is characterised by its
frequency. Frequencies capable of being heard are species and age dependent.
For humans, the range is from 20 Hz to 20,000 Hz and is called audio or sonic.
Our best hearing is at a frequency of 100 Hz. (Gelfand 2004).
The human hearing has a logarithmic response to the sound amplitude, as the eye
has to the illumination, and thus detects sounds with a wide range of amplitudes.
The sound pressure is often measured as a level on a logarithmic decibel scale (dB).
The subjective experience of doubling the sound strength corresponds to an
increase of the sound level by 10 dB, i.e. ten times higher sound level. Sound levels
of 90 dB and higher can cause hearing damage. However, sounds consist of a
6.5 Gestalt-Principles and Perception 127

Fig. 6.24 Gran€ o’s map of the ‘auditory phenomena’ near his summer house in Valosaari. (1)
sounds and noises produced by people always in summer, (2) produced by people sometimes in
summer, (3) produced by people frequently at all times of the year (boating route, ice road), (4)
birdsong in spring and summer, (5) clanging of cow bells in summer. The hatched area for each
auditory phenomenon terminates at the 25 m phenomenal curve (Gran€ o and Paasi 1997)

mixture of many frequencies at different strengths. Therefore, the sound level is

measured as dB(A), the average sound level weighted by frequency, which corre-
sponds better to the subjective sound experience. A difference of 1 dB(A) is just
Noise disturbance is a subjective interpretation depending on the sound level and
the nature of the sound, i.e. the psychological associations we make.
The auditory system is complementary to the visual system. The information
from both can be contradictory, for example when standing in a tranquil looking
countryside with loud noise disturbance. This illustrates how a visual scene creates
expectations for the other sensations. The same happens with olfactory sensation.
Farina (2014) defines soundscape as “an acoustical composition that results from
the voluntary or involuntary overlap of different sounds of physical or biological
origin”. The soundscape is an aspect of (landscape) ecology. He distinguishes three
types of sounds in the landscape: geophonies (produced by wind, running water,
waves, weather, earthquakes), biophonies (produced by non-human living organ-
isms) and anthrophonies (produced by humans). A specific mixture of the three
128 6 Sensing and Experiencing the Landscape

types characterises landscape types. He introduced the concept of a sonotope,

similar to ecotope, as a distinct spatial sonic unit, similar to a patch in landscape
ecology, and the result of overlapping geophonies, biophonies and anthrophonies.

6.6 Experiencing the Landscape

6.6.1 Landscape Experience Assessment

The same landscape can mean different things to different people, as if they use
different perceptual lenses (Howard 2013). Landscape experience is the result of
the interaction between landscape properties and physiological and psychological
and cultural factors related to the observer. Tuan (1974) named the “the affective
bond between people and place or setting” topophilia. Jacobs (2006) speaks of the
production of mindscapes and of the psychology of the visual landscape. He
considers environmental psychology the leading discipline in the study of human
responses to the visual landscape, but acknowledges that many other disciplines are
involved as well. However, all approaches share three core assumptions: (1) the
way people perceive landscapes is influenced but not determined by physical
landscape attributes, (2) a complex mental process of information reception and
processing mediates between the physical landscape and the psychological land-
scape, and (3) various biological, cultural and individual factors can exercise
influence on this mental process. L€orzing (2001) sees the construction of
mindscapes as successive stages in understanding and interaction with the land-
scape, each affecting our behaviour differently.
The way the landscape is perceived, experienced and appreciated – the psychol-
ogy of landscape as Jacobs (2011) called it – is studied by several disciplines and a
wide variety of methods is used. Important is that the psychological responses to
landscape prove to be partially innate and convergent. However, the explanations
vary a lot and the complexity of the subject is reflected by theories.
An important factor is also that the research design and method influence the
outcome of the experiments (Sevenant and Antrop 2010a, b). The material tangible
landscape interacts in a complex manner with the judgment context of the commu-
nity of observers and defines values, preferences, characterization and behaviour
(Fig. 6.25).

6.6.2 Theories

The holistic experience of the landscape results in an assessment of preference and

value, which is based on rational and affective criteria, both consciously and
unconsciously. Many landscape qualities are involved and often the expression of
6.6 Experiencing the Landscape 129

Fig. 6.25 The complex layout of landscape experience assessment (Sevenant 2010)

value and preference is associated with the aesthetic quality of the landscape. In
landscape protection, conservation, planning and environmental impact assessment
there has been an increasing demand in systematically describing these qualities,
formulating (quantitative) indicators to be used in scenario modelling and policy
assessment. Also, scenic qualities and aesthetics are the core business in (land-
scape) architecture and there is a growing need in understanding how preferences
are related to landscape characteristics. All this stimulated since the 1970s the
development of scientific research in describing and mapping scenic qualities and
typologies of the visual landscape and landscape preference. The following funda-
mental question arises: is the aesthetic quality inherent (or intrinsic) to the land-
scape or is it in the ‘eye of the beholder’, thus a mental or social construct? (Lothian
1999). Two research paradigms developed from this. The objectivist one aims to
identify physical landscape properties that can be related to certain preferences,
e.g. aesthetically pleasing, or disturbing, etc. The subjectivist paradigm focuses on
the psychological and sociological response when observing the landscape. The
first one is a landscape centred approach, the second one focuses on the observer
(Sevenant 2010).
Since the 1970s the main discourse in landscape preference research focuses on
nature versus nurture. One group of theories starts from a biological and evolution-
ary bias to explain our preferences regarding the landscape. Another group
130 6 Sensing and Experiencing the Landscape

emphasizes the importance of culture, including education and media. Finally, a

combined approach emerged as well.
The arousal theory of Berlyne (1971) is an early example of the evolutionary
approach used in environmental psychology. The hypothesis is that an optimal
number of sensory stimuli from the environment provoke a positive preference.
When the level of arousal is too low, the landscape is experienced as dull, while a
too high-level of arousal causes a chaotic and uncertain experience.
Appleton (1975) introduced the prospect-refuge theory to explain landscape
preference. The aesthetic satisfaction people experience depends on the capability
of the landscape to fulfil their biological needs. Two factors define the preference of
a landscape: prospect or the possibility to oversee the landscape, and refuge or the
potential to hide and be unseen. The whole concept is based on a lifestyle of
prehistoric hunters in savanna-like landscapes, hence the theory is also known as
the savanna theory and was used to explain why some studies indeed indicated this
type of landscape as most preferred. However, the empirical basis for the theory
remains poor.
For early humans who depended on hunting and gathering, spatial understanding
of their habitat and the ability to explore new areas was probably highly important
(Tveit et al. 2012).
Stephen and Rachel Kaplan (Kaplan and Kaplan 1995) used this concept to
develop their theory of landscape preference based on processing visual informa-
tion, which facilitates understanding and exploration of the environment. Fast,
unconscious processing of that information has been very important in survival
and would have been favoured by natural selection. They suggested that people’s
perception combines an immediate and unconscious dimension (a ‘two-dimen-
sional space’) and a slower cognitive inference with a future prospect (a ‘three-
dimensional space’).
Their empirical research showed that humans classify landscapes based on two
criteria: content and spatial configuration. Content opposes for example wet and dry
land, and spatial configuration open and enclosed landscapes. The knowledge
obtained to understand and explore these landscape categories depends on four
informational factors: coherence, legibility, complexity and mystery. These are
summarized in the preference matrix (Table 6.6).
Coherence is the degree different elements relate to each other and facilitates
insight and creates order. Complexity refers to the number of different types of the
visual elements in the scene. A certain amount of complexity is necessary to keep
interest in the landscape. Legibility refers to the ease to ‘read’, understand and

Table 6.6 The preference matrix for landscape preference

Understanding making sense of Exploring involvement
2D: immediate Coherence Complexity
Direct information
3D: Future of promised Legibility Mystery
Indirect information
Kaplan et al. (1989)
6.6 Experiencing the Landscape 131

Fig. 6.26 Ulrich’s model

for landscape preference
(after Ulrich 1983)

remember the place. Mystery refers to the potential of hidden information in the
landscape that still can be discovered.
Van der Jagt et al. (2014) tested the validity of the preference matrix and found
that each of the categories in the preference matrix is predictive of scene aesthetics
and that coherence and complexity interact in predicting scenic quality. They also
found that natural character was a positive predictor of scene attractiveness, while
built character and low levels of familiarity predicted scenic quality negatively.
Ulrich (1983) combined evolutionary and cultural aspects in his psycho-evolu-
tionary model of landscape preference. He distinguishes between a fast, affective
response and a slower cognitive response when perceiving landscapes (Fig. 6.26).
The fast affective response is mainly unconscious and primary evolutionary deter-
mined. It is triggered by landscape characteristics (Ulrich calls them preferenda),
which are abstract and holistic, such as complexity and coherence. The affective
response results in feelings ranging between safe and unsafe, and like and dislike.
This part of Ulrich’s theory is very similar to the ones of Berlyne and Appleton. The
132 6 Sensing and Experiencing the Landscape

cognitive response is slower and mainly determined by cultural factors. The

landscape is comprehended only after the identification of various elements,
which are recomposed and structured. The structuring follows an assessment
regarding vital needs and utility, ranging from food supply, safety and accessibility.
Finally, the affective and cognitive responses are combined. This explains why the
same landscape gives a similar affective experience, while the cognitive assessment
can be different.
In his summary of research in aesthetic, emotional and psychological responses
to visual natural and urban landscapes, Ulrich (1986) lists following characteristics
for liking unspectacular natural scenes:
(1) complexity, or the number of elements in the scene is moderate to high;
(2) the complexity is structured and order or patterning is present;
(3) there is a clearly defined moderate to high level of depth;
(4) the ground surface is relatively smooth and is judged to be favourable to
(5) a deflecting or curving sightline is present, conveying a sense that new land-
scape information lies immediately beyond the observer’s bounds;
(6) judged threat is negligible or absent.
Nohl (2001) sees the human perception as a process of gaining information or
‘sensory cognition’, which explains how aesthetic joy is experienced. The land-
scape is then seen as an aesthetic object characterised by its appearance and the
meaning it has. Following the aesthetical theory of the German philosopher
Baumgarten (1750–1758) and the findings by Kaplan and Kaplan (1995), Nohl’s
theory of the ‘aesthetic landscape’, i.e. the aesthetically perceived landscape, allows
better defining aesthetic categories, such as the beautiful and the sublime, and to
formulate a new paradigm for sustainable landscape aesthetics. Nohl recognises
four different levels of aesthetic cognition: the perceptual, the expressive, the
symptomatic and the symbolic level. The ‘aesthetic pleasure’ is larger and changing
the more information is extracted from these aesthetic cognitive levels. This means
that aesthetic values are not stable but vary in time and with new experiences. New
aesthetic categories emerge when landscapes change. Based on the main contem-
porary landscape types, Nohl defines four sustainable aesthetic landscape catego-
ries: the beautiful, the (new) sublime, the interesting and the plain. These categories
make the prototypes of the future aesthetical landscapes: the ‘traditional cultural
landscape’, the ‘spontaneous landscape’, the ‘urban-industrial landscape’ and the
‘rural functional landscape’.
The beautiful is a landscape where the observer more or less knows all elements
and perceived in one glance as arranged in a balanced and harmonic way. This
category is typical for the traditional cultural landscapes and thus of the past but still
significant for the future. The beautiful landscape is rich of symbolic meanings,
‘makes sense’ and has a story and history.
The traditional aesthetic category of ‘the sublime’ referred to landscapes of the
grandiose nature, wild and mysterious, huge and wide, terrifying and fascinating
simultaneously, unconquered and overwhelming humans. They symbolise the vital
6.6 Experiencing the Landscape 133

power of life, thus sustainability. Today, humans have the technological power to
create such landscapes as well, often by destruction or exploitation, or by building.
Nohl refers to this category as the (new) sublime.
The interesting consists of very heterogeneous landscapes created by apparently
chaotic processes as can be found on the urban fringe, industrial derelict land and
large construction sites. They are fascinating in their ugliness and enormity.
The plain is the aesthetical category found in vast landscapes of intensive
agricultural production. They are experienced purely functional, monotonous and
The contemporary landscapes loose following aesthetic qualities: variety, natu-
ralness, (rural) structuring, regional identity and vista quality. Nohl describes the
consequences these deficits on the aesthetic as coarsening, impoverishment,
destabilisation, and alienation of the human perceptual field, i.e. the beholder’s
field of vision.

6.6.3 Research Models and Paradigms

Research in landscape experience and preference can be subdivided in several

ways, each approach or model characterised by a specific paradigm and using
different methods (Sevenant 2010). Expert Models

Expert models consider landscape as an object, having an inherent aesthetic quality

which can be determined by any competent inspection by a trained expert (Daniel
2001). The approach is similar to the evaluation of art-pieces and monuments and
often used by landscape architects. Daniel and Vining (1983) further distinguish
between formal aesthetic and ecological models. A well-known formal aesthetic
model is the Visual Management System (USDA 1974), which assumes a direct
relation between scenic quality and landscape diversity. An early and well-known
ecological model is the Leopold’s matrix (Leopold 1969). Although the approach
aims to be objective, the lack of reliability and validity has been a source of
fundamental criticism (Palmer and Hoffman 2001) when defining a uniqueness
ratio based on ecological measures of the landscape. Public Preference Models

Public preference models are perception-based and use methods of survey-research.

Different approaches can be recognised (Sevenant 2010).
Psychophysical models aim to find mathematical relationships between physical
features in the landscape and the perceptual judgement or behaviour of the
134 6 Sensing and Experiencing the Landscape

observers. The focus is on the landscape qualities, which are considered objectively
measurable, and on the formulation of landscape indicators, preferentially ones
useful in policy and allowing making predictions of public’s preferences. For
example, Hägerhäll et al. (2004) used the fractal dimension of landscape silhouettes
to find relations with landscape preference, and Rogge et al. (2007) simulated the
openness on landscape photographs to predict landscape preferences.
Based on a review of the literature of landscape preference studies, Tveit et al.
(2006) identified nine key visual concepts and created a theoretical framework for
the analysis of the visual landscape. Using this framework, Ode et al. (2008)
derived indicators related to landscape preference and experience, which they
used to capture the visual character of the landscape related to aesthetics. The
concepts used were complexity, coherence, disturbance, stewardship, imageability,
visual scale, naturalness, historicity and ephemera. The selection of the indicators
happened stepwise using filters considering criteria such as data availability, trans-
ferability and potential to quantify. The visual concepts are interrelated and demand
careful interpretation, as landscape changes altering indicator values related to one
concept may cause an increase or decrease in indicator values of another concept.
Fry et al. (2009) explored the conceptual common ground of visual and ecological
landscape indicators and identified a set of indicators that capture important aspects
of both ecological and visual quality.
Cognitive or psychological models are more concerned with the observer and
aim to assess the cognitive and affective reactions evoked by the landscape. In
general, high-quality landscapes are the ones that evoke positive feelings, such as
security, happiness, etc., while low-quality landscapes evoke negative feelings,
such as stress, fear, etc. (Sevenant 2010).
Most literature in landscape perception and preference has focused on aesthetic
preference by people. Factors that influence the aesthetic preference include land-
scape dimensions and observer’s background characteristics. The latter are far less
examined, partially due to the increasing complexity of the dataset and the complex
multivariate analysis models that require more sophisticated statistical analysis.
Latent class analysis assigns probabilistically respondents in latent preference
classes in function of the respondent’s characteristics and background. This allows
including fuzziness in the model (Fig. 6.27). The latent classes can be related to
ratings of an aesthetic landscape quality, derived from a set of landscape charac-
teristics used as indicators.
These methods use questionnaires to collect observer’s response to the land-
scapes presented. This can be done with focus groups in the field, but more often
using stimuli as (manipulated) photographs or computer simulations. More recently
also eye-tracking is used in the analysis (Dupont et al. 2014).
Phenomenological models focus on the individual experiences and the aim is to
assess the relationship between person and landscape. Personal in-depth interviews
and questionnaires are used, but also sketches and cognitive mapping. A well-
known example is the pioneering work by Kevin Lynch (1960) ‘The Image of the
City’, introducing the concept of imageability.
6.7 Experts and Laypeople Experience the Landscape Differently 135

Fig. 6.27 Principle of latent classes analysis in landscape preference assessment. Indicators of
landscape characteristics are used as predictors of a given landscape quality. Characteristics of the
observers define probabilistically latent classes of preference. The relationship between the
landscape quality and the latent classes depend also on the research method used (after Sevenant
and Antrop 2010b)

6.7 Experts and Laypeople Experience the Landscape


Since the European Landscape Convention, the participation of the public should be
integrated in landscape management and policy, and their aspirations towards
landscape should be assessed. This means transdisciplinarity and implies a shift
from a top-down approach by experts to a bottom-up approach by laymen, insiders
and outsiders. It raises the question how well the expert judgement fits the aspira-
tions of the public (Sevenant and Antrop 2010a).
Tveit (2009) analysed how future landscape professionals (students) and lay
people differ in terms of perceiving and appreciating the visual scale in the
landscape. Correspondence between both could enable experts to encompass public
opinion in landscape planning and management. Visual scale indicators proved to
be good predictors of preference for the student group and confirmed the landscape
aesthetic theory, but this was not the case to predict public preferences.
Sevenant and Antrop (2010a) explored ways to involve local people and to grasp
their aspirations with regard to the landscape features of their surroundings. They
engaged multicultural inhabitants of the city of Ghent (Belgium) since urbanites
136 6 Sensing and Experiencing the Landscape

became important new users of the countryside and thinking and planning of the
rural landscape tends to happen with the urban needs in mind. It appeared that much
more factors played a role than only landscape. It was confirmed that ‘the urbanite’
does not exist and that many people actually living in the city do not necessarily
have an urban background as several of them lived in the countryside during some
period of their life. Also, it appeared that the language used in the survey, proved to
be too academic and abstract and that appropriate vocabulary to communicate
properly was often lacking.
Dupont et al. (2015) used eye tracking to analyse the difference in viewing
pattern among landscape experts and lay people. Acquired educational or profes-
sional expertise with respect to landscapes seemed to enhance efficient information
extraction in terms of an improved interpretation, identification and understanding
of landscape objects. The landscape experts seem to observe landscape photographs
in a holistic fashion, consisting of a global scanning of the image alternated with
more detailed inspections of particular components. In contrast, non-experts spend
considerably more time and attention to specific objects, in particular to buildings,
restricting their visual exploration of the landscape.
Makhzoumi (2009) gave an interesting case study of the landscape experiences
of the inhabitants of the rural village of Ebel-es-Saqi (Lebanon). The case is not
only interesting because it focuses on the discourses of the insiders but also deals
with a culture with a very different view on the concept of landscape. Indeed, the
term ‘landscape’ is absent in Arabic, but the findings confirmed that nevertheless, a
spatially explicit and linguistically layered conception of the village landscape was
present. The village ‘landscape’ acted as a medium for social and cultural actions,
rather than just a scenic background. Place names and generic location references to
landscape resources was common knowledge. Even without the word ‘landscape’,
landscape features constituted cultural content.


Antonson, H., Ahlstr€om, C., Mårdh, S., Blomqvist, G., & Wiklund, M. (2014). Landscape heritage
objects’ effect on driving: a combined driving simulator and questionnaire study. Accident;
Analysis and Prevention, 2014(62), 168–177.
Antrop, M. (2007). Perspectieven op het landschap (322 pp). Gent: Academia Press.
Antrop, M. (2012). Intrinsic values of landscapes. In T. Papayannis & P. Howard (Eds.),
Reclaiming the Greek landscape (pp. 31–42). Athens: Me-INA.
Appleton, J. (1975). The exprience of landscape (p. 293). London: Wiley.

Baumgarten, A. G. (1750–1758). Ästhetik. Ubersetzt von Mirbach, D., 2007 (H. R. Schweizer,
Ed.). Hamburg: Meiner Felix Verlag.
Berlyne, D. E. (1971). Aesthetics and psychology. New York: Appleton-Century-Crofts.
Bourassa, S. C. (1990). A paradigm for landscape aesthetics. Environment and Behavior, 22(6),
Bourassa, S. C. (1991). The aesthetics of landscape. London/New York: Belhaven Press.
van Campen, C. (1993). Beeldillusies. Gezichtsbedrog in afbeeldingen sinds de renaissance.
Canteleer, De Bilt, 160 pp., ISBN 90 213 2186 6.
References 137

Cosgrove, D. (2002). Chapter 12: Landscape and the European sense of sight – Eyeing nature. In
K. Anderson, M. Domosh, S. Pile, & N. Thrift (Eds.), Handbook of cultural geography
(pp. 249–268). London: SAGE Publications.
Daniel, T. C. (2001). Whither scenic beauty? Visual landscape quality assessment in the 21st
century. Landscape and Urban Planning, 54, 267–281.
Daniel, T. C., & Vining, J. (1983). Methodological issues in the assessment of landscape quality.
In I. Altman & J. Wohlwill (Eds.), Human behaviour and the natural environment (Vol. VI,
pp. 39–84). New York: Plenum Press.
Dramstad, W. E., Tveit, M. S., Fjellstad, W. J., & Fry, G. L. A. (2006). Relationships between
visual landscape preferences and map-based indicators of landscape structure. Landscape and
Urban Planning, 78(4), 465–474.
Dupont, L., Antrop, M., & Van Eetvelde, V. (2014). Eye-tracking analysis in landscape perception
research: Influence of photograph properties and landscape characteristics. Landscape
Research, 39(4), 417–432.
Dupont, L., Antrop, M., & Van Eetvelde, V. (2015). Does landscape related expertise influence the
visual perception of landscape photographs? Implications for participatory landscape planning
and management. Landscape and Urban Planning, 141, 68–77.
Farina, A. (2014). Soundscape ecology. principles, patterns, methods and applications (p. 315).
Dordrecht: Springer.
Fry, G., Skar, B., Jerpåsen, G., Bakkestuen, V., & Erikstad, L. (2004). Locating archaeological
sites in the landscape: A hierarchical approach based upon landscape indicators. Landscape
and Urban Planning, 67(1–4), 97–108.
Fry, G., Tveit, M. S., Ode, Å., & Velarde, M. D. (2009). The ecology of visual landscapes:
Exploring the conceptual common ground of visual and ecological landscape indicators.
Ecological Indicators, 9(5), 933–947.
Gelfand, S. A. (2004). Hearing an introduction to psychological and physiological acoustics (4th
ed.). New York: Marcel Dekker.
Gibson, J. J. (1950). The perception of the visual world. Westport: Greenwood Press.
Gobster, P. H., Nassauer, J. I., Daniel, T. C., & Fry, G. (2007). The shared landscape: What do
aesthetics have to do with ecology? Landscape Ecology, 22, 959–972.
Gran€o, J. G. (1929). Reine Geographie. Eine methodologische Studie beleuchtet mit Beispielen
aus Finnland und Estland. Helsinki, Helsingfors, Acta Geographica, 2(2), 202 pp.
Gran€o, O., & Paasi, A. (Eds.). (1997). Pure geography. J.G. Gran€ o., (M. Hicks, Trans.).
Baltimore/London: The John Hopkins University Press, 191 pp.
Hägerhäll, C. M., Purcell, T., & Taylor, R. (2004). Fractal dimension of landscape silhouette
outlines as a predictor of landscape preference. Journal of Environmental Psychology,
24(2004), 247–255.
Howard, P. (2013). Perceptual lenses. In P. Howard, I. Thompson, & E. Waterton (Eds.), The
Routledge compagnion to landscape studies (pp. 43–53). London: Routledge.
Jacobs, M. (2006). The production of mindscapes. A comprehensive theory of landscape experi-
ence. PhD Dissertation, Wageningen University, pp. 280.
Jacobs, M. (2011). Psychology of the visual landscape. In S. Nijhuis, R. van Lammeren, & F. van
der Hoeven (Eds.), Exploring the visual landscape, Research in urbanism series, Amsterdam
(Vol. 2, pp. 41–54). IOS Press.
Kaplan, R., & Kaplan, S. (1995). The experience of nature. A psychological perspective (340 pp).
Ann Arbor: Ulrich’s Bookstore.
Kaplan, R., Kaplan, S., & Brown, T. (1989). Environmental preference: A comparison of four
domains of predictors. Sage Publications, Environment and Behavior, 21(5), 509–530.
Koffka, K. (1935). Principles of Gestalt psychology. New York: Harcourt/Brace/World.
K€ohler, W. (1929). Gestalt Psychology. New York: London.
Leopold, L. B. (1969). Landscape esthetics. Natural History, 78, 36–45.
L€orzing, H. (2001). The nature of landscape. A personal quest (176 pp). Rotterdam:
010 Publishers.
138 6 Sensing and Experiencing the Landscape

Lothian, A. (1999). Landscape and the philosophy of aesthetics: Is landscape quality inherent in
the landscape or in the eye of the beholder? Landscape and Urban Planning, 44,
177, 30 pages–198.
Lynch, K. (1960). The image of the city (194 pp). Cambridge, MA/London: The MIT Press.
Lynch, D. K., & Livingston, W. (2001). Color and light in nature (2nd ed.p. 277). Cambridge:
Cambridge University Press. ISBN 0521775043, 9780521775045.
Makhzoumi, J. M. (2009). Unfolding landscape in a Lebanese village: Rural heritage in a
globalising world. International Journal of Heritage Studies, 15(4), 317–337.
McConnell, J. V. (1989). Understanding human behavior. New York: Holt, Rinehart and Winston
Inc. Intern. Edition, 645 pp.
Meienberg, P. (1966). Die Landnuztungskartierung nach Pan-, Infrarot und Farb-luftbildern.
Regensburg: Verlag M. Lassleben Kallmunz, 133 pp.
Middleton, W. E. K. (1968). Vision through the atmosphere (250 pp). Toronto: University of
Toronto Press.
Neuray, G. (1982). Des Paysages. In Pour qui? Pourquoi? Comment? Gembloux: Presses
Agronomiques de.
Nijhuis, S. (2011). Visual research in landscape architecture. In S. Nijhuis, R. van Lammeren, &
F. van der Hoeven (Eds.), Exploring the visual landscape, Resarch in urbanism series (Vol.
vol. 2, pp. 103–145). Amsterdam: IOS Press.
Nijhuis, S. (2015). GIS-based landscape design research. Stourhead landscape garden as a case
study. Delft University of Technology Architecture and the Built environment, 335 pp.
Nijhuis, S., van Lammeren, R., & van der Hoeven, F. (2011). Exploring the visual landscape,
Resarch in urbanism series (Vol. vol. 2, p. 334). Amsterdam: IOS Press.
Nohl, W. (2001). Sustainable landscape use and aesthetic perception – Preliminary reflections on
future landscape aesthetics. Landscape and Urban Planning, 54, 223–237.
Ode, Å. (2005). Analysing and communicating landscape change – the Visulands toolkit. Conf.
Proc. “Landscape Changes”, Ankara, 198–207.
Ode, Å., Tveit, M. S., & Fry, G. (2008). Capturing landscape visual character using indicators:
Touching base with landscape aesthetic theory. Landscape Research, 33(1), 89–117.
Ode, Å., Fry, G., Tveit, M. S., Messager, P., et al. (2009). Indicators of perceived naturalness as
drivers of landscape preference. Journal of Environmental Management, 90(1), 375–383.
Palmer, J. F., & Hoffman, R. E. (2001). Rating reliability and representation validity in scenic
landscape assessments. Landscape and Urban Planning, 54, 149–161.
Panero, J., & Zelnik, M. (1979). Human dimension and interior spece. New York: Whitney
Library of Design.
Pechère, R. (1995). Grammaire des Jardins. Secrets de métier (137pp). Bruxelles: Editions
Rogge, E., Nevens, F., & Gulinck, H. (2007). Perception of rural landscapes in Flanders: Looking
beyond aesthetics. Landscape and Urban Planning, 82, 159–174.
Schafer, R. M. (1977). The tuning of the world. Toronto: Random House Inc.
Schafer, R. M. (1993). The soundscape: Our sonic environment and the tuning of the world
(320 pp). Rochester: Destiny Books.
Schneider, S. H. (1972). Cloudiness as a global climatic feedback mechanism: The effects on the
radiation balance and surface temperature of variations in cloudiness. Journal of the Atmo-
spheric Sciences, 29, 1413–1422.
Sevenant, M. (2010). Variation in landscape perception and preference. Experiences from case
studies in rural and urban landscapes observed by different groups of respondents. Doctoral
dissertation. Ghent University, Department of Geography, Ghent. p. 480.
Sevenant, M., & Antrop, M. (2009). Cognitive attributes and aesthetic preferences in assessment
and differentiation of landscapes. Journal of Environmental Management, 90, 2889–2899.
Sevenant, M., & Antrop, M. (2010a). Transdisciplinary landscape planning: Does the public have
aspirations? Experiences from a case study in Ghent (Flanders, Belgium). Land Use Policy, 27,
References 139

Sevenant, M., & Antrop, M. (2010b). The use of latent classes to identify individual differences in
the importance of landscape dimensions for aesthetic preference. Land Use Policy, 27,
Tuan, Y.-F. (1974). Topophilia: A study of environmental perception, attitudes and values.
Englewood Cliffs: Prentice Hall.
Tveit, M. S. (2009). Indicators of visual scale as predictors of landscape preference; a comparison
between groups. Journal of Environmental Management, 90, 2882–2888.
Tveit, M. S., Ode, Å., & And Fry, G. (2006). Key concepts in a framework for analysing visual
landscape character. Landscape Research, 31(3), 229–225.
Tveit, M. S., Ode-Sang, Å., & Hägerhall, C. M. (2012). Chapter 4: Scenic beauty: Visual landscape
assessment and human landscape perception. In L. Steg (Ed.), Environmental psychology: An
introduction (pp. 38–46). Chichester: British Psychological Society/Wiley.
Ulrich, R. S. (1983). Aesthetic and affective response to natural environment. In I. Altman & J. F.
Wohlwill (Eds.), Behavior and the natural environment (pp. 85–125). New York: Plenum
Ulrich, R. S. (1986). Human responses to vegetation and landscapes. Landscape and Urban
Planning, 13, 29–44.
USDA – US Department of Agriculture Forest Service. (1974). The visual management system;
agriculture handbook 462. In National forest landscape management (Vol. Vol. 2).
Washington, DC: US Government Printing Office.
Van der Ham, R., Schut, G., & Iding, J. (1970). Een voorstel voor een nieuwe landschapstypologie
naar visuele kenmerken. Stedebouw en Volkshuisvesting, 51, 421–438.
Van der Jagt, A. P. N., Craig, T., Anable, J., Brewer, M. J., & Pearson, D. G. (2014). Unearthing
the picturesque: The validity of the preference matrix as a measure of landscape aesthetics.
Landscape and Urban Planning, 124, 1–13.
Wrightson, K., 2000. An introduction to acoustic ecology. Soundscape: The Journal of Acoustic
Ecology: 10–13.
Chapter 7
Landscape Dynamics and Evolution

Abstract Landscapes are dynamic and change continuously. The frequency, speed
and magnitude of these changes vary from place to place. Some parts of the
landscape may stay stable for long periods and witness from past situations. The
landscape is a patchwork of time layers, like a palimpsest. Landscapes evolved over
thousands of years and are marked by short periods of profound, revolutionary
changes. The most important break with the past occurred during the Age of
Revolution starting in the eighteenth century in western Europe. The break marks
the transition from traditional landscapes to modern new ones, characterised by a
profound change in attitude towards the environment, which is reflected in the
landscape. The main driving forces of the changes are demography, economy,
politics and natural calamities. Each of them drives characteristic processes: supply
in natural resources, movement, urbanisation and industrialisation. These drivers
act at different scales: in the past local and regional scales of action dominated,
today the global scale has become so important that it is considered as a new driving
force: globalisation, including climatic change. Special focus is given on the
diffusion process of urbanisation, shaping most of the contemporary landscapes
and where the majority of the population lives. Landscape biographies describe the
evolution and history of local and regional landscapes. Time depth and landscape
paths or trajectories are new concepts in the study of landscape dynamics.

Keywords Landscape evolution • History • Palimpsest • Change • Processes •

Drivers • Urbanisation • Time depth • Landscape path

7.1 Introduction

In this chapter, we look at the different dynamical aspects of the landscape. Firstly,
the landscape genesis and history are discussed as well as concepts and principles of
landscape changes. Next, the contemporary processes of change and their driving
forces are studied in a network society. In particular, attention is given to processes
of urbanisation, which affect the majority of the population. Lastly, concepts and
approaches to study landscape dynamics and history are discussed considering
concepts as landscape paths, trajectories and landscape biography as well as
methodological aspects of surveying and landscape monitoring.

© Springer Science+Business Media B.V. 2017 141

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_7
142 7 Landscape Dynamics and Evolution

7.2 The Landscape is Dynamic

The linguistic meaning of the word ‘landscape’ suggests something static. Most
representations of landscapes in drawings, paintings, maps and photographs are
static. Nevertheless, landscape changes and transformations catch most of our
attention today. The ‘crisis of the landscape’, as suggested by the Dutch philosopher
Ton Lemaire (2002), conveys to the magnitude and speed of the changes of
landscapes. However, history learns that landscape change is not something recent.
In fact, landscapes change all the time and dynamics are an essential property of
landscapes: changes are inherent to landscape, changes make landscapes (Clark
et al. 2003). This is also recognised in the formal definition of the European
Landscape Convention.
Dynamics of landscapes are related to their functioning as a system. Hence, the
paradigm of landscape ecology: spatial patterns and processes are interacting
continuously (Forman and Godron 1986). Landscape dynamics vary in space and
time and different dimensions of change can be recognised (Table 7.1) and specific
concepts are used (Fig. 7.1). In general, when the intensity of the dynamics
increase, changes alter faster and larger areas are affected. Some areas are more
stable than others and changes occur slowly and gradually, almost unnoticed. Other
areas transform suddenly and completely, wiping out everything that existed
before. In most cases, landscapes only change partially while keeping some conti-
nuity and “memory” from the past.
Richard Muir (2003) sees change as an essential characteristic of landscape with
cascading effects: “every change produces effects that are changes”. The amount of
change is proportional to the power, freedom and prosperity that the landowners,
communities and policymakers possess. The result of all changes makes the unique
history of a location or region. Many elements of this history are represented in the
landscape and in local narratives.

Table 7.1 Dimensions of landscape dynamics

Factor • Examples
Frequency • Duration of alternating periods of change and stability
• Cyclic changes or irreversible changes
Speed • Fast: i.e. catastrophic events
• Slow: Gradual changes such as global warming or demographic variations
Magnitude • Large: Everything is transformed; tabula rasa
• General or specific: All features change or just some (e.g. only land use changes)
• Local or global: Some places or areas change, while other remain stable; or
change affects all places
Reversibility • Cyclic
• Irreversible ¼> history
Causes • Natural causes or human-made, or combined
• Direct (causal) or indirect (induced, collateral)
After Antrop and Van Eetvelde (2008)
7.3 Landscape Genesis and History 143

Fig. 7.1 Concepts related to the study of the history of landscapes and their changes

7.3 Landscape Genesis and History

Landscape genesis and history describe and explain the landscape trajectories or
paths through time. This allows making a biography of a landscape and to deter-
mine the time depth or ‘age’ of its features.
Generally, landscape genesis covers the period in which geomorphological
processes shaped the land and is mainly focused on the physical, natural landscape
as it was formed before the permanent human imprint. Depending on the landscape,
the time covered in the study may vary a lot. Essentially, landscape genesis aims to
explain how landforms, soils and vegetation patterns developed through time.
Special attention is given to the Quaternary period (starting approximately 2.6 mil-
lion years ago) and characterised by important climatic changes and the appearance
of humans. Landscape genesis gives a synthesis of research from geology, physical
geography and geo-archaeology.
Landscape history starts with the written accounts and with descriptions and
representations of landscapes in particular. Historical maps are important docu-
ments. Ancient written documents may also indirectly refer to the landscape as tax
records or records of food production and trade. It is important to link landscape
changes with other events in nature, politics, society, demography and economy
(Marcucci 2000).
When landscape genesis focuses on natural processes, landscape history focuses
upon human actions to make the landscape. The popularity of the study of land-
scape history did not come first from academic interest, but from local people
144 7 Landscape Dynamics and Evolution

interested in better knowing their place and landscape. In England in particular,

W.G. Hoskins’ book ‘The Making of the English Landscape’ (Hoskins 1955) was an
influential stimulant.

7.3.1 From Traditional to New Landscapes

The concept of traditional landscapes refers to idealised models landscapes that

were gradually shaped over a long period, characterised short sudden ‘revolutions’
of change and longer periods of stability and adaptation (Antrop 2000). In most of
the western world, these are agrarian landscapes as they came to be before the
‘Revolution Age’ when devastating and rapid transformations started (Vos and
Meekes 1999). Recently, Renes (2015) criticised this distinction between a stable
pre-1900 and a dynamic post-1900 history and showed that changes in the past were
often quick and intensive. More important than the speed of the changes is the
shifting attitude towards the landscape caused by these changes. Traditional land-
scapes show the spatial, ecological and historical coherence between all landscape
components in a rather multifunctional and sustainable way. They form the basis
for a generic landscape typology. They possess a distinct identity and a strong
regional diversity depending on natural and cultural characteristics of the area. In
Europe, traditional landscapes are well represented on the historical maps of the
seventeenth to early nineteenth century.
Antrop (1997) used the distinction between traditional and new landscapes to
indicate the changed attitude towards the landscape (Fig. 7.2). The traditional
landscapes remained stable enough for several generations – except for disastrous
events. It was ‘natural’ to see the landscape as a commodity, a legacy to be passed
on to the next generation. For the generations at that time, a debate on sustainability
was not relevant and the maintenance of the landscape was obvious. The attitude of
people towards the contemporary new landscapes is completely different. Nowa-
days, the landscape changes several times during one’s lifetime. Also, the pace and
scale of changes are still increasing. Landscape does not symbolize any more
stability encompassing several generations; landscape became a commodity that
is used and then dumped when no longer useful. Landscapes can be written off and
recreated, landscapes are ‘make-able’. Landscapes are important in so far they offer
‘services’ to the society.
The important changes since the second half of the eighteenth century, in a series
of entangled ‘revolutions’, caused an irreversible change in attitude towards land-
scape. The basic driving force behind these changes was the subsistence of the
growing population reducing the effects of disasters as famine and epidemics.
Already in the sixteenth century, agricultural experiments led to a better land
management and more productive farming in the densely populated regions of
Flanders and Holland. The Dutch pioneered in soil restoration and drainage, canal
building and land reclamation technology. The Flemish farmers intensified the land
use and increased the production by introducing the more effective four-field crop
7.3 Landscape Genesis and History 145

Fig. 7.2 Periods of landscape change in Europe and traditional and new landscapes (After Antrop

rotation system using turnips and clover as forage. Rapidly, many of these innova-
tions were exported to the neighbouring countries. The British Agricultural Revo-
lution enabled an unprecedented population growth, freeing the workforce from
farming and moving people to the cities (Williamson 2002; Overton 2002). The
Agricultural Revolution happened simultaneously with the cultural movement of
the Enlightenment and both helped to drive other revolutions: the Scientific and
Industrial Revolution and political revolutions such as the American and French
In less than 300 years, landscapes in mining and industrial areas and the
periphery around towns and cities completely transformed from rural to
industrial-urban. Simultaneously, the countryside elsewhere also transformed
with new modes of transportation such as canals and railways, new farming
practices, the enclosure of commons and the creation of landscape parks. Gradually,
new landscapes emerged, which reflected more globalizing trends of a networking
society (Rudbeck Jepsen et al. 2015).
Revolutions and new technology bring innovations that gradually diffuse from
the places they originated from (Hägerstrand 1967). Similar processes affect var-
ious places at different times. The shift from natural and traditional landscapes to
new landscapes within the western culture is different for north-western Europe and
North America (Moss and Okey 2004) (Fig. 7.3) and even varies within Europe
(Vos and Meekes 1999; Palang et al. 2000; Peil et al. 2004) (Fig. 7.4).
146 7 Landscape Dynamics and Evolution

Fig. 7.3 Landscape genesis in the northeast of North America (a) and in north-western Europe
(b). The estimated proportion (%) of the different landscape types is given horizontally; the
vertical axis shows the time since the seventeenth century. (w) is the proportion of wilderness,
(ic) the proportion of ancient autochthone cultural landscapes, EM indicates the beginning of the
European immigration, AR marks the agricultural revolutions or innovations. (1) transect of
traditional landscapes before 1800, (2) transect of gradually developed traditional landscapes
after 1800, with (2a) relics and non-mechanised parts, (2b) ordinary and mechanised landscapes,
(3) modern landscapes (After Moss and Okey 2004)

Fig. 7.4 A comparison of

the landscape development
in Western Europe and
Estonia (After Vos and
Meekes 1999; Palang et al.
2000; Peil et al. 2004)
7.4 Contemporary Driving Forces and Processes of Landscape Change 147

7.4 Contemporary Driving Forces and Processes

of Landscape Change

The contemporary trends of landscape change and the main driving forces are
generally known (Vos and Klijn 2000) (Table 7.2). However, how these driving
forces act locally and the processes and mechanisms that induce real changes are
still rather vague (Pinto-Correia et al. 2006; Antrop and Van Eetvelde 2008).
In general, the ongoing changes result in a polarisation of geographical space: a
concentration of people, activities and infrastructure in rather small areas, while
vast areas receive less attention and eventually become abandoned (Fig. 7.5).

7.4.1 A Global View – Driving Forces

Several, simultaneously acting processes initiate and control landscape dynamics.

Generally, these are called driving forces (Bürgi et al. 2004; Brandt et al. 1999;
Jongman and Bunce 2000) or simply drivers (Wood and Handley 2001) or keystone
processes (Marcucci 2000). Four main groups of forces can be recognized: demog-
raphy, economy, politics and natural calamities (Fig. 7.6). Each of them drives
characteristic processes that interact. Important processes are the supply in natural
resources (soil, food, fibre, water and space), production (agriculture, forestry,
industry), urbanisation (living, leisure, working) and networking (accessibility,
mobility, communication, information), urbanisation and industrialization. They
act at different scales: in the past, local and regional scales of action dominated,
today the global scale has become so important that it is regarded as a separate
driving force: globalization. Climate change has become a new main driving force
related to natural disasters at the local and regional scale.
Demography is the most direct driving force for landscape transformation as the
population size determines the extraction of natural resources, land use and settle-
ment density. Besides demography, other human driving forces are economy and

Table 7.2 Trends in landscape change processes

Maximizing land productivity <> Maximizing labour productivity
Sustainable <> Consumable
Multifunctional <> Mono-specialisation
High diversity <> Low diversity
Small (scale) <> Big (scale)
Internal, local market <> External, international market
From local to Global
From involution to Replacement
From engagement to Alienation
From integration to Segregation
After Vos and Klijn (2000)
148 7 Landscape Dynamics and Evolution

Fig. 7.5 Polarisation of

geographical space

politics (Fig. 7.6). These are strongly related and interact by the law of supply and
demand. The feedback is fast and many processes are short-term. Natural driving
forces are slow but powerful, only interacting with human drivers at long-term and
gradually transform the processes. The way in which the processes function
depends also on the available technology and the scale of action.
The first conference devoted to “Man’s role in changing the face of the earth” in
1956 showed how broad, varied and complex the human impact already was in the
mid-twentieth century (Thomas 1956). All indicators revealed that the magnitude
of change has increased dramatically during the last 300 years (Goudie 2000;
Antrop 2005, 2008) (Fig. 7.7). All changes are related to population growth, in
particular the growing proportion of the urban population. Since 2008 more people
live in urban places than in rural areas. Hence, the importance of urbanisation
processes in understanding the landscape dynamics and change.
However, Zhang et al. (2007) demonstrated that – during the pre-industrial
period and at a continental scale – using Europe and China as case studies, changes
in population size, as well as phenomena as frequencies of war, famine and
epidemics all fluctuated in a successive order corresponding to global temperature
changes. Deteriorating climate conditions resulted in a decline of the agricultural
7.4 Contemporary Driving Forces and Processes of Landscape Change 149

Fig. 7.6 Contemporary driving forces and processes of landscape change. Human drivers interact
fast by strong social and economic relations, natural drivers act more independent and at long-
term. The drivers initiate a series of actions and processes that interact with feed back loops. All
these processes have different implications according to the scale

Fig. 7.7 Indicators of environment change are related to the logistic growth of the urban
population since the Age of Revolutions (After Goudie 2000; Antrop 2005, 2008)
150 7 Landscape Dynamics and Evolution

Fig. 7.8 The system of climate change and population dynamics at a continental scale and its
effects on landscape changes. Dotted rounded boxes represent different domains, rectangular
boxes show specific processes; the thickness of the black arrows corresponds to the degree
correlation. For more explanation see Zhang et al. 2011. Climate change is expressed here only
by temperature change (Modified after Zhang et al. 2011)

production causing food prices going up until a threshold causing wars and famine
to erupt. The simultaneous ecological and social disasters caused a negative feed-
back with catastrophic population collapses in Europe and China in the fourteenth
and seventeenth centuries. It could be noticed that the time delay between the
temperature drops and population decline has decreased from 20 to 40 years
respectively, indicating that some social, political and technological mechanisms
became more effective to remediate. Since the Age of Revolutions, these causal
relations changed completely (Fig. 7.8). In a more detailed study of Europe between
7.4 Contemporary Driving Forces and Processes of Landscape Change 151

Fig. 7.9 Relations between climate change, food supply and the main periods of crisis and
population collapse in Europe between 1200 AD and 1800 AD. Indicators are: the temperature
anomaly, the agricultural production index and real grain price and the population size. Important
periods in history affecting landscape transformations are indicated (Modified after Zhang et al.

1200 and 1800 AD (Zhang et al. 2011), significant causalities were found, which
could be correlated to important landscape changes (Fig. 7.9).

7.4.2 Processes Sustainable Supply of Vital Natural Resources

Chisholm (1962) considered five primary resources for a traditional agrarian set-
tlement, which were vital for sustainable survival: permanent water supply, suffi-
cient area for arable land and grazing land, availability of fuel resources (wood,
peat) and building materials (wood, stone). If the territory occupied by the settle-
ment could not supply most of these resources, exchange of resources with other
places became necessary.
152 7 Landscape Dynamics and Evolution

The rise of urban places and cities changed the whole picture. Urbanisation is a
complex demographic and economic process that creates new landscape types. The
rapid urbanisation that accelerated with the Industrial Age also caused an ecological
transformation, which affected the global ecological stability and might have led to
geopolitical insecurity. In this context, Rees (1992) introduced the concept of the
ecological footprint. Rees and Wackernagel (1996) formulated the effect of urban-
isation upon the ecological footprint as follows:
. . .ecological footprint analysis shows that they [the cities] act as entropic black holes,
sweeping up the output of whole regions of the ecosphere vastly larger than themselves.
(Rees and Wackernagel 1996, p. 245).

Water is a vital natural resource and people have devised systems to obtain and
use water more conveniently throughout history. With the growth of the population
and the concentration of people and activities in urban places, the provision of good
quality drinking water became problematic. Water became a scarce commodity and
wars were fought to get it. During the last centuries, the huge demand of water in
industry and agriculture has led to ecological problems of sustainable supply, in
particular of drinking water. Huge disparities exist between urban and rural areas
(UNICEF and WHO 2012).
Since the discovery of fire, humans have started to exploit fossil fuels. Rodrigue
et al. (2013) recognised five phases in the evolution of the use of non-renewable
fossil fuels: from biomass sources (mainly firewood and peat) until the Industrial
Revolution, followed by the use of coal, then oil (from which humanity became
dependent in the late twentieth century) to alternative forms as natural gas, nuclear
power and fractioning of shale. As fossil energy sources are running out, the shift is
now towards renewable energy sources, such as solar, geothermal, wind and tidal
sources. All these demand vast areas of space on land or sea and have an important
impact on the visual morphology of the landscape. Production

Agriculture, Pastoralism and Forestry

With the Neolithic, a lifestyle of hunting and gathering gradually changed to new
lifestyles of nomadic pastoralism and of agriculture and settlement. These new
forms of subsistence permanently modified the natural environment and both
competed for space and land. Sedentary societies developed, based on permanent
agriculture, which supported an increasingly large population that settled in villages
and towns.
Towns took advantage of the agglomeration economies, based on the concen-
tration in one place of very diverse activities, labour specialization and diversifi-
cation and on trade (Chatterlee 2003). Thus, varied cultures developed,
characterized by arts, architecture and political structures. Towns grew into cities,
7.4 Contemporary Driving Forces and Processes of Landscape Change 153

which soon became centres of innovation and power and formed networks with
other cities connected by continental trade routes.
The land became organized territory and landownership and land use became
strictly regulated. As agriculture was developing, nomadic pastoralism needed to
adapt. Grazing rights had to be determined and the movement of herds between
winter and summer pastures became a regulated system of transhumance along
fenced routes.
In forested regions, forests were cleared for expanding agricultural land. Forests
were also exploited for grazing and wood production and silvopastoralism devel-
oped. Emanuelsson (2009) considers this “fight between food and fiber” as an
important driving force in shaping the landscape. He considers the silvopastoral
landscapes being amongst the most sustainable landscapes and characterised by a
very high biodiversity. Rackham (1990, 2003) showed how humans utilized and
transformed ancient woodlands and vegetation for their subsistence, making char-
acteristic landscapes and creating new landscape elements such as hedges, ditches,
coppices and introducing new plant species.

Industry and Energy

Before the Industrial Age, most settlements were villages and hamlets and only a
few towns and cities existed. Often there was only one primal city that was also the
capital. The mobility of most people was limited to the local territory and a nearby
market town (Robb 2008). Although the urban lifestyle is very ancient and spread
with the innovation of agriculture, the process of urbanisation accelerated during
the nineteenth century in particular in northwest Europe with the upcoming
industrialisation and an exponential growth of the population (Jordan-Bychkov
and Bychkova Jordan 2002; Vandermotten 2000).
For several reasons, urbanisation and industrialisation can be seen as multifac-
eted driving forces of landscape transformation. Industries were located at sites that
met economic criteria as availability of natural resources, accessibility for trans-
portation and labour forces. Early industrialisation was mainly located where
natural resources were found, i.e. coal, iron ore and labour was attracted from the
surrounding countryside. Other industries (e.g. textile) developed in existing towns,
transforming them rapidly into conurbations.
When the resources early industries were based on became exhausted, these
industrial areas became derelict and cover today vast areas of brownfields with
polluted soils and ground water. Wallonia (Belgium), in the nineteenth century the
most industrialised area on the European continent after the United Kingdom,
inventoried 2735 brownfields ( friches industrielles) covering an area of
ca. 13,000 ha, or 0,77% of the total area of the region.
154 7 Landscape Dynamics and Evolution Urbanisation and the Deep Rural

Urbanisation refers to (1) the transformation of a rural landscape into an urbanised

one, and (2) a process of a changing life style, which can be regarded as a functional
change preceding the morphological transformation (Antrop 2003). Thus, urbani-
sation does not remain restricted to areas near cities but can also occur in the
Both urbanisation and industrialisation introduced and stimulated new modes of
transportation. Important are tramways and railways, as these increased the mobil-
ity of a large number of people. These were the means by which the countryside
became disclosed in the nineteenth century, causing a massive migration towards
the cities and the new industrial centres. In a similar way, the general use of the car
after the Second World War and the creation of motorways increased the mobility
again in an even more flexible way. This mode of transportation became the most
important factor in urban sprawl and the urbanisation of the countryside (Figs. 7.10
and 7.11).
The model of differential urbanisation (see also Fig. 7.18) also shows that
villages and hamlets in the rural countryside can be affected by urbanisation in
various ways. It happens often in two phases. In the first phase, functional changes
happen to landscape elements, e.g. a farm becomes a restaurant, manor houses
become cultural centres, pastures become fenced horse pastures, etc. Morpholog-
ical changes are minimal. Præstholm and Kristensen (2007) call this form ‘urban-
isation in disguise’. In a second phase, also morphological changes become
important as new housing, commercial centres, etc. Van Eetvelde and Antrop

Fig. 7.10 Crossroads in the countryside always attract economic activities and initiate urbanisa-
tion (Photo Antrop 2005)
7.4 Contemporary Driving Forces and Processes of Landscape Change 155

Fig. 7.11 When new roads disclose villages in the remote countryside, urbanisation is imminent
and rapid. The mountain village Kastelli (Crete) shows a new video store next to a traditional farm
(Photo Antrop 2003)

Fig. 7.12 Some models for

the urbanisation of rural
villages: (O) original
(historical) village, (a)
expansion, (b) axial
development, (c) explosion
and differentiation, (d )
beady ring development
and (e) development of
satellites (After Van
Eetvelde and Antrop 2004)

(2004) formulated some models of the possible transformation of rural villages

(Fig. 7.12).
The population in areas situated in remote areas from the growing urban and
industrial centres gradually declines and these peripheral areas even became aban-
doned. Many villages lost their functions and services and land use became more
extensive, characterized by random reforestation (called ‘reboisements en timbre
poste’ by Nebout 1985) and subject to rewilding (Perreira and Navarro 2015).
David Lowenthal (1997) called these areas the ‘rural residue’.
156 7 Landscape Dynamics and Evolution

7.4.3 Networking: The Network Society

of the Information Age

Castells (2000a, c) introduced the concept of a network society, which is charac-

teristic for the contemporary ‘Information Age’ (Castells 2000b). It is a social
structure organized by a ‘timeless time’, which became possible using new com-
munication technologies. Castells sees it as a historical phase following the ‘clock
time’ of the Industrial Age and the ‘biological time’ that controlled the
pre-industrial times. Reducing the time of communication between different places
also means reducing distance. Besides a space of places, Castells defines the space
of flows. The space of flows makes it possible to organize social activities simul-
taneously without having spatial continuity. However, the space of flows requires a
technological infrastructure that operates from specific places and that is rooted in
the space of places. Yet, relations and flows of information in the network define the
meaning and function of the space of flows and not the relation to the place
(Fig. 7.13).
General global processes result in very different responses at the local level
(Table 7.3). For example, worldwide, economy aims to increase productivity and
reduce time and costs. This can be achieved by reducing the geographical distance
by a faster and better mobility, hence the importance of networking, communica-
tion and moving the activities where the costs are minimal. However, at the local

Fig. 7.13 A network society: space of flows and space of place (Photo Antrop 2006)
7.4 Contemporary Driving Forces and Processes of Landscape Change 157

level, displacement is not an option as activities and people are bound to the place
and moving out does not solve the local problem. This may cause a loss of
landscape maintenance and ultimately land abandonment in rural and pastoral
regions. In order to prevent this, local responses must be varied, creative, resource-
ful and multiple.
Swaffield and Primdahl (2006) discussed how two rather opposite, global
agendas affect changes of rural landscapes in a network society. The first one is
the free open market, as fostered by the World Trade Organisation (WTO) and the
second one is sustainable development as encouraged by the United Nations. These
agendas work through different scales (down) and only national, regional and local
policy levels have competences in making or steering changes in the landscape
(Hägerstrand 2001; Primdahl 2010) (Fig. 7.14).
Table 7.3 summarizes the global processes and the local responses on these
global drivers. The most important difference between the global and local scale, is
that on the local scale, spatial displacement is not an option to adapt to the global

Table 7.3 Global processes and local responses

Global processes Æ Local responses
are essentially not sustainable attempt sustainability
Increasing productivity ! Specialisation, intensification of core activities,
up-scaling; resulting in landscape
Reducing time/cost distance ! Restructuring space and landscape
Networking: information and communi- ! Where is the local place situated? (core-periph-
cation infrastructure, world cities as nodes ery), how good is the accessibility? Attempt to
get better connections.
Replacing and/or displacing depending on ! Maintenance more important than replacement,
investment, profit, depreciation more multifunctional land use, innovation and
diversifying, challenging new opportunities

Fig. 7.14 Global policy

agendas and competences
as driving forces in
changing local agricultural
landscapes. The feedback
loop between the two
agendas is weaker than their
sectorial implementation.
Darker grey shades indicate
larger competences and
authority concerning the
landscape (Based on
Primdahl and Swaffield
2010a, b; Hägerstrand
158 7 Landscape Dynamics and Evolution

7.4.4 Calamities

Any natural event with catastrophic consequences for living things in the vicinity is
called a calamity or disaster. The United Nations (UN) defines a disaster as “a
serious disruption of the functioning of society, causing widespread human, mate-
rial or environmental losses, which exceed the capacity of the affected society to
cope using only its own resources.” Climate change is likely to increase the
frequency and severity of extreme weather events and hence natural catastrophes
could increase significantly. Other consequences are political instability and war,
which eventually stimulate mass migrations of the population.
A report issued on November 2015, 25th by the UN, “The Human Cost of
Weather Related Disasters” (UNISDR 2015) shows that over the last 20 years,
90% of the major disasters have been caused by 6457 recorded floods, storms, heat
waves, droughts and other weather-related events. The report also highlights data
gaps, noting that economic losses from weather-related disasters are much higher
than recorded, as only 35% of the records include information about economic
losses. UNISDR estimates that the true figure on disaster losses – including
earthquakes and tsunamis – exceeds US$250 billion annually. Economic losses
are a major development challenge for many least developed countries battling
climate change and poverty. These economic losses also affect the insurance sector
significantly (Botzen et al. 2010). On the other hand, climate change can also bring
new business opportunities for insurers.

7.5 Models and Processes of Urbanisation

Modelling urbanisation allows understanding new spatial patterns that are formed
and helps predict trends in the transformation of the surrounding countryside.
According to the criteria used to define “urban”, several models have been built.
These allow to analyse the complex process of urbanisation and to define new types
of landscapes that are created.
Early models of suburbanisation, also called peri-urbanisation, focused in rather
static zones of the transformation of the countryside as a function of the distance to
the city centre. An early example is the concentric model of Bryant et al. (1982),
introducing a series of new concepts (Fig. 7.15). The sphere of influence from the
city was called the urban field and several urban fringes were recognised. Each of
the concentric zones had a specific character. Successive stages of urbanisation are
dependent on the accessibility of places and affect in variable ways the existing land
use, agriculture in particular (Bryant and Johnston 1992). The model of Lewis and
Maund (1976) links the successive phases of urbanisation to the accessibility of
places by car from the city centre. They showed how the urban sprawl follows the
roads (Fig. 7.16). Landscape dynamics increase in well accessible areas, while
7.5 Models and Processes of Urbanisation 159

Fig. 7.15 Urbanisation around cities creates concentric zones with different functions and
morphology according to distance: (1) city, (2) inner urban fringe, (3) outer urban fringe, (4)
urban shadow, (5) the rural hinterland, countryside, (6) weekend and seasonal dwelling zone; (a)
maximal range of commuting zone, (b) urban field, (x) small urbanized centre, ( y) isolated
settlement (After Bryant et al. 1982)

Fig. 7.16 Urbanisation

related to transportation
infrastructure: (C) city
centre, (S) urban sprawl, (a)
main road, (b) secondary
road, (1), (2), (3) first,
second and third phase of
urbanisation of rural
settlements. us: areas of
urban shadow (After Lewis
and Maund 1976)

areas that are not disclosed by roads remain less dynamic. They are situated in the
urban shadow (see also Fig. 7.17).
This theoretical context helps to understand the processes of urban sprawl and
the urbanisation of the countryside. All settlements are nodes in an urban network at
a certain hierarchical level. This has been studied by urban geographers, such as
Edgar Kant (Jauhiainen 2005; Buttimer 2005) and Walter Christaller, who took it as
a basis for his central place theory (Christaller 1933). However, these early
approaches were still based on the ‘clock time age’ where exchanges between
places happened by roads and railways.
The higher a place in the hierarchy of an urban network, the better its commu-
nication with places at the same level. Thus, between the main centres, better, faster
160 7 Landscape Dynamics and Evolution

Fig. 7.17 The principles of urban implosion and urban shadow. Large cities develop a better
communication with each other than with smaller places between them. Good indicators are the
frequency of public transportation and travel times. In time-space, this results in an urban
implosion as large centres are closer together and smaller places become remote. Consequently,
the less served areas nearby large cities become less dynamic

and more frequent connections develop so time-distance between them is decreas-

ing. Smaller places near the main centres benefit less from these better communi-
cations and hence became more distant in time-dimension. Haggett (2001) called
this process urban implosion, which partly explains the existence of urban shadow
zones (Fig. 7.17).

7.5.1 Urban or Not Urban, That’s the Question

Defining the urbanised landscape is difficult, but so is the dichotomy between urban
and rural. Up to the Industrial Age, the distinction was clear, as towns and cities
were places well separated from the countryside. They were defended by walls and
had specific privileges and even legislation. The United Nations Demographic
Yearbook of 1955 (United Nations 1955) formulated the problem in distinguishing
between urban and rural today as follows:
. . .there is no definite point in the continuum between “rural” of the particular and “urban”
where the one ends and the other begins. The line drawn between the two is, therefore,
arbitrary. (p. 16).

Countries use different concepts and criteria to define urban, which makes an
international comparison difficult (Frey et al. 2001). Concepts as, town, village and
hamlet mean very different realities from country to country. The main difference
lies in the way in which the size of the settlement is defined. This can be based on
the population or on the build-up area. Both methods have specific difficulties. The
population is often counted by administrative units, such as municipalities. How-
ever, urban agglomerations seldom follow administrative boundaries. Also, urban
agglomerations have increasingly fuzzy borders, making a clear demarcation with
the rural difficult (Pacione 2001).
Complex processes related to urbanisation resulted in numerous new concepts
and terminology to describe all new phenomena. Terminologies as suburban, semi-
7.5 Models and Processes of Urbanisation 161

urban, peri-urban, urban fringe are common and sometimes used as synonyms but
sometimes have subtle different meanings depending on the linguistic context. See
Meeus and Gulinck (2008) for a recent overview. These landscapes are the ones
most people live and work in, also called ordinary landscapes (Meinig 1979; Groth
and Bressi 1997), the everyday landscape, in French paysage du quotidien
(Luginbühl 2012).

7.5.2 A Process of Diffusion: From Urban Sprawl

to Functional Urban Areas

Geyer and Kontuly (1993) developed a model of differential urbanisation that

describes how different phases in urbanisation interact between large and small
places (Fig. 7.18). It shows how urbanisation gradually spreads as a diffusion wave
in space, affecting ultimately the remotest countryside. The basis of the model is the

Fig. 7.18 The model of differential urbanisation. (a) The net migration balance of the population
towards the city centre defines the urbanisation phases. Different phases are recognised. The first
phase consists of migration from the countryside to the city centre, causing a decline in the nearby
towns and villages. First, the city centre grows (U ), followed by an expansion by suburbanisation
(S). In the second phase, the growth of the city slows down and is followed by an exodus of the
population. This phase is called counter-urbanisation and causes the nearby towns to grow by
immigration, followed by a population increase of villages. The last phase is one of re-urbanisation
of city centres attracting people again, causing gentrification of the (often renovated) historical
centre. (b) shows systematically what happens in space (After Geyer and Kontuly 1993; Champion
and Paddison 2001; Antrop 2004a)
162 7 Landscape Dynamics and Evolution

recognition of different phases in the urbanisation process. Geyer and Kontuly

(1993) defined these by the net migration in the (historical) centre of the place.
The first phase corresponds to the population growth in the city centre by migration
of people from smaller places in the countryside. Two stages are identified: first
when the net migration is still increasing and in a second stage when this slows
down and is called suburbanisation (Antrop 2004). This can be related to a shift
from the growth of the centre to a growth of the fringe. The moment that the
population of the urban place is declining is called the population reversal. Simul-
taneously, but out in phase, smaller places as towns and villages make a similar
transition. The decline of the population growth of the city and the growth of the
population of towns and villages is called the phase of counter-urbanisation, which
corresponds to what is commonly called peri-urbanisation or urbanisation of the
countryside. Geyer and Kontuly (1993) also defined a third phase in the cycle they
called re-urbanisation, corresponding to the gentrification of the renovated city
The causes of the shift from urbanisation to counter-urbanisation are important.
These are multiple and the following reasons have been proposed:
• Loss of qualities in suburbs due to the cheap housing quality, buildings often not
compliant with the current environmental and safety rules (energy consumption,
etc.), prices, traffic congestion and landscape deterioration
• Services and industry seeking rural locations: growing job opportunities in the
outer urban fringe with good accessibility (edge cities), loss of time due to traffic
congestion near cities and increase of telework
• The countryside becoming empty and cheap, more available space and often less
severe environmental restrictions
• Changes in agriculture: subsidies for development of rural activities (e.g. EU
CAP for rural development), more part-time and hobby farmers who are
“attached to the land” but not part of the “maintenance of the fabric of rural
society” (Robinson 1990)
• Second homes: often in abandoned rural buildings (farms), clustering of second
houses in ‘summer suburbs’
• Retirement migration: elderly people moving to coastal areas.
In Europe, the population reversal showed a shift in time from north to south
starting between 1950 and 1960 in north-western Europe, moving around the 1970s
to France and Italy and occurring in southern Europe between 1980 and 1990
(Antrop 2004; Champion and Paddison 2001, Cheshire 1995). This also indicates
a diffusion of the urbanisation process at a continental scale.
Conceiving urbanisation as a diffusion process explains why complex forms of
urbanisation are found all over the countryside. Figure 7.19 summarises the evolu-
tion to this complexity. The initial setting is a hierarchical network of places
consisting of a city (A), some towns (B and C) and villages (D) spaced according
to the central place theory and connected by main roads. The city is best situated
and has the highest connectivity. Hence, it attracts economic activities and labour
force, expanding in successive phases (1–7). This development is mainly controlled
7.5 Models and Processes of Urbanisation 163

Fig. 7.19 Star-shaped model of the complex urbanisation process. An urban network consists of
places of different size: (A) city, (B) town, (C) small town, (D) village, (E) village absorbed in
urban agglomeration, (F), (G) and (I ) isolated villages in the countryside and (H ) crossroads
clusters. These are connected by different means of transportation: (a) river, (b) main (paved) road,
(c) railroad, (d ) motorway, (e) rural road. Large places have an old historical centre (1) and former
town walls became ring roads (2). Successive phases of urban sprawl are (3) the nineteenth-
century fringe, (4) districts around railway stations, (5) industrial zones, (6) the first phase of high-
density suburbanisation, absorbing adjacent rural villages, (7) low-density suburbanisation of the
surrounding countryside (rurban). River valleys constitute greenways with important natural
qualities (n). Other developments are: (8) ribbon-building along the roads offering a good
accessibility to the main centres, (9) urban-commercial corridors, (10) new activity zones near
crossroads and edge cities, (11) crossroads settlements old (h) and new (g), (12) weekend houses,
(13) exurbs, (14) up-scaled farms move out of the village, (15) wind farm, (16) shopping mall, (17)
compound. Figure not on scale (After Antrop 2000)

by the transportation modes. Initially waterways and land routes, from the nine-
teenth century also the railway, since World War II the generalised mobility by cars
and motorways demanded new infrastructure. This is the phase of urban sprawl,
ribbon-building along roads in the countryside (8) and the development of urban
corridors (9). How small places and the countryside benefit from the economic
growth and how they are affected by urbanisation, depends on their disclosure by
the transportation infrastructure. Places connected by railway develop a new district
(4) and attract new industry and commerce. Good motorway connection allows
bypassing the traffic congestion near the main centre. As wet valleys are not
suitable for building, most of the urban expansion follows higher ground,
transforming the agricultural land. Valleys remain long untouched and conserve
important natural qualities (n). In planning, they are conceived as greenways or
164 7 Landscape Dynamics and Evolution

‘green fingers’ in the urban matrix but are still threatened by industry, commerce
and housing in search of new space nearby the city centre. The natural and rural
qualities also attract recreation and weekend housing (12). The development of
smaller places in the countryside depends on local land qualities and opportunities.
Some villages decline and may become deserted (G), others more attractively
situated (F) attract city-dwellers who restore old houses or build their own com-
munity in an exurb (13). Villages in good farmland (I) transform in secondary
homes while farms up-scale outside the village (14). Crossroads (11) possess a good
connectivity and attract new economic activities. Well-connected areas at the edge
of the urban agglomeration attract new industrial zones (5) or edge cities where no
people work, but only work) (10). The latest developments in the countryside
consist of installations in open landscapes for renewable energy production such
as wind farms (15), easy accessible, huge shopping malls (16) where a lot of space
is available to build from scratch (even on dispense of natural and rural qualities)
and expensive residential compounds (17) where the landscape is redesigned,
forming an alien body inside the traditional landscape.
The result of this diffuse development is a fragmented, urbanised rural landscape
with a lot of new elements and more multifunctional character. With a high density
of urban places, the landscape becomes a mix of rural and urban elements. Antrop
(2004) called it rurban, others speak of neo-rural (Gulinck 2004), fractal city
(Batty and Longley 1997), ‘ville émergente’ (Dubois-Taine and Chalas 1997),
hybrid landscapes (Qvistr€om 2013), peri-urban landscapes (Præstholm and
Kristensen 2007; Moreira et al. 2016) and many other terms are used as well
(Hidding and Teunissen 2002; Meeus and Gulinck 2008).
Characteristic for these urbanised landscapes are the functional relations. Hence,
a whole new series of new concepts has emerged, such as the Functional Urban
Zone (FUZ) , Functional Urban Area (FUA) (OECD 2012), Functional Urban
Region (FUR) (Antikainen 2005), Urban Metropolitan Zone (UMZ) (EEA 2000) or
Polycentric Metropolitan Areas (poly-FUAs) in particular in highly urbanised
Europe. They all have different meanings depending on the definitions, methods
and used data. Most definitions combine different criteria such as administrative
entities (e.g. boundaries), morphological characteristics (e.g. specific land uses) or
functional features (e.g. population density or thresholds, travel patterns) and these
urban areas typically extend beyond the administrative and morphological bound-
aries of a town or city. For example, an Urban Morphological Zone is defined as “a
set of urban areas lying less than 200m apart”, where “urban areas” are defined from
CORINE land cover classes that “contribute to the urban tissue and function”.
Consequently, the UMZs change over time. The results are compiled and defined in
the Urban Atlas (EEA 2002) and linked to the Urban Audit by Eurostat (Eurostat
2010). UMZs are used as a spatial basis to define environmental indicators,
e.g. related to sustainability and vulnerability to climate change. It is obvious that
these ways to define urban areas are remotely based on landscape reality. However,
as planning tools, they have an important impact on the landscape transformation in
the future.
7.6 Concepts and Models to Study Landscape Change 165

7.6 Concepts and Models to Study Landscape Change

7.6.1 The Biography of a Landscape

The biography of a landscape gives a holistic account of the genesis and history of a
landscape in a broad interdisciplinary perspective, covering a period from prehis-
tory until today. Samuels (1979) introduced the concept ‘Biography of Landscape”.
He considered the landscape as the ‘expression of authorship’, referring to the
particular role of individuals in the shaping of landscapes. Hence, a biography of
landscape requires a historical knowledge of the role of individuals, of their ideas
and actions, in the making of landscapes (Kolen 2005; Kolen et al. 2015).
A landscape biography is not merely a description of the evolution of elements,
structures and themes but also focuses on the story of their relations and interactions
and on the causes of their transformation. It also relates to the cultural biography of
places or regions (Elerie and Spek 2010). The landscape is conceived in successive
time layers, which are still partially recognisable in the actual landscape. Narratives
enhance the values and meanings of the relics still present and support the public’s
awareness of landscape as heritage (Bloemers et al. 2010).
Although the concept landscape biography is new, the approach is not. Earlier
regional studies of landscapes were also holistic and covered the same concept but
did not use the name. This was the case for most regional studies of geography,
e.g. the account of Alexander von Humboldt about Cuba and Mexico, Gran€o’s
description of the Altai and the regional monographs of the French school of
geography. Also, some approaches of landscape atlases resulted in regional land-
scape monographs and are very similar to the Dutch concept of landscape biogra-
phy. Examples are the Irish Atlas of the Rural Landscape (Aalen et al. 1997), the
Atlas Paysagers in Wallonia (Maréchal 2007) and France (Luginbühl 1994) and
landscape atlases in Germany (Gotzmann 2010).
Palang et al. (2011) argued that the biography approach may help to encourage
planners, citizens and stakeholders to think in the longer term and that the decisions
made today influence the processes in the future in a better way when considering
the lessons from the past. Landscape biography and the path dependency approach
help local decisions to preserve and enhance the uniqueness of each landscape
against the general, globalising driving forces.

7.6.2 Space-Time Models to Study Change

Landscape change analyses are based on the combined, integrated and simulta-
neous interpretation and analyses of multiple data sources, such as remotely sensed
data, historical and current topographical maps and field measurements. The choice
of analysis approaches, methods and spatial and temporal scales depends on
research aims and landscape characteristics as well as on the available data sources
166 7 Landscape Dynamics and Evolution

Fig. 7.20 Models and concepts in studying spatio-temporal landscape changes. (a) snapshot
model, (b) base state with amendments-model, (c) space-time composite-model. T1, T2, T3, T4
series of time-slice snapshots, time intervals are not necessarily equal; T1 – T2 etc. consecutive
changes; T1 + T2 + T3 + T4 space-time composite with corresponding transition matrix showing
the trajectory type. TDx is the time depth of place x on the map. CF is the frequency of changes.
The hypothetic example shows a transition from a rural landscape (R) to an urban one (U). (Van
Eetvelde and Käyhk€ o 2009, based on Langran 1993)

(Bürgi and Russell 2001; Antrop and Van Eetvelde 2008). The cartographic space-
time models, which are applied in landscape change detection, can be grouped into
three main conceptual models: snapshot, base state with amendments and space-
time composite (Fig. 7.20) (Van Eetvelde and Käyhk€o 2009; Langran 1993). These
models are based on the use of a sequence of cartographic (map) data, extending
(retrospectively) several decades, even centuries. Landscape change applications
are practically built on integrated spatio-temporal analyses of different time layer
maps, which represent landscape patterns through time.
The snapshot model emphasises the state of the landscape at each time layer,
resulting in a time series of time slices of significant periods (e.g. Bender et al.
2005; Van Eetvelde and Antrop 2009; Van Eetvelde and Antrop 2009; Ruiz and
Domon 2009). The selection of the periods is mostly derived by the availability of
cartographic and remote sense data.
The two other models emphasise the processes of change enabling interpretation
of the amount, patterns and nature of changes. The spatial comparison of consec-
utive time periods fits into the base state with an amendments-model. This gives the
possibility to assess the changes between two time periods by using a change matrix
or binary change maps, giving quantitative measurements of the degree of change
and no change between two adjacent time layers (Van Eetvelde and Käyhk€o 2009).
7.6 Concepts and Models to Study Landscape Change 167

The time depth, landscape change trajectory and landscape paths illustrate
applications of the space-time composite-model (Skånes and Bunce 1997; Mertens
and Lambin 2000; Fairclough and Rippon 2002; Fairclough 2003; Crews-Meyer
2004; Käyhk€ o and Skånes 2006). Landscape change trajectory analyses (LCTA)
focus on the identification of long-term dynamics and evolution of landscape
patterns and processes and they proved to be useful in the investigations of the
relation between land cover patterns and various environmental variables (Bürgi
et al. 2004; Hietel et al. 2004; Reger et al. 2007), in the exploration of deforestation
and reforestation processes (Mertens and Lambin 2000; Lambin et al. 2003;
Kennedy et al. 2007; Hartter et al. 2008), in the assessment of the status of valuable
habitats and species (Johansson et al. 2008; Käyhk€o and Skånes 2008) and the
protection of the cultural landscape (Fairclough 1999).

7.6.3 Time Depth and Landscape Paths/Trajectories

Space-time models allow describing the unique development of a landscape and

new concepts such as time depth (Clark et al. 2004), landscape trajectory and
landscape path (Vuorela et al. 2002; Käyhk€o and Skånes 2006; Ruiz and Domon
2005) were introduced.
Time depth defines the period elements and structures originated that can still be
recognised in the landscape. Landscape paths or trajectories describe the successive
stages in the development of the landscape. This is often based on an analysis of key
elements in the landscape. Since the availability of detailed topographical maps,
landscape paths can easily be constructed since the eighteenth century, covering the
fast transformation from traditional to new landscapes. Landscape trajectories are
represented by time series of maps, composite maps and (carto-)diagrams. Impor-
tant changes are linked to general or local events (political, social, economic, etc.).
The example of Mol (Kempenland, Belgium) shows social and economic events
and technology has transformed the landscape several times during the last
300 years (Figs. 7.21, 7.22 and 7.23). Place names and historical facts help to
trace back events long before maps existed, showing the roots of places in time-
depth layers, as for example the different rural villages that are all absorbed now in
the agglomeration of the town Landen (Fig. 7.24).
168 7 Landscape Dynamics and Evolution

Fig. 7.21 Landscape characteristics and elements that initiated changes are identified on histor-
ical maps. In this example of Mol (Kempenland, Belgium) a time series of maps of 1770, 1832,
1934, 2003 and 2007 was used. The maps of 1770 and 1973 are shown (maps: Carte de Cabinet of
de Ferraris 1770 (KBB), Institut Cartographique Militaire ICM 1872 (National Geographic
Institute NGI 2003)
Fig. 7.22 The example of Mol: some exemplary landscape trajectories mapped on an aerial image
of 2007. The transitions (>) between the five maps are indicated for the selected spots 1–27.
Letters represent the different landscape types at each moment: H heathland, O open field,
E enclosed files, M marshland, Mi irrigated land and water-meadows, W woodland, C conifer
plantation, L lake, V village, U urbanised, R recreation, S sand pit, N (new) nature, X crossroads
infrastructure. The number at the end of a trajectory gives the number of changes between 1770
and 2007 (background image Google Earth 2007)

Fig. 7.23 Landscape transformation and events for the example of Mol. Letters refer to the main
landscape types as in the previous figure. Events: 1 digging of the canals, 2 creation of irrigated
water-meadows, 3 new cropland with open fields, 4 exploitation of white sands, 5 restoration of
sand pits into lakes, 6 creation of new nature on former water-meadows, 7 suburbanisation of
former open fields; dates of important events: 1830 Belgium independent, 1847 introduction of the
new land development law, 1980 disclosing the area by motorways
170 7 Landscape Dynamics and Evolution

Fig. 7.24 Time depth of places and sites represented in a 3D diagram. The example of Landen
(Belgium): the market town Landen is situated at the margin of the fertile loess plateau of
Haspengouw, which was already intensively cultivated during Roman times. The earliest medieval
settlement was at a place now called St.Gitter and the tomb (motte Pepijn) of Pepin the Elder, the
founder of the Merovingian dynasty, illustrates the importance of the place. During the medieval
period many small agrarian villages were founded (place names and years indicated), belonging to
competing counties and bishoprics (dioceses). To compete with the prince-bishop of Liège, the
duke of Brabant founded nearby a new town Landen, which he granted much freedom. Conse-
quently, Landen grew and St.Gitter became deserted. In 1837, the town was connected by a
railway, which was the start of a rapid urban sprawl, which absorbed nearby villages
References 171


Aalen, F. H. A., Whelan, K., & Stout, M. (Eds.). (1997). Atlas of the Irish rural landscape
(352 pp). Cork: Cork University Press.
Antikainen, J. (2005). The concept of functional urban area – Findings of the ESPON project 1.1.1.
Informationen zur Raumentwicklung, Heft, 7, 447–456.
Antrop, M. (2003). Continuity and change in landscapes. In U. Mander & M. Antrop (Eds.),
Multifunctional landscapes, Continuity and change (Vol. 3, pp. 1–14). Southampton: WIT
Antrop, M., & Van Eetvelde, V. (2008). Mechanisms in recent landscape transformation. In Ü.
Mander, C. A. Brebbia, & J. F. Martin-Duque (Eds.), Geo-environment and landscape evolu-
tion III (pp. 183–192). Southhampton: WIT Press.
Antrop, M. (1997). The concept of traditional landscapes as a base for landscape evaluation and
planning. The example of Flanders region. Landscape and Urban Planning, 38, 105–117.
Antrop, M. (2000). Changing patterns in the urbanized countryside of Western Europe. Landscape
Ecology, 15(3), 257–270.
Antrop, M. (2004). Landscape change and the urbanisation process in Europe. Landscape and
Urban Planning, 67(1–4), 9–26.
Antrop, M. (2005). Why landscapes of the past are important for the future. Landscape and Urban
Planning, 1-2, 21–34.
Antrop, M. (2008). Landscapes at risk: About change in the European landscapes. Evolution of
geographical systems and risk processes in the global context (2008). In P. Dostal (Ed.),
Evolution of geographical systems and risk processes in the global context (pp. 57–79).
Prague: Charles University Prague.
Batty, M., & Longley, P. (1997). The fractal city. Architectural Design, 67(129), 74–83.
Bender, O., Boehmer, H. J., Jens, D., & Schumacher, K. P. (2005). Using GIS to analyse long-term
cultural landscape change in southern Germany. Landscape and Urban Planning., 70(1–2),
Bloemers, J. H. F., Kars, H., van der Valk, A., & Wijnen, M. (Eds.). (2010). The cultural landscape
heritage paradox. Protection and development of the Dutch archaeological-historical land-
scape and its European dimension. Amsterdam: Amsterdam University Press.
Botzen, W. J. W., van den Bergh, J. C. J. M., & Bouwer, L. M. (2010). Climate change and
increased risk for the insurance sector: A global perspective and an assessment for the
Netherlands. Natural Hazards, 52, 577–598.
Brandt, J., Primdahl, J., & Reenberg, A. (1999). Rural land-use and landscape dynamics—
Analysis of ‘driving forces’ in space and time. In R. Kr€ onert (Ed.), Land-use changes and
their environmental impact in rural areas in Europe, Man and the biosphere series (Vol.
24, pp. 81–102). Paris: Parthenon Pub.
Bryant, C. R., & Johnston, T. R. R. (1992). Agriculture in the city’s countryside (233 pp). Toronto:
University of Toronto Press.
Bryant, C., Russwurm, L., & McLellan, A. (1982). The city’s countryside: Land and its manage-
ment in the rural urban fringe. London: Longman.
Bürgi, M., Hersperger, A. M., & Schneeberger, N. (2004). Driving forces of landscape change –
Current and new directions. Landscape Ecology, 19, 857–868.
Bürgi, M., & Russell, E. W. B. (2001). Integrative methods to study landscape changes. Land Use
Policy, 18, 9–16.
Buttimer, A. (2005). Edgar Kant (1902–1978): A Baltic pioneer. Geografiska Annaler: Series B,
Human Geography, 87(3), 175–192. September 2005.
Castells, M. (2000c). Materials for an exploratory theory of the network society. British Journal of
Sociology, 51(1), 5–24.
Castells, M. (2000a). The rise of the network society. Oxford: Blackwell.
Castells, M. (2000b). The information age: Economy, society and culture. (Vol. 3, Updated
edition). Oxford: Blackwell.
172 7 Landscape Dynamics and Evolution

Champion, T. (2001). Urbanisation, suburbanisation, counterurbanisation and reurbanisation. In

R. Paddison (Ed.), Handbook of urban studies (pp. 143–161). SAGE Publications: London.
Chatterlee, S. (2003). Agglomeration economies: The spark that ignites a city. Businees Review,
Q4, 6–13.
Cheshire, P. (1995). A new phase of urban development in Western Europe? The evidence for the
1980s. Urban Studies, 32(7), 1045–1063.
Chisholm, M. (1962). Rural settlement and land use. London: Hutchinson University Library.
Christaller, W. (1933). Die zentralen Orte in S€ uddeutschland. Jena: Fischer.
Clark, J., Darlington, J. & Fairclough, G. (2003). Pathways to europe’s landscape (114 pp).
European Pathways to Cultural Landscape (EPCL),
Clark, J., Darlington, J. & Fairclough, G., (2004). Using historic landscape characterisation
(63 pp). English Heritage & Lancashire County Council.
Crews-Meyer, K. A. (2004). Agricultural landscape change and stability in northeast Thailand:
Historical patch-level analysis. Agriculture, Ecosystems and Environment, 101, 155–169.
Dubois-Taine, G., & Chalas, Y. (1997). La ville ́emergente. La Tour d’Aigues: Éditions de l’Aube.
EEA. (2000). Urban morphological zones 2000.
EEA. (2002). Towards an urban atlas: Assessment of spatial data on 25 European cities and urban
areas. Environmental issue report No 30.
Elerie, H. & Spek, Th., (2010). The cultural biography of landscape as a tool for action research in
the Drentsche Aa National Landscape (northern Netherlands). In: Bloemers, J.H.F., H. Kars,
A. van der Valk & M. Wijnen (eds) The cultural landscape heritage paradox. Protection and
development of the Dutch archaeological-historical landscape and its European dimension.
Amsterdam University Press. Amsterdam. pp. 83–113.
Emanuelsson, U. (2009). The rural landscapes of Europe. How man has shaped European nature
(p. 383). Stockholm: The Swedish Research Council Formas.
Eurostat. (2010). Urban audit database.
region_cities/city_urban. Accessed 1 Sept 2010.
Fairclough, G., & Rippon, S. (Eds.). (2002). Europe’s cultural landscape: Archaeologists and the
management of change. Europae Archaeologiae Consilium, Occasional Paper 2, Brussels.
Fairclough, G. (1999). Protecting the cultural landscape: National designation and local character.
In J. Grenville (Ed.), Managing the historic rural landscape (pp. 27–39). London: Routledge.
Fairclough, G. (2003). The long chain: Archaeology, historical landscape characterization and
time depth in the landscape. In H. Palang & G. Fry (Eds.), Landscape interfaces. Cultural
heritage in changing landscapes (pp. 295–318). Dordrecht: Kluwer Academic Publishers.
Forman, R., & Godron, M. (1986). Landscape ecology (619pp). New York: J. Wiley.
Frey, W. H., & Zimmer, Z. (2001). Defining the city. In R. Paddison (Ed.), Handbook of urban
studies (pp. 14–35). SAGE Publications: London.
Geyer, H. S., & Kontuly, T. M. (1993). A theoretical foundation for the concept of differential
urbanisation. International Regional Science Review, 15(12), 157–177.
Gotzmann, I. (2010). Die Kulturlandschaft und ihre Bestandaufname. Betrachtungen aus
geographisch-landeskundlicher Sicht (pp. 72). Bonn: Bund Heimat und Umwelt in
Goudie, A. (2000). The human impact on the natural environment (511 pp). Oxford: Blackwell
Groth, P., & Bressi, T. W. (Eds.). (1997). Understanding ordinary landscapes (272 pp). New
haven: Yale University Press.
Gulinck, H. (2004). Neo-rurality and multifunctional landscapes. In: J. Brandt & H. Vejre (Eds.),
Multifunctional landscapes, Vol. 1: Theory, values and history, Vol. 14 of Advances in
ecological sciences (pp. 63–74). Southampton: WIT Press.
References 173

Hägerstrand, T. (1967 [1953]). Innovation diffusion as a spatial process [Innovationsf€orloppet ur

korologisk synpunkt]. Postscript and translation by Allan Pred; Translated with the assistance
of Greta Haag. Chicago: University of Chicago Press. OCLC 536383.
Hägerstrand, T. (2001). A look at the political geography of environmental management. In
A. Buttimer (Ed.), Sustainable landscapes and lifeways: Scale and appropriateness
(pp. 35–58). Cork: Cork University Press.
Haggett, P. (2001). Geography: A global synthesis (672 pp). Harlow: Prentice-Hall.
Hartter, J., Lucas, C., & Caughan, A. E. (2008). Detecting tropical dry forest succession in a
shifting cultivation mosaic of the Yucatán Peninsula, Mexico. Applied Geography, 28,
Hidding, M. C., & Teunissen, A. T. J. (2002). Beyond fragmentation: New concepts for urban-
rural development. Landscape and Urban Planning, 58(2–4), 297–308.
Hietel, E., Waldhardt, R., & Otte, A. (2004). Analysing land-cover changes in relation to
environmental variables in Hesse, Germany. Landscape Ecology, 19, 473–489.
Hoskins, W. G. (1955). The making of the English landscape (third ed.). London: Hodder and
Stoughton. 1988.
Jauhiainen, J. (2005). Edgar Kant and the rise of modern urban geography. Geografiska Annaler.
Series B, Human Geography, 87(3), 193–203.
Johansson, L. J., Hall, K., Prentice, H. C., Ihse, M., Reitalu, T., Sykes, M. T., & Kindstr€ om,
M. (2008). Semi-natural grassland continuity, long-term land-use change and plant species
richness in an agricultural landscape on Öland, Sweden. Landscape and Urban Planning, 84,
Jongman, R. H. G., & Bunce, R. G. H. (2000). Landscape classification, scales and biodiversity in
Europe. In U. Mander & R. H. G. Jongman (Eds.), Consequences of land use changes
(pp. 11–38). Southampton: Wit Press.
Jordan-Bychkov, T. G., & Bychkova Jordan, B. (2002). The European culture area. A systematic
geography (437 pp). New York: Rowman & Littlefield Publishers.
Käyhk€o, N., & Skånes, H. (2006). Change trajectories and key biotopes—Assessing landscape
dynamics and sustainability. Landscape and Urban Planning, 75(3), 300–321.
Käyhk€o, N., & Skånes, H. (2008). Retrospective land cover/land use change trajectories as drivers
behind the local distribution and abundance patterns of oaks in south-western Finland. Land-
scape and Urban Planning, 88, 12–22.
Kennedy, R. E., Cohen, W. B., & Schroeder, T. A. (2007). Trajectory-based change detection for
automated characterization of forest disturbance dynamics. Remote Sensing of Environment,
110, 370–386.
Kolen, J. (2005). De biografie van het landschap. Drie essays over landschap, geschiedenis en
erfgoed. Amsterdam: Vrije Universiteit.
Kolen, J., Renes, H., & Hermans, R. (Eds.). (2015). Landscape biographies. Geographical,
historical and archaeological perspectives on the production and transmission of landscapes,
Landscape and heritage studies (p. 437). Amsterdam: Amsterdam University Press.
Lambin, E. F., Helmut, J., Geist, H. J., & Lepers, E. (2003). Dynamics of land-use and land-cover
change in tropical regions. Annual Review of Environmental Resources, 28, 205–241.
Langran, G. (1993). Time in geographical information systems. London: Taylor and Francis.
Lemaire, T. (2002). Met open zinnen. Natuur, landschap, aarde (358pp). Amsterdam: Ambo.
Lewis, G. J., & Maund, D. J. (1976). The urbanisation of the countryside: A framework for
analysis. Geografiska Annaler, 58B, 17–27.
Lowenthal, D. (1997). European landscape transformations: The rural residue. In P. Groth & T. W.
Bressi (Eds.), Understanding ordinary landscapes (pp. 180–188). New Haven: Yale University
Luginbühl, Y. (2012). La mise en scène du monde. Construction du paysage européen (p. 432).
Paris: CRNS Editions.
174 7 Landscape Dynamics and Evolution

Luginbühl, Y. (1994). Méthode pour des atlas de paysages – Identification et qualification; Strates/
CNRS, Ministère de l’Aménagement du Territoire, de l’Equipement et des Transports/Direc-
tion de l’Architecture et de l’Urbanisme.
Marcucci, D. J. (2000). Landscape history as a planning tool. Landscape and Urban Planning, 49
(1–2), 67–81.
Maréchal, L. (Ed.), (2007). Atlas des Paysages de Wallonie. CPDT, Ministère de la Région
Wallonne, Namur.
Meeus, S. J., & Gulinck, H. (2008). Semi-urban Aeras in landscape research: A review. Living
Rev. Landscape Research, 2(2008), 3.
Meinig, D. W. (1979). In J. B. Jackson, P. F. Lewis, D. Lowenthal, D. W. Meinig, M. S. Samuels,
D. E. Sopher, & Y.-F. Tuan (Eds.), The interpretation of ordinary landscapes. Geographical
essays. New York/Oxford: Oxford University Press.
Mertens, B., & Lambin, E. F. (2000). Land-cover-change trajectories in southern Cameroon.
Annals of the Association of American Geographers, 90, 467–494.
Moreira, F., Fontes, I., Dias, S., Batista e Silva, G., & Loupa-Ramos, I. (2016). Contrasting static
versus dynamic-based typologies of land cover patterns in the Lisbon metropolitan area:
Towards a better understanding of peri-urban areas. Applied Geografie, 75, 49–59.
Moss, M. R., & Okey, B. W. (2004). A landscape-ecological appraisal of the diversity and
relevance of ultural landscapes. In R. H. G. Jongman (Ed.), The new dimensions of the
European landscapes, Wageningen UR Frontis Series (pp. 219–235). Dordrecht: Springer.
Muir, R. (2003). On change in the landscape. Landscape Research, 28(4), 383–403.
Nebout, J.P. (1985). Le movement d’extension forestière en Montagne Bourbonnaise aux XIXe et
XXe siècles. Courrier de la Montagne Bourbonnaise, n 22, 17–47.
OECD. (2012). Definition of functional urban areas (FUA) for the OECG metropolitan database
(p. 9). Paris: OECD.
Overton, M. (2002). Agricultural revolution in England 1500–1850. London: BBCi History. http://
Pacione, M. (2001). Models of urban land use structure in cities in the developed world. Geog-
raphy, 86(2), 97–119.
Palang, H., Alumäe, H., & Mander, Ü. (2000). Holistic aspects in landscape development: A
scenario approach. Landscape and Urban Planning, 50, 85–94.
Palang, H., Spek, T., & Stenseke, M. (2011). Digging in the past: New conceptual models in
landscape history and their relevance in peri-urban landscapes. Landscape and Urban Plan-
ning, 100(4), 344–346.
Peil, T., Sooväli, H., Palang, H., Oja, T., & Mander, Ü. (2004). Estonian landscape study:
Contextual history. Belgeo, 2-3, 231–244.
Perreira, H. M., & Navarro, L. (Eds.). (2015). Rewilding European landscapes (227 pp). Cham:
Springer International Publishing.
Pinto-Correia, T., Gustavsson, R., & Pirnat, J. (2006). Bridging the gap between centrally defined
policies and local decisions – Towards more sensitive and creative rural landscape manage-
ment. Landscape Ecology, 21(3), 333–346.
Præstholm, S., & Kristensen, S. P. (2007). Farmers as initiators and farms as attractors for
non-agricultural economic activities in peri-urban areas in Denmark. Geografisk Tidsskrift,
Danish Journal of Geography, 107(2), 13–27.
Primdahl, J., & Swaffield, S. (2010a). Globalisation and the sustainability of agricultural land-
scapes. Landscape Research, 39(2), 123–140.
Primdahl, J., & Swaffield, S. (2010b). Globalization and rural landscape change – Key concepts,
development trends and some implications for policy. In J. Primdahl & S. Swaffield (Eds.),
Globalization and agricultural landscapes – Change patterns and policy trends in developed
countries (pp. 61–73). Cambridge: Cambridge University Press.
References 175

Primdahl, J. (2010). Globalization and the local agricultural landscape: Current change patterns
and public policy interventions. In J. Primdahl & S. Swaffield (Eds.), Change patterns and
policy trends in developed countries, Studies in landscape studies (pp. 149–168). Cambridge:
Cambridge University Press.
Qvistr€om, M. (2013). Peri-urban landscapes: From disorder to hybridity. In P. Howard,
I. Thompson, & E. Waterton (Eds.), The Routledge companion to landscape studies
(pp. 427–437). London: Routledge.
Rackham, O. (1990). Trees and woodland in the British landscape. London: Dent.
Rackham, O. (2003). Ancient woodland: Its history, vegetation, uses in England. Colvend:
Castlepoint Press.
Rees, W., & Wackernagel, M. (1996). Urban ecological footprints: Why cities cannot be sustain-
able – And why they are a key to sustainability. Environmental Impact Assessment Review, 16,
Rees, W. E. (1992). Ecological footprints and appropriated carrying capacity: What urban
economics leaves out. Environment and Urbanisation, 4(2), 121–130.
Reger, B., Otte, A., & Waldhardt, R. (2007). Identifying patterns of land-cover change and their
physical attributes in a marginal European landscape. Landscape and Urban Planning, 81,
Renes, H. (2015). Historic landscapes without history? A reconsideration of the concept of
traditional landscapes. Rural Landscapes: Society, Environment, History, 2(1), 1–11.
Robb, G. (2008). The discovery of France. A historical geography from the revolution to the first
world war (454 pp). New York: W. W. Norton & Co.
Robinson, G. M. (1990). Conflict and change in the countryside (482 pp). London: Belhaven Press.
Rodrigue, J.-P., Comtois, C., & Slack, B. (2013). The geography of transport systems (3th ed.p.
416). New-York: Routledge.
Rudbeck Jepsen, M., Kuemmerle, T., Mülkler, D., Erb, K., Verbur, P. H., Haberl, H., Vesterager,
J. P., Andrič, M., Antrop, M., Austrheim, G., Bj€ orn, I., Bondeau, A., Bürgi, M., Bryson, J.,
Caspar, G., Cassar, L. F., Conrad, E., Chromý, P., Daugirdas, V., Van Eetvelde, V., Elena-
Rosselló, R., Gimmi, U., Izakovicova, Z., Jančák, V., Jansson, U., Kladnik, D., Kozak, J.,
Konkoly-Gyuró, E., Krausmann, F., Mander, Ü., McDonagh, J., Pärn, J., Niedertscheider, M.,
Nikodemus, O., Ostapowicz, K., Pérez-Soba, M., Pinto-Correia, T., Ribokas, G., Rounsevell,
M., Schistou, D., Schmit, C., Terkenli, T. S., Tretvik, A. M., Trzepacz, P., Vadineanu, A.,
Walz, A., Zhllima, E., & Reenberg, A. (2015). Transitions in European land-management
regimes between 1800 and 2010. Land Use Policy, 49, 53–64.
Ruiz, J., & Domon, E. (2009). Analysis of landscape pattern change trajectories within areas of
intensive agricultural use: Case study in a watershed of southern Quebec, Canada. Landscape
Ecology, 24, 419–432.
Ruiz, J., & Domon, G. (2005). Integrating physical and human dynamics in landscape trajectories:
Exemplified at the Aulnages watershed (Québec, Canada). In B. Tress, G. Tress, G. Fry, &
P. Opdam (Eds.), From landscape research to landscape planning: Aspects of integration,
education and application, Wageningen UR Frontis Series, n 12 (pp. 67–81). Berlin: Springer
Science+Business Media.
Samuels, M. S. (1979). The biography of landscape: Cause and culpability. In D. W. Meining
(Ed.), The interpretation of ordinary landscape (pp. 51–88). Oxford: Oxford University Press.
Swaffield, S., & Primdahl, J. (2006). Spatial concepts in landscape analysis and policy: Some
implications of globalisation. Landscape Ecology, 21(3), 315–331.
Thomas, W. L. (Ed.). (1956). Man’s role in the changing face of the earth (1193 pp). Chicago: The
University of Chicago Press.
UNICEF and World Health Organization. (2012). Progress on drinking water and sanitation:
2012 Update. WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation.
UNISDR (2015). The human cost of weather related disasters.
46793 Accessed 1 Dec 2015.
176 7 Landscape Dynamics and Evolution

United Nations. (1955). Demographic yearbook 1955. New York: United Nations, Statistical
Van Eetvelde, V., & Antrop, M. (2004). Analyzing structural and functional changes of traditional
landscapes – Two examples from southern France. Landscape and Urban Planning., 67(1),
Van Eetvelde, V., & Antrop, M. (2009). Indicators for assessing changing landscape character of
cultural landscapes in Flanders (Belgium). Land Use Policy, 26, 901–910.
Van Eetvelde, V., & Käyhk€ o, N. (2009). The applicability of quantitative techniques for assessing
spatio-temporal patterns of landscape changes. In J. Breuste, M. Kozová, & M. Finka (Eds.),
European landscapes in transformation. Challenges for landscape ecology and management.
European IALE conference 2009. 70 years of landscape ecology in Europe (pp. 379–382).
Salzburg: Univ. of Salzburg.
Vandermotten, C. (2000). Building a continental area: Identities, differences and urban develop-
ments in Europe. BELGEO Belgian Journal of Geography, 1, 114–142.
Vos, W., & Klijn, J. (2000). Trends in European landscape development: Prospects for a sustain-
able future. In J. Klijn & W. Vos (Eds.), From landscape ecology to landscape science
(pp. 13–30). Netherland: WLO, Wageningen/Kluwer Academic Publ.
Vos, W., & Meekes, H. (1999). Trends in European cultural landscape development: Perspectives
for a sustainable future. Landscape and Urban Planning, 46(1–3), 3–14.
Vuorela, N., Alho, P., & Kalliola, R. (2002). Systematic assessment of maps as source information
in landscape change research. Landscape Research, 27(2), 141–166.
Williamson, T. (2002). The transformation of rural England: Farming and the landscape,
1700–1870. Exeter: Exeter University Press.
Wood, R., & Handley, J. (2001). Landscape dynamics and the Management of Change. Landscape
Research, 26(1), 45–54.
Zhang, D.D., Brecke, P., Lee H.F., He, Y.-Q. and Zhang J., 2007. Global climate change, war, and
population decline in recent human history. PNAS, 104,49, pp.19214–19219.
Zhang, D.D., Lee H.F., Wang, C., Li, B., Pei, Q., Zhang J., Ylun A., 2011. Causality analysis of
climate change and large-scale human crisis. PNAS, 108,42, pp.17296–17301.
Chapter 8
Analysing Landscape Patterns

Abstract Very diverse phenomena compose the landscape, configuring complex

patterns at different spatial scales. The properties of these patterns allow describing
the transcendent and holistic qualities of the landscape, such as landscape hetero-
geneity, coherence, connectedness and diversity. Often methods and techniques are
borrowed from different disciplines, adapted and used in a new context. Landscape
ecology has been a catalyst in this process and laid the basis for quantitative pattern
analysis and the use of landscape metrics. To reduce the complexity, several models
are used to describe the spatial properties of landscape. Five are discussed based on
the primitives considered: (1) element, component, structure, (2) point, line, poly-
gon, surface, (3) patch, corridor, matrix, mosaic, (4) mass, screen, space, and
(5) landmark, district, path, node, edge. Mapping landscape patterns causes an
abstraction of the landscape, which does not guarantee to keep its holistic character.
However, landscape metrics allow to describe some holistic meta-properties of the
landscape such as its heterogeneity, diversity, information content and connectivity.

Keywords Landscape models • Patterns and processes • Entropy • Order •

Landscape metrics • Diversity • Heterogeneity • Coherence • Connectivity

8.1 Introduction

To reduce the complexity of landscapes, different models can be conceived allowing

to study properties that are focusing on specific features, or on structures that
represent a holistic meta-reality, the ‘more’ in the expression “landscape is more
than the sum of its composing parts.” Five models for decomposing the landscape are
discussed. Each model uses its proper terminology and some specific methods.
Nowadays, GIS-analysis, spatial statistics and visualization techniques have become
common techniques, allowing quantitative descriptions and extensive map produc-
tion and computer visualization accessible to all. However, the perspective narrows
to a vertical, bird’s-eye view and pitfalls in map use and statistical interpretation are
common. Some basics in spatial pattern analysis and information theory as well as the
use of maps as tools for analysis will be discussed. Special attention will be given to
landscape metrics, which became important tools to quantify transcendent landscape
properties such as heterogeneity, diversity and connectedness.

© Springer Science+Business Media B.V. 2017 177

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_8
178 8 Analysing Landscape Patterns

8.2 Decomposing the Landscape: Models for Analysis

The landscape is composed of very diverse phenomena configuring complex

patterns at different spatial scales. Describing, mapping and analysing become
easier when the landscape is decomposed in thematic layers and spatial dimensions.
Various models have been designed in order to facilitate this.

8.2.1 Model 1: Element, Component, Structure

Elements, components and structures are the components of a landscape based on

spatial properties.
Elements are discrete phenomena with crisp borders, such as objects: a tree, a
house, a field. The edges or borders are clear, material or formally defined as for
example in administrative borders and some property boundaries. The scale of
the study defines how the elements are described. At a large scale, details of a
house and its footprint can be mapped. On small scales, houses are symbolised
by dots or aggregated to build up areas represented as a polygon. On a small
scale, adjacent fields with similar land cover will be grouped into one polygon,
while fields smaller than the pre-set Minimum Mapable Unit (MMU) will be
Components are phenomena that vary continuously through space and seldom
present clear edges or borders. The terrain surface is a typical example, with
varying slopes, exposition and elevation. Other components are soil conditions,
groundwater and microclimatic variations. In order to describe and map compo-
nents, these must be sampled and subdivided in well-defined classes or categories.
The choices of the describing attributes and variables, sampling scheme, number of
classes, class-limits and classification method are made according to the objectives
of the analysis.
Structures are ways in which elements or components are linked and related
spatially or functionally. Thus, structures are essentially holistic and have a mean-
ing that transcends the importance of its composing elements. Structures are
conceived mentally to reduce the complexity of landscape for a better understand-
ing. They are the first step to modelling of the landscape. Consequently, a same set
of elements can be ordered in different structures according to the aim or function
Structures are defined by:
• the nature of the selected elements that compose the structure
• the nature of the relations taken into consideration
• the type of structure.
8.2 Decomposing the Landscape: Models for Analysis 179

For example, the following types can be recognised:

• spatial structures
• relational structures
• temporal structures
• functional structures.
The most obvious representations of spatial structures are maps or plans. These
are two-dimensional structures with a spatial basis defined by the scale and map
projection. However, gradually more 3D and even 4D representations become
available. Without a spatial basis, they become schemes. Spatial structures in the
landscape are often represented as patterns, networks or configurations. Examples are
field systems, settlement patterns, road networks and networks of hedgerows. These
are all tangible, but also intangible structures exist, such as territorial subdivisions.
Relational structures focus more on the nature and strength of the relationship
between the composing elements. Examples are sociograms, settlement hierarchies
and topological graphs, such as network schemes of public transportation.
Temporal structures describe the coherence and transformation between succes-
sive events or conditions. Examples are the geological timetable, crop calendars
and phenology schemes of the vegetation, seasonal movements of people and cattle
as in transhumance (Fig. 8.1).

Fig. 8.1 A landscape is composed of discrete objects (elements), continuous components and
structures that forming networks. The landscape of Mount Cassel (France) (Photo M. Antrop 2005)
180 8 Analysing Landscape Patterns

8.2.2 Model 2: Point, Line, Polygon, Surface

In a digital GIS-environment, all real objects and landscape phenomena are

geocoded in a format that is the most suitable for the purpose. For example, at a
small-scale analysis of settlement patterns, buildings can be coded as points, having
X-, Y- and Z-coordinates for their location, and an ID that links this point to an
attribute table. For a large-scale analysis, the shape and height of the building can
be important and then the geocoding will use polygons instead.
Basically four primitives are used for geocoding: point (node), line (poly-line,
arc), polygon and surface (volume). The choice of the geocoding determines which
further analyses are possible (Figs. 8.2 and 8.3).
Points, lines and polygons are used to represent objects. Landscape mosaics, as
for example formed by land cover or territories, are represented by tessellations of
polygons. Surfaces are constructed from discrete observations to represent contin-
uous phenomena, such as the terrain and landform.

8.2.3 Model 3: Patch, Corridor, Matrix, Mosaic

This is a typical model used in landscape ecology and was introduced by Forman
and Godron (1986) (Fig. 8.4). A patch is an area in the landscape that is beneficial
for the species under consideration, while all space that is not, is referred to as the
matrix. Linear elements connecting patches are called corridors; they fragment the
matrix, but also connect patches and form networks. In the mosaic version, the
landscape is simplified as a tessellation of polygons representing landscape types or
land cover types.
Shape, size, composition and configuration of these primitives are used in spatial
analysis and geostatistics. For patches a core and an edge can be identified, which
have different ecological properties. Edge effects increase when the size (area) of
the patch decreases and its shape becomes less compact (Fig. 8.5). The width of the
corridor also defines the core and edge. A particular case appears when corridors are
process-oriented, such as a stream (Fig. 8.6). Corridors and patches form spatial
networks, which can be described and studied using network analysis (Fig. 8.26).
The matrix is defined as the landscape feature that has the largest extent and
highest connectedness, i.e. it occupies most of the space in a continuous way. The
percolation theory (Gardner and O’Neill 1990) relates the connectedness of the
matrix to the area the patches occupy. It defines when a matrix inversion occurs
(Fig. 8.7). An example is the transformation from a rural landscape to an urbanised
one (Fig. 8.8).
8.2 Decomposing the Landscape: Models for Analysis 181

Fig. 8.2 Landscape elements can be geocoded in different ways according to the needs. As
landscape ecological primitives: (a) patch, (b) corridor and (c) matrix; as GIS-primitives: (1)
points, (2) lines, (3) polygons and (4) surfaces. The choice depends on the purpose of the analysis,
scale and data format. (Photo J. Semey 1989)

8.2.4 Model 4: Mass, Screen, Space

Mass, screen and space are the primitives used when describing the visual structure
of the landscape scenery (Fig. 8.9). Mass consists of volumes, which hide the
landscape behind, such as buildings, hills and woods. Screens are tall linear
elements obstructing the view. They can be walls or linear vegetation and may
form networks. A space is an open area without any elements obstructing the view
and can be bordered by masses or screens. Masses and screens can be either biotic
182 8 Analysing Landscape Patterns

Fig. 8.3 Landscape

elements and components
are coded in a different way
according to the intended
analysis and cartographic

or abiotic. They often have some ‘transparency’, which varies according to the
nature of the screen, its density and orientation in relation to the view direction
(Piessens 1985). Successive semi-transparent screens can result in a ‘closed’ land-
scape type although the coverage and density of the screen elements is scarce. This
is typical in certain bocage-type landscapes that have a ‘filtered’ appearance.
These primitives are used to map visual characteristics of the landscape (De Veer
and Burrough 1978). Figure 8.10 gives an example. Simon Bell (2004) uses a more
elaborated set of ‘basic elements’ for visual design in landscape architecture: point,
line, plane, solid volume and open volume. Each of them is characterized by a series
of variables: number, position, direction, orientation, size, shape, interval, texture,
density, colour, time, light, visual force and visual inertia. Combinations of elements
are possible and their spatial composition and configuration (Bells calls this organi-
zation) form patterns and define the visual character of the landscape.
8.2 Decomposing the Landscape: Models for Analysis 183

Fig. 8.4 The patch-

corridor-matrix model for
landscape analysis.
Important properties are
connectivity and
connectedness. In the case
of a landscape mosaic all
patches have more or less
the same size and are
characterised by a scale-
dependent heterogeneity

Fig. 8.5 Variables to

describe patches: the ratio
between the edge and core
depends on the size (area)
and shape. A shape index
expresses the compactness
of the patch in comparison
to a circle

8.2.5 Model 5: Landmark, District, Path, Node, Edge

Kevin Lynch introduced these terms in his book The Image of the City (1960) as the
elements that describe the image of a place and which we use when constructing
mental maps of places. Landmarks are special tall elements, often eye-catchers,
184 8 Analysing Landscape Patterns

Fig. 8.6 Different types of corridors: (a) linear, (b) strip corridors, (c) stream corridors. C: core,
E edge, e: ecoduct. 1-7: cross-sections: 1, 3, 6 are corridors made by space bordered by vegetation,
2, 5 are corridors made by vegetation bordered by open space, 7 stream corridors often are
complexes of previous types

Fig. 8.7 Matrix inversion in a landscape conceived of patches in a matrix. The matrix is the
landscape type with the largest extent and connectedness. When increasing the number of patches
as well as the summed size of the patches reaches a proportion of 0.5928, these become the matrix
with the largest connectedness. This threshold is generally valid in all cases and is called the
percolation threshold (after Gardner and O’Neill 1990)

which we use to orient ourselves in space and which are often regarded as an icon of
the place. Paths, such as streets and waterways, make networks allowing move-
ment. Paths intersect and form nodes that are special elements that exert attraction,
such as crossroads or a square. Building blocks, parks, lakes form areal units with a
8.2 Decomposing the Landscape: Models for Analysis 185

Fig. 8.8 Example of a matrix inversion: the transition of a rural landscape in 1862 (matrix ¼ agri-
cultural land) to an urbanised one today (matrix becoming built-up area). Example from the
suburbs of Ghent (Belgium): the built-up area in 1862 is shaded grey (Photo M. Antrop 2000)

Fig. 8.9 Successive semi-transparent screens of tree rows result in a cumulative filtering of the
open space, while biotic masses (as wood) and abiotic masses (as buildings) obstruct the view
(Photo M. Antrop 2005)

distinct character and are called districts. The interface between districts and other
elements are called edges with proper characteristics.
The importance and significance of these elements may vary for different
groups, such as inhabitants and tourists. Lynch developed different methods to
describe and map these elements to characterise the cognitive map of a place or city
and to define its legibility or imageability. These maps were intended to support
urban design and spatial planning.
Although designed for urban places, the ideas and methods proposed by Lynch
proved to be useful in landscapes as well (Fig. 8.11).
186 8 Analysing Landscape Patterns

Fig. 8.10 Example of

mapping visual primitives:
(1) open space, (2) water,
(3) woods, (4) buildings and
constructions, (5) screen:
droves, (6) screens:
hedgerows and coppiced

Fig. 8.11 The elements of legibility of places according to Kevin Lynch (1960)
8.3 The Map Is Not the Landscape 187

8.3 The Map Is Not the Landscape

Aerial images show a birds-eye view of the landscape in its holistic appearance
revealing its ‘total character’ and need to be interpreted. The result is often a series
of thematic maps representing the different landscape layers of components. Maps
are often a basic data set for analysing landscape patterns. Also, they are most
commonly used to visualise and to communicate the results of analyses and
interpretations. With the introduction of GIS, mapping became easy to all, causing
a deluge of maps. However, many users are not professionally trained in cartogra-
phy and hence – often unconsciously – lying with maps became quite common
(Monmonier 1996). The same could be said for using statistics (Huff and Geis

8.3.1 Mapping Landscape Features

Different patterns and structures exist according to the elements taken into consid-
eration. Different types can be recognised and scale-dependency is an important
criterion to make the differentiation and to model and code the elements. Specific
methods have been developed to analyse each type.
Spatial point patterns are formed by elements modelled as points, described by
spatial co-ordinates (x, y and eventually z) and one or several attributes, which can
be qualitative (nature, category) of quantitative (size, etc.). This is a typical way of
modelling discrete objects, such as buildings or settlements, at a course, regional
scale. A typical density and spatial distribution are important characteristics of
point patterns. Basic questions for the analysis are: are the point elements randomly
scattered? or do the show spatial aggregation and clustering or some regular
(geometric of planned) order?
Depending on the scale, elements can be modelled as polygons and considered a
patch pattern. For example, at a more detailed scale, buildings are coded as poly-
gons representing their footprints. In landscape ecology, a large number of tools are
available to analyse patches and patch patterns.
Linear elements form networks, characterised by the nature of the linear ele-
ment, but also by density and connectivity of the whole pattern and eventually the
orientation when transportation or stream networks are concerned. Borders and
edges can also be considered as networks. In this case, often the ‘transparency’, the
crispness and fuzziness of the edges are properties to be studied as well.
Surface patterns are formed by continuous phenomena, such as topography,
landform, groundwater table and temperature. In general, they are referred to as
geographical surfaces. To analyse surface patterns, methods developed for the
analysis of the relief and terrain models can be used: defining (relative) elevation
differences, slope degree and orientation, flow directions, etc. As the topographical
188 8 Analysing Landscape Patterns

surface is ‘carrying’ most of the discrete elements composing the landscapes, the
analysis of terrain models is important in any form of landscape analysis.
Geographical surfaces are modelled in different ways. Continuous surfaces can
be constructed from discrete point data. For example, the groundwater table can be
modelled from observations of the water level in wells, a continuous population
density of potential surface can be modelled from the population size of settle-
ments, etc. Geographical surfaces are constructed from the interpolation of discrete
observations and a wide variety of geostatistics exists for their analysis and is
available in GIS.
In order to study spatial patterns and to compare different geographical surfaces,
it is often necessary to reduce their continuous character into a limited set of
discrete classes. For example, the continuous variation of the slope degree can be
reduced to three classes: flat, moderate and steep. The choices of the number of
classes and the method to define them, depend on the purpose of the analysis and
will also determine the way their spatial pattern will be represented on a map.
Mosaic patterns are formed by a tessellation of polygons of different nature.
Land cover and vegetation patches are typical examples. To this group also belong
the map units on choropleth maps, such as administrative divisions and territories,
or building density by census district. A particular case is the grid map used for
sampling of aggregating data in a regular tessellation of squares or hexagons.

8.3.2 Size and Scale Dependency in Choropleth Maps

Basically, the units on choropleth maps are conceived as spatially uniform and
homogeneous. The map units define the grain with varying sizes and shapes, which
has important consequences when using them to analyse spatial mosaic patterns and
when visualising them on maps. Apparent spatial associations and patterns are size
and scale dependent (Figs. 8.12 and 8.13).

8.4 Everything Is Related to Everything Else

Tobler’s first law in geography (“Everything is related to everything else, but near
things are more related than distant things.” Tobler 1970) emphasises the impor-
tance of distance in connectivity matters. Relations result from the interaction
between different landscape components. They can be deterministic or stochastic,
and stable or variable. Relations create spatial or functional coherence. Relations
define landscape heterogeneity and diversity. They also depend on the spatial
configuration of landscape features, which is expressed in their connectedness
and connectivity.
8.4 Everything Is Related to Everything Else 189

Fig. 8.12 The area-dependence problem in classification in choropleth maps. The theoretical grid
map (a) contains cells with randomly assigned values between 0 and 9, representing, for example,
the number or density of landscape elements within the grid cell. Map (b) shows an overlay of nine
zones (e.g. administrative units), each equal to four adjacent cells. The total count of the grid
values is assigned to each zone and visualised with a grade of shades on the choropleth map, which
shows a concentration of high values on the lower edge. Map (c) shows a different overlay of
equal-sized zones (e.g. parish boundaries), which removes the concentration completely after
visualisation. Note also that the standard deviation S of the mapped values dropped from 7.645 to
4.714. (after Unwin 1981)

Fig. 8.13 The size-dependency problem in choropleth mapping. In the theoretical map (a) a
zoning of a regular grid is used to sample two variables x and y (values between brackets) in order
to calculate the correlation between them. For map (a) the product-moment correlation is
r ¼ 0.715. In map (b) one of the zones has doubled in area and once again in map (c). Clearly,
the overlay of data-sampling zones determines the outcome. This kind of juggling with zone
boundaries is the basis for the gerrymandering by redrawing electoral boundaries (after Unwin

8.4.1 Vertical and Horizontal Relations

Coherence refers to the degree in which features of different kinds are associated
with each other in space or time. Mander et al. (2010) discuss the different ways in
which spatial coherence is defined. Phipps (1984) related coherence to the concept
of order in the landscape and makes the distinction between vertical relations
190 8 Analysing Landscape Patterns

Fig. 8.14 Vertical and horizontal relations in the landscape. The two sections through hypothet-
ical landscapes give an observation of the soil type (B) and land cover (L) at regular intervals.
Section (a) shows a clear association between both components: land cover changes each time
when the soil type changes. There is a strong vertical (ecological) relation between both resulting
in a spatial co-variation. Section (b) shows no relation between both components. However, the
landscape is not showing a chaotic pattern but possesses a repeating spatial zonation formed by a
horizontal relationship between land cover types with a sequence FcpB

(ecological order) and horizontal relations (topological order). Van Mansvelt

(1997) distinguished three groups of ecological coherence: the vertical (on site),
the horizontal (landscape-level) and the cyclical (temporal) coherence. Mander
et al. (2010) used the spatial autocorrelation as an estimate of landscape coherence
(Fig. 8.14).

8.4.2 Correlation and Coherence

The strength of the relation between landscape components can be expressed in

several ways, depending on the properties of the used data. Data can consist of
qualitative descriptions or qualitative measurements and different scales of mea-
surement can be used. Data can refer to discrete objects or continuous phenomena.
Often sampling is used, so a basic question is how representative the data set
is. Quantitative expressions of the strength of the relation or association use
correlation coefficients or similarity indices. However, they never indicate if the
relationship is causal. An idea of the nature of the dependency can be found using
regression analysis. Applying such statistical methods often show low values of
association between landscape components, in particular when comparing to the
one expected from a holistic observation of the landscape. Much has to do with the
scale and sampling properties.
An example of the analysis of an aerial photograph of Han-sur-Lesse (Belgium)
illustrates this (Fig. 8.15 and Table 8.1). Croplands, meadows, deciduous and
coniferous forests each occupy approximately one-quarter of the sample. The
inceptisols are clearly dominant and take half of the sample and the slope classes
are equally distributed. Cropland mainly occupies the gentle slopes and colluvial
soils and inceptisols, while meadows are mostly located on the flat, alluvial soils.
Also, the settlements and orchards are found on these soils. Forests are distributed
8.4 Everything Is Related to Everything Else 191

Fig. 8.15 Aerial photographs easily show relations between landscape components, especially
when examined stereoscopically. From relations between land cover, landform and slope relations
with soil conditions and the geological substrate can be deduced. In this case, meadows and hay
land are found on the alluvial valley soils, while cropland is located on gentle to moderate slopes
on colluvial soils and inceptisols on a shale or limestone substrate. Deciduous and coniferous
forests are found on steep slopes and stony limestone hills. Sampling by means of a regular grid
allows a quantitative assessment of the relation between these components (aerial photograph of
Han-sur-Lesse, Department of Geography, Ghent University, 1970)

over moderate and steep slopes and inceptisols and stony grounds. In this case, all
observed frequencies are significantly non-random.
Figure 8.16 illustrates the principle in one dimension using a transect. The grain
and the extent define the scale of the investigation. The grain gives the smallest
spatial difference (resolution) between values and often depends on the technique
or instrument used to measure (Fig. 8.17).
192 8 Analysing Landscape Patterns

Table 8.1 Relations between the landscape components land cover, slope degree and soil type for
the sample of the Han-sur-Lesse photograph

Land cover Soil type

A W B D E L N P a c s g r Totals
Slope degree 0 0 13 2 2 2 0 0 0 17 5 1 0 0 23
1 18 5 0 0 0 4 4 0 2 5 22 4 0 33
2 4 4 0 0 0 14 6 1 0 4 18 2 5 29
3 0 0 0 0 0 11 14 0 0 0 3 0 21 25
Totals 26 21 2 4 2 29 24 2 19 14 45 6 26 110
Soil type a 0 13 2 2 2 0 0 0
c 11 3 0 0 0 0 0 0
s 14 0 0 2 0 16 11 2
g 1 5 0 0 0 0 0 0
r 0 0 0 0 0 13 13 0

Absolute frequencies are given. Land cover: A cropland, W meadows, B orchards, D village,
E water, L Deciduous forest, N coniferous forest, P heathland. Soil type: a alluvial, c colluvial,
s inceptisol, g soil on gravel, r rocky ground. Slope degree: 0 flat, 1 gentle, 2 moderate, 3 steep

Fig. 8.16 Scale and measurement of a continuous spatial variable. The figure represents a series
of measurements made by sampling a continuous varying variable at regular intervals (black
circles) along a transect. The variation between the measurements partially depends on the
distance of lag between the measurements. Also, values can be aggregated. The grain and extent
define the scale of the investigation
8.4 Everything Is Related to Everything Else 193

Fig. 8.17 Landscape heterogeneity as a function of the scale (grain) (after Burel and Baudry
2003). The most detailed image has a pixel resolution of 2. Coarser maps are obtained by pixel
thinning. Consequently, the difference between the adjacent pixels and thus the heterogeneity of
the image increases. At a certain value (here 20) the maximal heterogeneity is obtained, reflecting
the macro structure of the landscape (represented here by three grey values). The increasing
variation is the result of the decreasing number of pixels. This kind of analysis is similar to the
shuttle analysis as described by Forman and Godron (1986)

8.4.3 Landscape Heterogeneity and Diversity: Applications

of the Information Theory

Kilchenmann (1973) proposed a method to use information entropy as a tool to

correlate qualitative thematic maps and to assess regional classifications. Later,
ocker and Bergmann (1978) applied the information theory to formulate two
models to quantify the relationship between landscape elements.
The information entropy is expressed quantitatively by the formula of Shannon-
Weaver, which is also used in landscape ecology to measure the heterogeneity of
the landscape. The average information entropy is given by:
H ðAÞ ¼  pi :log2 pi

S is the number of different signs, pi the proportional occurrence of a sign (or class)
i of a phenomenon A (the ‘message’). The logarithmic base number 2 is used to
194 8 Analysing Landscape Patterns

express the information entropy H in bits. The maximal entropy is obtained when
all classes have an equal chance to occur in the message, thus when:
p1 ¼ p2 ¼ . . . ¼ pn

or calculated as:

Hmax ðAÞ ¼ log2 ðSÞ

The evenness expresses the degree in which the different signs are represented
equally, thus:
EðAÞ ¼ H ðAÞ=Hmax ðAÞ

The redundancy is its complement and is often represented as a percentage:

RðAÞ ¼ 1  EðAÞ and as percentage : RðAÞ% ¼ 100:ð1  EðAÞÞ

Expressing the information entropy in bits, thus 1 or 0, yes or no corresponds to

the number of Boolean questions one must ask to get the right answer. E.g. S ¼ 2
and p1 ¼ forest and p2 ¼ cropland, than H ¼ 1 bit corresponding to one Boolean
question: is it forest? Yes gives a1 and no logically a2.
In the following table (Table 8.2), the theoretical series of different arrange-
ments of elements are compared according to their information properties. These
can be conceived as different landscape elements along transects. Sequence 1 is
composed of all similar elements and shows no variation: the information entropy
H ¼ 0 and the redundancy is maximal. The information entropy is a measure of the
amount of information the series gives, or in what degree the series of brings ‘new’
information, i.e. is unexpected. Simultaneously information entropy also expresses
order: higher values signifying more disorder and chaos. The evenness (E)
increases when some elements dominate in the series. Redundancy (R), expressed
as a percentage here is the complement of evenness. Sequence 1 contains no
information, as all successive signs are the same and no new sign is unexpected.
The evenness is zero as no sign dominates over another. Sequence 2 is composed of
two elements (o and x) occurring at an equal frequency. Both the entropy H and
evenness E are maximal and equal to one, which is the maximal value that can be

Table 8.2 Sequences of Sequence H (bits) E R%

different arrangements of
1 oooooooooooooooo 0.000 0.000 100%
16 signs and the
2 oxoxoxoxoxoxoxox 1.000 1.000 0%
corresponding information
3 oooxoooxoooxooox 0.811 0.811 19%
entropy H expressed in bits,
4 oo_o_oo_o_oo_o_o 0.954 0.954 5%
evenness E and redundancy R
5 oloxoloxoloxolox 1.501 0.947 5%
expressed in %
6 pbacefdghjklmioq 4.000 1.000 0%
7 abcdefghijklmopq 4.000 1.000 0%
8.4 Everything Is Related to Everything Else 195

obtained (Hmax ¼ 1). Consequently, the redundancy is zero. Sequence 3 is also

composed of two elements, but one dominates the other: the information entropy
and evenness decrease, and the redundancy rises accordingly. Sequence 4 shows a
rhythmic pattern with a varying distance along the transect. It is formed by two
elements, where ‘o’ can be conceived as ‘mass’ (patches, woods, buildings) and ‘_’
as ‘space’ (matrix). Here ‘o’ dominates ‘_’ but less than ‘o’ dominates ‘x’ in
sequence 3; the information entropy rises and the redundancy decreases. Sequence
5 is composed of three different elements and is thus more diverse than the previous
series, which is expressed by higher values for the information entropy. No element
dominates over another and consequently, the evenness approaches one, while the
redundancy remains low. Finally, sequence 6 consists of 16 different elements that
only occur once. The diversity is maximal and so is the information entropy, but the
evenness is one and the redundancy 0%: no pattern can be recognised. Sequence
7 contains the same signs as series 6 and has the same values for the information
metrics. Here it becomes clear that these do not contain semantic information,
which allows us to identify the series as the alphabet.
The Shannon-Weaver information entropy is a metric that gives a synthetic
value of the composition and configuration of the whole ‘message’ (i.e. a series
of data or area). Thus, it belongs to the landscape-level metrics.
ocker and Bergmann (1978) introduced the distinction between the distribu-
tive (Hd) and the summed information entropy (Hs), depending on how the
different categories (‘signs’) are used in the summing. When the summed propor-
tion of each category is applied regardless if it is fragmented over several spatial
units, the summed information entropy Hs is obtained. If the proportion of each
spatial unit is used, the distributive information entropy Hd is obtained (Figs. 8.18,
8.19, 8.20 and Table 8.3).
The information entropy also shows the structural shift in time when a
landscape gradually transforms into another type. The transition from a rural into
an urban landscape passes through a phase of higher heterogeneity (Figs. 8.20 and
8.21 case A).

8.4.4 Networks: Connections and Fragmentation Graph Theory – Network Typologies

Connectivity is a basic concept of graph theory and important in the study of

network flow problems. Some specific terminology is used (Fig. 8.22). Connectivity
refers to the minimum number of elements (vertices or nodes, and edges or paths)
needed to connect all nodes to each other. It defines a network, represented by a
connected graph in which paths connect any two vertices to each other. If one node
is removed so the graph becomes disconnected, two subgraphs are formed. A graph
is said to be directed when the movement along the paths can go only in one way,
196 8 Analysing Landscape Patterns

Fig. 8.18 The information entropy is a measure of the diversity and heterogeneity combined.
Consider a landscape mosaic composed of different landscape types (A, B, C, D and E). The
number of the occurring types gives the diversity (‘richness’). With a constant diversity and
coverage of the types, fragmentation into smaller spatial units or patches (a and b) causes an
increase in the landscape heterogeneity proportion. The heterogeneity increases further when the
spatial units differ in size and the diversity increases (c, d, e). The information entropy expresses
this in a quantitative way (here in bits). It makes a difference whether the summed or distributive
entropy is calculated

e.g. in hydrographical networks and in some cases also in transportation networks.

Different types of networks are defined (Figs. 8.23 and 8.24). Landscape Connectivity: Multiple Meanings

The concepts connectedness and connectivity became popular in conservation

biology and landscape ecology, using the patch-corridor-matrix-mosaic model to
study spatial landscape patterns. A wide range of literature is available on this
subject (Forman and Godron 1986; Baudry and Merriam 1988; Schreiber 1988;
Turner 1990; Gardner and Urban 2007; Metzger and Décamps 1997; McGarigal
et al. 2002; Burel and Baudry 2003; Kindlmann and Burel 2008).
However, both concepts are also used in other contexts of landscape analysis.
Networks analyses developed from mathematics and topology and became popular
in geography for the study of hydrographical networks, settlement patterns, in
transportation networks and in the analysis of the visual landscape. In landscape
design and architecture, these concepts refer to visual relations between spaces and
objects. Similar uses are found in landscape archaeology.
In all cases, both concepts have different meanings and sometimes the defini-
tions are vague and generic. Often, in domains without a formal definition of
connectivity, the word is as a synonym for connectedness, while in other disciplines
the difference is important. In modern computer science and information technol-
ogy, both concepts are basic. Here, the common noun form is connectivity but the
common adjectival form is connected.
Another type of connectivity is found in regular tessellations. Here, connectivity
describes the number of neighbours accessible from a central tile. Grid based
analysis is common on raster maps, such as landform analysis on digital elevation
models, viewshed analysis and cost path analysis. Also, territories in the landscape
form tessellations. Two types of tessellations are commonly used in relation to
connectivity (Fig. 8.25).
8.4 Everything Is Related to Everything Else 197

Fig. 8.19 (a) CORINE Land Cover west of the city of Ghent (Belgium). (b) the information
entropy of the land cover defined for each square kilometre and mapped in four classes (grey
shades); superimposed white lines indicate the landscape types resulting from a holistic classifi-
cation based on a visual map interpretation. The high heterogeneity of the outer urban fringe is
clearly visible (After Antrop and Van Eetvelde 2000)

Merriam (1984) introduced the concept of landscape connectivity, which he

defined as the degree in which the landscape facilitates or impedes movements of
species among resource patches. Baudry and Merriam (1988) use the term con-
nectedness to refer to the fact that two adjacent patches of the same type are
198 8 Analysing Landscape Patterns

Fig. 8.20 Aerial photographs of two sample areas: an urban landscape (Destelbergen near the city
of Ghent) and a transition between two types of rural landscapes (Verrebroek between the enclosed
Land van Waas and the open polder land of the Scheldt river). Land use is sampled in both cases
using a 250 m by 250 m grid (see also Fig. 8.21)

Table 8.3 Landscape metrics related to the information entropy of Fig. 8.18
Metric Case (a) Case (b) Case (c) Case (d) Case (e)
Diversity 2 2 4 4 5
# patches 2 4 4 7 7
Hmax 1.000 1.000 2.000 2.000 2.322
Hs 1.000 1.000 2.000 1.906 2.031
Hd 1.000 2.000 2.000 2.500 2.500
Es 1 1 1 0.953 0.875
Ed 1 2 1 1.250 1.077
Hmax maximal information entropy in bits, summed Hs and distributive Hd information entropy
and corresponding evenness Es and Ed

spatially joined and the term connectivity for the possible movement of an individ-
ual of a given species move between patches, whether or not they are spatially
connected. However, both concepts were often used without a clear distinction.
Therefor, to remove the ambiguity, the concepts of structural and functional
connectivity are used instead (Vogt et al. 2009). Structural connectivity refers to
8.4 Everything Is Related to Everything Else 199

Fig. 8.21 Frequency distribution of the information entropy of the land use, calculated for each
grid of the sample areas Destelbergen and Verrebroek on topographical maps from three time
periods. Low values of information entropy indicate homogeneous landscapes with a low diversity
in land use types or with a dominating category. A high information entropy indicates a large
diversity in land use and/or a large spatial heterogeneity or fragmentation. The distributions for
case Destelbergen show the transition from a rural to an urban landscape: in 1910 the distribution
of the information entropy is rather normal. In 1962, the suburbanization started and the diversity
in land use categories dropped. Finally, in 1995, the whole area turned into a heterogeneous urban
landscape. During the same period, the landscape in case Verrebroek remained unchanged and
characterised by a homogenous cropland with a low diversity in land use categories. Sampling
according to Fig. 8.20 (After Antrop 1998)

the physical arrangement of landscape elements, i.e. determines connectedness.

Functional connectivity refers to the species-specific movement potential through a
landscape. A spatial analysis of the structural connectivity is used to assess or
predict the functional connectivity (Goodwin and Fahrig 2002). However, measures
of structural connectivity often have no link with movement behaviour (Metzger
and Décamps 1997, Fig. 8.26).
200 8 Analysing Landscape Patterns

Fig. 8.22 Components of a planar network: V vertex, node (1–8), E edge, arc, link (a–i; i is a loop
connecting a vertex with itself), R region, face (in this case R ¼ 3 as the surrounding space counts
also; regions defines circuits of cycles of edges forming an path), A and B are two subgraphs in the
space considered (dotted border). Euler’s formula describes the relationship between the number
of vertices, edges and regions of a planar network in an infinite space

Fig. 8.23 Network typologies: A mesh (not fully connected), B hub and spoke, C linear, path, and
D tree or stream (directed)

8.5 Landscape Metrics and Indicators

The central paradigm of landscape ecology states that the spatial patterns formed by
landscape elements is the result of ecological processes and that these patterns are
simultaneously determined by these processes (Turner 1990). This stimulated the
development of methods to describe and measure spatial patterns quantitatively and
to relate their properties to processes. A wide variety of landscape metrics and
landscape indicators, sometimes also called landscape indices, have been devel-
oped. Most of these are based on a patch-corridor-matrix-model of the landscape
(model 3) and are applied upon digital maps in raster or polygon format showing
mosaics. The development of landscape metrics and tools for spatial analysis was
made possible by the rapid development of GIS and techniques of spatial analysis
and geostatistics. The most common analysis tool that was developed especially for
this purpose is the Fragstats package (McGarigal and Marks 1995).
8.5 Landscape Metrics and Indicators 201

Fig. 8.24 (a) past landscapes had few superimposed networks of different kind, (b) contemporary
landscapes have much more non-planar networks with complex interaction

Fig. 8.25 Two commonly

used tessellation in
landscape analysis: (a)
6-connectvity in hexagonal
tiling and (b) 8-connectivity
in square tiling. Note that in
b distance equity is not kept
in all directions
202 8 Analysing Landscape Patterns

Fig. 8.26 Summary of the network analysis in a patch-corridor model. Different types of nodes
can be recognised in a network. For example, the fragmentation of a hedgerow network shows a
change in frequency of the different node-types: T-nodes decrease and end-nodes O increase (after
Burel and Baudry 2003). The detour-index gives the ratio between the real length (At) of a
segment (road, corridor, etc.) and the direct straight-line connection (Av) between the two
end-nodes. The topological structure of a network is represented in a graph. The properties of a
graph are indicated by the number of nodes, number of paths, number of circuits and number of
subnets. The cyclomatic number m gives the number of circuits. A connectivity matrix shows all direct
links between the nodes. The nodality of a node is the number of the connecting paths. The alpha-index
gives the number of circuits compared to the maximal possible circuits. The beta- and gamma-index
are the ratios between the number of nodes and paths and express the connectivity of the network

Landscape metrics can quantify landscape composition or configuration and can

be applied on three levels:
• patch-level: these metrics describe spatial properties of individual patches and
their context;
• class-level: metrics describing properties of all patches belonging to the same
• landscape-level: metrics describing spatial properties of the pattern in the whole
studied area.
Some metrics can be applied on different levels and refer to the whole study,
which defines the landscape. Typical landscape metrics to describe the composi-
tion, are:
8.5 Landscape Metrics and Indicators 203

• Dominance: the proportion of occurrence of each category in the landscape.

• Richness: the number of different categories of patches.
• Evenness: the degree in which different categories occur evenly; complement of
• Diversity: combines richness and evenness.
• Fractal dimension: a measure of the space-filling of a pattern, used in the
percolation theory.
Typical landscape metrics describing the spatial configuration are:
• Frequency of the size or area of the patches and their density within the
• Shape properties of the patches and their complexity.
• Core-edge ratio of patches.
• Proximity of patches (isolation measure).
• Contrast between adjacent patch types.
• Distribution of patches.
• Contagion: degree of aggregation of patches of the same category.
• Subdivision: degree of fragmentation of a patch type.
• Connectivity: degree of functional links between patches.
Haines-Young and Chopping (1996) proposed the following grouping of land-
scape metrics:
• Areal indices: describe the proportion of different patches of landscape types, as
well as the shape characteristics and core-edge ratio.
• Linear indices: describe the borders, shapes and network properties of linear
structures at a landscape level, such as connectivity.
• Topological indices: describe spatial relations between landscape elements and
spatial units regardless size and shape. They express spatial distribution and
association, isolation, heterogeneity and diversity.
Several hundreds of landscape metrics have been formulated and many of them
are correlated and confusing (Riitters et al. 1995; Botequilha Leit~ao and Ahern
2002; Cushman et al. 2008). Haines-Young and Chopping (1996), Wu (2004) and
many others also discussed the effects of measurement scales, scale, grain size,
zoning, areal extent and data aggregation on the outcome. Special precautions are
needed to avoid misuse (Li and Wu 2004). Cushman et al. (2008) tried to define a
set of ‘universal and consistent’ landscape metrics.
Uuemaa et al. (2009) analysed the occurrence of the terms landscape metrics,
indexes and indices in research papers dealing with landscape ecology published in
international peer-reviewed scientific journals from 1994 to October 2008. Their
use appeared to be very broad, and the following groups of themes and applications
could be recognised:
1. use/selection and misuse of metrics,
2. biodiversity and habitat analysis;
3. water quality;
204 8 Analysing Landscape Patterns

4. evaluation of the landscape pattern and its change;

5. urban landscape pattern, road network;
6. aesthetics of landscape;
7. management, planning and monitoring.
The term landscape metrics is used most frequently, in particular in relation to
software packages as Fragstats, while the term landscape indices is applied in a
broader sense. The use of landscape metrics in the international journals has shown
a strong increase since the year 2000, in particular in biodiversity and habitat
analysis and in evaluation of landscape patterns and change. However, a clear
decline in the use of the terminology can be observed from the year 2005 onwards.
Dramstad et al. (1998) already discussed the difficulties to interpret the land-
scape metrics and to relate them to indicators useful in monitoring, planning and
policy. An NIJOS/OECD expert meeting (Dramstad and Sogge 2003) revealed the
main problems in using landscape metrics or indices in policy:
• they are often very abstract and not transparent, in particular when participation
of non-experts is essential;
• they offer no absolute values and thresholds, which can be used to support policy
decisions objectively;
• there is a lot of uncertainty involved when comparing landscapes from different
areas and moments, in particular due to the different data properties and
Although most of the applications of landscape metrics relate to the analysis of
land cover patterns forming mosaics of landscape types, several attempts have been
made to use landscape indicators to describe visual landscape patterns and proper-
ties as well (Dramstad et al. 2006; Fry et al. 2009).
Essentially, landscape metrics aiming to describe quantitatively aspects such as
diversity, complexity, heterogeneity, order and chaos, all refer to the transcendent
and holistic characteristics of the landscape, to properties of ‘the whole is more than
the sum of the composing parts’. Antrop and Van Eetvelde (2000) compared the use
of landscape metrics with a holistic interpretation of suburban patterns on aerial
imagery (Fig. 8.19). The outer urban fringe with a ‘rurban’ landscape type scores
the highest values on landscape heterogeneity.
Different landscape types are also characterised by different profiles in their
landscape metrics, as showed in six square kilometre samples of Belgian landscapes
(Fig. 8.27, Tables 8.4, 8.5 and 8.6).

Fig. 8.27 CORINE Land Cover samples of six landscapes in Belgium (after Antrop 2007)
8.5 Landscape Metrics and Indicators 205

Table 8.4 Characterisation of six Belgian landscapes based upon landscape metrics derived from
the CORINE Land Cover
1 2 3 4 5 6
Region: Westhoek Straatdorpen Kempen Haspengouw Famenne Ardennen
%Build-up <1 87 49 9 8 2
%Agriculture 100 14 45 91 47 18
%Forest 0 0 4 0 44 80
%Other 0 0 <1 0 0 0
%Water 0 <1 1 0 0 00
PR 4 8 13 8 10 11
NP 60 169 124 73 152 190
MPS 4361 393 185 5326 109 139
ED 37 68 71 29 79 67
MSI 4.45 2.76 2.25 3.75 2.20 2.18
MPFD 1.31 1.30 1.29 1.31 1.30 1.29
SDI 0.23 0.68 1.52 0.55 1.77 1.75
SEI 0.13 0.33 0.59 0.26 0.77 0.73
The terminology of Fragstats is used: PR Patch Richness, NP number of patches, MPS Mean Patch
Size in m2, ED Edge Density in m/m2, MSI Mean Shape Index (circle ¼ 1, increases with
elongation of shape), MPFD Mean Patch Fractal Dimension (between 1 ¼ simple, straight borders
and two very distorted borders, SDI Shannon Diversity Index (increases with number of categories
and patches), SEI Shannon Evenness Index (as SDI, but varying between 0 ¼ homogeneous and
1 ¼ very heterogeneous) (see Fig. 8.27)

Table 8.5 Seven highly universal and consistent class-level landscape structure components
across many different cover classes in 531 landscapes across three very different and disjointed
regions of North America (after Cushman et al. 2008)
Component name Description
Edge contrast Degree of “contrast” between the focal class and its neighbourhood,
where contrast is user-defined and represents the magnitude of difference
between classes for one or more attributes.
Patch shape Shape complexity of patches of the focal class, where shape is defined by
complexity perimeter–area relationships.
Aggregation Degree of aggregation of cells of the focal class, where large, compact
clusters of cells of the focal class are considered aggregated.
Nearest neighbour Proximity of patches of the focal class, based on the average or area–
distance weighted average distance between the nearest neighbours.
Patch dispersion Spatial dispersion of patches across the landscape, reflecting whether
patches of the focal class tend to be uniformly distributed or dispersed
(clumped) based on the variability in the nearest neighbour distances.
Large patch The degree of concentration of the focal class area in few, large patches
dominance with large core areas.
Neighbourhood Degree of isolation of patches from nearby patches of the same or similar
similarity class (i.e., the degree of similarity of the neighbourhood surrounding
patches of the focal class in terms of patch composition).
206 8 Analysing Landscape Patterns

Table 8.6 Seven universal landscape structure components derived from 531 landscapes across
three very different and disjunctive regions of North America (after Cushman et al. 2008)
Component name Description
Contagion/diversity Degree of aggregation of patch types (or the overall clumpiness of the
landscape) and the diversity/evenness of patch types. Contagion and
diversity are inversely related; clumped landscapes containing large,
compact patches and an uneven distribution of area among patch types
have high contagion and low diversity.
Large patch Degree of landscape dominance by large patches.
Interspersion/ Degree of intermixing of patch types.
Edge contrast Degree of “contrast” among patches, where contrast is user-defined and
represents the magnitude of difference between classes for one or more
Patch shape Variability in the patch shape complexity, where the shape is defined by
variability perimeter–area relationships.
Proximity Degree of isolation of patches from nearby patches of the same class.
Nearest neighbour The proximity of patches to neighbours of the same class, based on the
distance area–weighted average distance between nearest neighbours.


Antrop, M. (1998). Landscape change: Plan or chaos? Landscape and Urban Planning, 41,
Antrop, M. (2007). Perspectieven op het landschap. Gent: Academia Press.
Antrop, M., & Van Eetvelde, V. (2000). Holistic aspects of suburban landscapes: Visual image
interpretation and landscape metrics. Landscape and Urban Planning, 50(1-3), 43–58.
Baudry, J. & Merriam, H.G., (1988). Connectivity and connectedness: Functional versus structural
patterns in landscapes. In: Schreiber, K. F. (Ed.), Connectivity in landscape ecology. Pro-
ceedings of the 2nd International Association for Landscape Ecology. Münstersche
Geographische Arbeiten, 29, 23–28.
Bell, S. (2004). Elements of visual design in the landscape (p. 196). London: Spon Press.
Botequilha Leit~ao, A., & Ahern, J. (2002). Applying landscape ecological concepts and metrics in
sustainable landscape planning. Landscape and Urban Planning, 59(2), 65–93.
Burel, F., & Baudry, J. (2003). Landscape ecology: Concepts, methods and applications (362 pp).
Enfield: Science Publishers.
Cushman, S. A., McGarigal, K., & Neel, M. C. (2008). Parsimony in landscape metrics: Strength,
universality, and consistency. Ecological Indicators, 8(5), 691–703.
De Veer, A., & Burrough, P. (1978). Physiographic landscape mapping in the Netherlands.
Landscape Planning, 5, 45–62.
Dramstad, W. & Sogge, C. (Eds.), (2003). Agricultural impacts on landscapes: Developing
indicators for policy analysis. Proceedings from the NIJOS/OECD Expert Meeting on Agri-
cultural Landscape Indicators in Oslo, Norway, Oct. 7–9, 2002. NIJOS Report 7, 350 pp.
Dramstad, W.E., Fjellstad, W.J. & Fry, G.L.A., (1998). Landscape Indices – useful tools or
Misleading Numbers? In: Dover, J.W. and Bunce, R.G.H., Key concepts in landscape ecology.
Proceedings of the 1998 European Congress of IALE, IALE (UK) 3- September 1998, 63–68.
References 207

Dramstad, W. E., Tveit, M. S., Fjellstad, W. J., & Fry, G. L. A. (2006). Relationships between
visual landscape preferences and map-based indicators of landscape structure. Landscape and
Urban Planning, 78(4), 465–474.
Forman, R., & Godron, M. (1986). Landscape ecology (619pp). New York: J. Wiley.
Fry, G., Tveit, M. S., Ode, Å., & Velarde, M. D. (2009). The ecology of visual landscapes:
Exploring the conceptual common ground of visual and ecological landscape indicators.
Ecological Indicators, 9(5), 933–947.
Gardner, R.H., & O’Neill, R.V. (1990). Pattern, process and predictability: the use of neutral
models for landscape analysis. In: Turner, M.G., Gardner, R.H. (Eds.), Quantitative methods in
landscape ecology (pp. 289–307). New York: Springer-Verlag.
Gardner, R. H., & Urban, D. L. (2007). Neutral models for testing landscape hypotheses. Journal:
Landscape Ecology, 22(1), 15–29.
Goodwin, B. J., & Fahrig, L. (2002). How does landscape structure influence landscape connec-
tivity? Oikos, 99, 552–570.
Haines-Young, R., & Chopping, M. (1996). Quantifying landscape structure: A review of land-
scape indices and their application to forested landscapes. Progress in Physical Geography., 20
(1), 418–445.
Huff, D., & Geis, I. (1954). How to lie with statistics (142 pp). New York: Norton & Company.
Kilchenmann, A. (1973). Die Merkmalanayse für Nominaldaten: eine Methode zur Analyse von
Qualitativen geographischen Daten. Geoforum, 15, 33–45.
Kindlmann, P., & Burel, F. (2008). Connectivity measures: A review. Landscape Ecology, 23,
Li, H. B., & Wu, J. G. (2004). Use and misuse of landscape indices. Landscape Ecology, 19(4),
Lynch, K. (1960). The image of the City (194 pp). Cambridge/London: The MIT Press.
Mander, Ü., Uuemaa, E., Roosaare, J., Aunap, R., & Antrop, M. (2010). Coherence and fragmen-
tation of landscape patterns as characterized by correlograms: A case study of Estonia.
Landscape and Urban Planning, 94(1), 31–37.
McGarigal, K., Marks, B.J. (1995), “FRAGSTATS: Spatial pattern analysis program for quanti-
fying landscape structure”, General Technical Reports, PNW-GTR-351, Portland (USDA
Forest Service, Pacific Northwest Research Station). Related online version (cited on
24 February 2009).
McGarigal, K., Cushman, S.A., Neel, M.C., Ene, E. (2002), “FRAGSTATS: Spatial pattern
analysis program for categorical maps”, project homepage. Amherst: University of Massa-
chusetts. URL (cited on 24 February 2009).
Metzger, J.-P., & Décamps, H. (1997). The structural connectivity threshold: An hypothesis in
conservation biology at the landscape scale. Acta Oecologica, 18(1), 1–12.
Monmonier, M. (1996). How to lie with maps (2nd ed.187 pp). Chicago: University of Chicago
Phipps, M., (1984). Rural landscape dynamics: The illustration of some concepts. In: Brandt, J.,
Agger, P. (Eds.), Methodology in landscape research and planning. Proceedings of the 1st
international seminar of the international association of landscape ecology, vol. I. Roskilde:
Roskilde University Centre, October 15–19, 1984, pp. 47–54.
Piessens, C. (1985). Studie van de landschapstransparantie in het Land van Waas. De
Aardrijkskunde, 9(2), 155–163.
Riitters, K. H., O’Neill, R. V., Hunsaker, C. T., Wickham, J. D., Yankee, D. H., Timmins, S. P.,
Jones, K. B., & Jackson, B. L. (1995). A factor analysis of landscape pattern and structure
metrics. Landscape Ecology, 10(1), 23–39.
Schreiber, K.-F. (ed.), (1988). Connectivity in landscape ecology. Proceedings 2nd. Intern.Sem. of
IALE, Münster, Sch€ oning, Paderborn, Münsterische Geographische Arbeiten, 29.
ocker, G. and Bergmann, A., (1978). Zwei einfache Modelle zur Quantifizierung der
Beziehungen von Landschaftselementen. In: Richter, H. (Ed.), Leipzig, Georg. Gesellschaft
der DDR, pp. 91–100.
208 8 Analysing Landscape Patterns

Tobler, W. (1970). A computer movie simulating urban growth in the Detroit region. Economic
Geography, 46(2), 234–240.
Turner, M. G. (1990). Spatial and temporal analysis of landscape patterns. Landscape Ecology, 4
(1), 21–30. doi:10.1007/BF02573948.
Unwin, D. (1981). Introductory spatial analysis. London: Methuen.
Uuemaa, E., Antrop, M., Roosaare, J., Marja, R., & Mander, Ü. (2009). Landscape metrics and
indices: An overview of their use in landscape research. Living Reviews in Landscape
Research, 3, 1.; http://landscaperesearch.;
Vogt, P., Ferrari, J. R., Lookingbill, T. R., Gardner, R. H., Riitters, K. H., & Ostapowicz,
K. (2009). Mapping functional connectivity. Ecological Indicators, 9, 64–71.
Wu, J. G. (2004). Effects of changing scale on landscape pattern analysis: Scaling relations.
Landscape Ecology, 19(2), 125–138.
Chapter 9
Building Blocks of the Landscape

Abstract The landscape consists of elements, components and structures forming

characteristic spatial configurations at different scale levels. Many of these are
material, tangible objects we call building blocks and be grouped into a physical
and a cultural system that interact. In a holistic system, both groups are equally
important and are studied according to hierarchically ordered scale levels. For the
purpose of landscape research human settlement in the context of their territorial
sphere of influence are considered the main holistic building blocks of the land-
scape. These are studied at three scale levels: the settlement site as a control centre,
the territory where the land is organized for specific uses and the mosaics formed by
settlement patterns at a geographic scale. The number of natural and cultural
combinations is high, and complex typologies have been conceived for specific
regional conditions. However, many common traits can be found to create generic
typologies. This is illustrated with examples and well-documented case studies
mainly from Europe. Spatial properties, morphology, functions and historicity are
considered universal dimensions, which allow making similar applications else-
where in the world. In a holistic approach, spatial patterns and dynamical processes
in a historical perspective are needed as well. A fundamental difference can be
made between settlements with ancient historical roots and more recently created
settlements, which relates also to a principal distinction between traditional and
new landscapes.

Keywords Natural-cultural system • Settlement • Territory • Mosaic • Land use •

Settlement typology

9.1 A Substrate Carrying Human Life

The landscape consists of elements, components and structures forming character-

istic spatial configurations at different scale levels. Many of these are material,
tangible objects we call building blocks. Building blocks can be grouped into two
main layers: (1) a physical system that forms the substrate of the land and defines
the natural settings, and (2) a cultural system inhabited by people who use and
organize the land and ultimately shape the landscape. Settlement territories

© Springer Science+Business Media B.V. 2017 209

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_9
210 9 Building Blocks of the Landscape

integrated all these features and made the landscape a holistic unit, forming a
building block of the landscape at a regional scale.
The physical, material system characterizes the land. Often the natural compo-
nent is still important. This substrate offers the opportunities and sets the constraints
humans had in organizing the environment into a habitat. Our ancestors possessed a
particular sensitivity, which we have lost now, to select the most appropriate place
to settle and to choose the most efficient land use. The assessment of the local
natural resources was most important and explains the choice of the initial sites of
settlement. Numerous examples show the subtlety of these choices: the proximity
of permanent water resources, avoiding the risk of calamities as flooding or
landslides, the selection of a strategic position to observe the valuable cropland
and cattle, a site that is well exposed to the sunlight and protected from dominating
harsh winds and heavy rains, and a place that offers a variety of natural resources
Each community had its space, and the size, distance and nature of the terrain
determined where to settle and how to organize the land into a territory. Most often
the control centre of the territory was a settlement, whether a town, village or farm.
The territorial space connected all features and made the landscape a holistic unit.
Space was organized into zones of different land use. Successive stages of this
historically evolving organization had their imprints in the landscape and the
landscape itself became an archive of this evolution. However, each settlement
was located in a broader spatial context. It had relations with the adjacent territories
and neighbours, and with larger places, such as towns and cities, and eventually
with regions far away. The geographical situation of a place defined which new or
foreign influences interfered with its development.
Each place or region has a unique past, has its own history, which is the result of
continuous, local adaptations to the environment and an evolving society. These
changes were culture-bound and depended on customs and the available technology
in a given period. All successive cultures and technologies left their mark,
transforming the older landscape in various degrees or even wiping it away. This
can clearly be observed by the successive, often superimposed transportation
networks. Many innovations and new landscapes were caused by economic
necessities and demographic changes. The general growth of the population
necessitated the settlements and territories to expand, often resulting in conflicts
with neighbours and the need to have strict rules according to the ownership and the
rights of using the land. The landscape, in its etymological territorial meaning, was
a spatial unit of organization with specific rules and rights. Increasing population
also caused densification of settlements and the reclamation of wasteland and new
land. Common land became privatized and agricultural innovations such as new
crops or new technology intensified the land use.
9.1 A Substrate Carrying Human Life 211

9.1.1 The Terrain as Foundation: The Natural Settings

The earth’s surface is a component of the landscape, but simultaneously the

‘carrier’ of many other components such as vegetation, land use and settlements.
In times of limited technology, humans had to adapt more to the natural settings
than they could manipulate and change them. Vital natural resources such as water
supply, soils, sources of raw materials and energy determined the way humans
behaved. More workers and increasingly more powerful technology made it grad-
ually possible to change the terrain itself. Mastering and transforming entirely the
earth’s surface became possible with mechanical excavators and explosives.
Different aspects of the natural settings are studied by different disciplines and
their findings contain significant information to understand the development of
landscapes and their functioning (Fig. 9.1). The Main Natural Building Blocks

The substrate (geology and soils) and the climate are the fundamental building
blocks that shape the land. They act at continental and global scales. Geology
describes the nature and age of rock formations and deposits, modelled by tectonics
forces and weathered and eroded by processes often dependent on the climate.


Geology - tectnonics
- lithology
- structure
- stratigraphy
Soils - parent material, texture,, substrate, stoniness...
- natural drainage
- soil type, profile development, depth,...

Slopes - steepness
- aspect (orientation) Land cover
- shape Vegetation
Particular landforms

Surface water
Ground water
Climate zone

Fig. 9.1 The physical-geographical building blocks of the landscape

212 9 Building Blocks of the Landscape

Geomorphology describes the various landforms and explains their genesis and the
processes involved. Geomorphological processes act differently on different sub-
strates according to the climatic conditions. For example, calcareous rocks will
dissolve and erode rapidly in warm and humid conditions, while they are very
resistant in cold and dry conditions. Understanding landscape genesis needs under-
standing the climatic changes in the past as well as the geomorphological processes
involved. Generally, the Pleistocene events were most important in modelling the
actual terrain. All these processes interact over long periods of time, although
sudden catastrophic events, such as earthquakes, tsunamis, volcanic eruptions,
floods and landslides occurred, separated by long periods of relative stability
(Rackham and Moody 2012, Fig. 9.2).
The soil is the top layer of the earth’s crust and is composed of loose mineral
particles, organic matter, water, air and living organisms and is where plants root.
The soil is the end product of a weathering process where the nature of the parent
material, climate (temperature and precipitation), relief (slope), organisms (fauna
and flora), human activities and time (duration) are important interacting factors.
The soil body is characterised by the chemical composition and size (texture) of
the minerals, by its colour and structure, natural drainage capacity, profile (soil
horizons) and parent material. These properties are used in soil classification, which
is mainly done from an agricultural perspective. Internationally most notorious is

Fig. 9.2 The landscapes of the Condroz, Famenne and Ardennes regions (Southern Belgium) are
largely determined by the geological substrate of this part of the eroded and uplifted Hercynian
mountains. Sediments from the Lower Devonian form an anticline of resistant rocks, mainly
quartzite (q). The steep slopes at the northern border remained forested and on the flat top zone,
small isolated agrarian settlements developed. North of the anticline, a large synclinorium
stretches out, consisting of smaller synclines and anticlines and exposing Middle Devonian
rocks of different resistance. From south to north (right to left in the figure) following formations
make the substrate: (kf) layers of limestone and schist forming a elongated plateau were most
villages developed; (k1) reef limestone forming wooded buttes of the cave country of the
Calestienne; ( fg) a broad zone of soft schist that eroded deeply and where the main rivers converge
in broad flood plains and where gravel banks (g) witness of successive stages of incision; next a
broad, mainly forested zone (pf) of resistant psammite (micaceous sandstone) ( p) and deeply
incised schist (f); and next a series of synclines in limestone (k2) alternating with anticlines in
psammite ( p) and covered with fertile aeolian loamy soils. Each of these natural regions was
occupied by humans in a different way, characterised by different building materials of the houses,
different settings and types of the villages, and different land use, thus forming landscapes with a
distinct identity
9.1 A Substrate Carrying Human Life 213

the USDA Soil Taxonomy (USDA 2010). Nevertheless, many national and regional
classification systems also exist, using local and common names for the different
soil types, which are often easier to understand by non-experts.
The soil is a complex non-renewable resource that performs many vital functions
(services), such as biomass and food production, transforming minerals and organic
matter; it is a habitat and gene pool, provides raw materials and contains essential
information to study the archaeology and history of the landscape (Fig. 9.3).
Water is essential for life and thus is an important factor in understanding the
location of human settlements. There is a need for a sufficient and permanent supply
of drinking water and its quality is essential for health care. Simultaneously the site
needs to be safe from flooding.
Waterways of various kinds dissect the terrain and combine multiple functions
such as barriers, natural borders, but are also corridors and routes. The hydrography
is strategically and economically important.

Fig. 9.3 Landform and soil conditions determined the settlement and land occupancy of the karst
polje of Didima (Argolid, Greece). The polje developed by dissolution of the limestone resulting in
a deep basin with heavy clay soils (vertisol, ‘terra rossa’) in the flat bottom, which can flood
seasonally. The steep mountain slopes are covered with pediments and colluvial fans, which
became terraced for a combined culture of permanent olive groves, wheat production and grazing
the fallow land in winter. The mountains are covered with sparse garrigue vegetation (phrygana)
and used as extensive grazing land. Some sinkholes occur at the border of the polje basin. The
whole territory has only one village in its centre, situated at the border of the clays soils of the basin
floor. Land use shows a circular zoning of the land use, determined by the natural conditions and
with decreasing land use intensity with distance from the village. The sinkholes were used as
settlement sites in prehistoric times and still have an important religious meaning. They augment
the local identity of the place. (© Photo M.Antrop 1976)
214 9 Building Blocks of the Landscape

People living along the shores of lakes or in coastal zones live at the edge of two
completely different resources spaces and adapt their way of living and create
adapted settlement forms.
Climate is determinant for geomorphological processes, availability of water
and vegetation zoning. Climate also determines evapotranspiration and thus salini-
zation, desertification and irrigation. It also defines the nature of the water flows and
the occurrence of snow and ice, which create distinct landscapes. Climate acts at a
global scale but also makes the variation in the landscape at regional and local
scales (topo-climate) and even on the scale of small habitats and sites (micro-
The physical settings and the climate determine the natural vegetation that can
develop and characterise the landscape further. Land cover is a generic term
referring to the vegetation formations and plant species, or the absence of vegeta-
tion. Land use refers to the particular ways humans use the land and will be
discussed later in detail. A broad-leaved forest can be a natural reserve, or a
plantation for economic timber production or a recreation area, and sometimes a
combination of different uses altogether. Land cover is commonly registered by
means of remote sensing, but defining land use demands additional data such local
customs in exploitation, economic choices, etc. Different land use practices created
different landscapes in a similar land cover type. Natural Land Units

The interaction between climate, geology, geomorphology, soil conditions and the
related land cover creates distinct land units, characterised by unique vertical,
ecological relations between the components. These are often called natural land
units or natural regions. Different types of land units form spatial patterns
organised in a hierarchical way according to scale. They define the systems thinking
and the holon-structure of the landscape.
Different types of land units can easily be recognised in the field, in particular
from a bird’s eye perspective as offered by aerial photographs. Identifying land
units by photo interpretation is a common practice in land classification and
evaluation and allows a quick mapping of the landscape. Most methods for land-
scape classification are, therefore, entirely holistic (Zonneveld 1995; Antrop and
Van Eetvelde 2000) (see Chap. 10).

9.1.2 Cultural Building Blocks: The Human Impact Cultural Holons

Most landscapes we experience today are the result of remodelling the land by
humans during many centuries. Each phase of remodelling superimposes over the
9.1 A Substrate Carrying Human Life 215

older ones, sometimes wiping them out, more likely transforming them partially.
Consequently, some like to compare the landscape with a palimpsest, which needs
to be decrypted to be understood (Claval 2004; Bloemers et al. 2010). To under-
stand the contemporary landscape structure, we need knowledge about the pro-
cesses and tools humans used during successive historical periods to organise
their land.
The human building blocks of the landscape vary over several scale levels, from
local objects to regional and sometimes interregional structures. They consist of
material objects, such as buildings, constructions and fields, which often carry
symbolic meanings, such as signs referring to property and land use rights, and
place and field names.
Settlements and territories should be regarded as the primary building blocks of
the cultural landscape. They have been extensively studied by archaeologists,
historical geographers and cultural geographers, such as Aston (2004), Roberts
and Wrathmell (2000), Muir (1999), Roberts (1987), Lebeau (1979), Haggett et al.
(1977), Aston and Rowley (1974), Uhlig and Lienau (1972), Chisholm (1962) and
many others. Their research includes many case studies, the development of
typologies and the study of the historical development of settlement forms
(Table 9.1).
The easiest approach to read the cultural landscape is to start with the initial
settlements from which the landscape was organised. The term settlement is used
here to denote any kind of settlement regardless of its size or population. Settle-
ments vary from individual farms, over hamlets and villages to towns and urbanised
places. Each settlement has a territory forming a distinct spatial unit that can be
regarded as a holon embedded in a hierarchical structured spatial scale and evolving
through time. Townships were the building blocks of most of the emerging cultural
the landscapes (Muir 2000). Different types of settlements and different models of
land use systems and territorial organisation developed, as well as different trajec-
tories through time.

Table 9.1 The human • Local: settlements as territorial centres of control

building blocks of the – Site, situation and territory
landscape – Settlement types
– Land use and land cover
– Field systems
– Small elements: hedges, pools, etc.
– Landmarks
• Regional and interregional
– Settlement patterns
– Territorial models
– Boundaries
– Transport and communication
– Agrarian systems
• Symbols, place names (toponyms), narratives
216 9 Building Blocks of the Landscape

The settlement-territory holons are the places where people of a community live
and organise their land and shape the landscape. Two main types of settlements can
be recognised considering the spatial field of influence of the place. The simplest
form is a territory-based settlement that subsists mainly on what the surrounding
land provides. More complex forms are settlements that are the focal centre in a
wider area, controlling smaller territory-based settlements that support these ‘cen-
tral places’. Their field of influence works at a higher hierarchical scale. Three Scales

An appropriate method of studying the cultural landscape starts from a detailed

analysis of one settlement and step-wise widens the scale to explore the broader
spatial context (Fig. 9.4). At each scale level, different properties are studied using
specific methods.

9.2 People Inhabit the Landscape: Settlements

Most early settlements were based on a self-sufficient economy based on agricul-

ture, husbandry, fishing or combinations. The natural resources available in the
territory of the settlement could lead to specialisation, which offered opportunities

Fig. 9.4 Settlement analysis at three scale levels

9.2 People Inhabit the Landscape: Settlements 217

in trading with other specialised settlements nearby. Specialisation in trading,

industry and military defence demanded particular site and situation properties.
Important site properties (intrinsic land qualities) to consider are:
• topography, relief: suitable for building; prospect and refuge
• permanent supply of drinking water
• no risk of flooding
• a microclimate offering natural shelter
• controlled accessibility
Properties of the situation (extrinsic to the site) are:
• a maximal diversity of natural resources nearby
– soils suitable for agriculture
– pasture land
– fuel sources (wood, peat. . .)
– building materials (wood, stone, minerals. . .)
• maximal control over the territory (often visual)
• assessment of risks
• spare space to expand
• accessibility (time, passable terrain)
The choice of the site for a new settlement did not depend only on the available
land qualities, but also on the social context and constraints. Other settlements had
already land rights and newcomers had to choose ‘free’ space and agreements had
to be made about the use of the common borderland. Understanding a site neces-
sitates knowledge about the situation as well.
Initial territory-based settlements lived from the land and their location was
often chosen to have a wide diversity of natural resources nearby. Depending on
whether the focus was on agriculture, husbandry, forestry or extraction, the land
qualities where assessed in different ways and the site selection varied accordingly.
The oldest settlements formed isolated agrarian ‘islands’ in a vast wilderness
(Fig. 9.5). When the spatial distribution of the land qualities was isotropic, circular
territories developed and the land use was organised in concentric belts according to
the distance to the settlement and the intensity needed to work the land. The
wilderness was the borderland between adjacent settlements. When the population
grew, territories expanded, reclaiming the wilderness. This expansion and new
additional settlements a retreat of the wilderness until the reclaimed land reached
that of neighbouring settlements. Instead of ‘natural’ borders that delimited a
territory, social and legal borders had to be agreed upon, sometimes after long
conflicts. The location of the settlement within its territory, the spatial organisation
of the land use in the territory, its shape and nature of the borders are all charac-
teristics of the settlement-territory model that can be considered a holon.
The focal centres are characterised by a spatial field containing several terri-
tories, forming not necessarily one compact area. This spatial field of influence is
also called hinterland. Focal centres provide services to all sub-ordered settlements.
218 9 Building Blocks of the Landscape

Fig. 9.5 Evolution of settlements and territories. (a) The initial settlement of the area consisted of
small groups of people who cleared the wilderness around the settlement. The best and most easy
to clear soils were selected and the settlement was placed in its centre. (b) The growing population
demanded the expansion of the infields and outfields; when the territories of neighbouring villages
touched each other, border conflicts needed to be settled and the use of the common wilderness
needed to be regulated. (c) Ultimately all land was developed, including marginal soils, and
wilderness and woodland were reduced to small patches. Some places became more powerful
and took the land of smaller places. (d) When population decreased or activities concentrated in
towns, the land use in marginal and poorly accessible areas became extensive and finally, the land
was abandoned; wilderness and woodland expanded (light grey). (e) Continuing marginalising
leads to abandonment of the villages. The neighbouring villages absorbed their territories

Thus, broader diversification of activities, exploitation of natural resources

becomes possible, beneficial to the whole hinterland. Focal centres became towns
and cities with a proper specialisation. Besides their relation to the sub-ordered
settlements, focal centres also developed economic, commercial and political
relations with more distant focal centres, establishing a hierarchy of central places
at different scales, which are all interdependent.

9.2.1 The Settlement Site

The concept site refers to properties of the place where a settlement was located at
its origin. The situation refers to the properties of the surrounding land in a broader
geographical context, as well as the position of the place in relation to other
settlements in the neighbourhood and to larger urban centres, which have an
influence on the place.
The founders of a settlement assessed carefully the local land qualities before
selecting the site of their settlement and considered simultaneously the potential of
the geographical situation, in particular regarding natural resources. Site Evaluation Considering Land Qualities

The general assumption is that the site of a settlement is the most suitable place
fulfilling all expectations and needs of the community. It was selected after a
careful evaluation of all significant land qualities. Chisholm (1962) developed a
cost-benefit model to assess the suitability of a site for an agrarian settlement
considering five primary resources: availability of arable land and grazing land,
9.2 People Inhabit the Landscape: Settlements 219

Fig. 9.6 A cost-benefit

assessment of a site
selection as a function of
distance to and the
importance of primary
natural resources (After
Chisholm 1962, Table 9.2)

Table 9.2 Cost-benefit analysis of two hypothetical sites

B: distance to
site in km BC
A: natural resource Site X Site Y C: relative ‘cost’-units/km to site Site X Site Y
Water 0.1 0.5 10 1.0 5.0
Arable land 2.0 1.0 5 10.0 5.0
Grazing land 2.5 1.5 3 7.5 4.5
Fuel source 2.5 2.0 3 7.5 6.0
Building materials 3.0 2.0 1 3.0 2.0
Sum 29.0 22.5
After Chisholm (1962)

water supply, availability of fuel resources (wood, peat) and building materials
(wood, stone). Figure 9.6 and Table 9.2 show a hypothetical example to illustrate
the method to define the most optimal location (site X or Y) considering the cost-
distance to the natural resources. In the example, the cost-distance per kilometre to
the daily water supply is estimated ten times higher than the cost to get occasionally
building materials. The table gives the cost-benefit calculation for both sites X
and Y, showing that site Y has a better location to get the necessary resources.
Another method to assess the location of a settlement site is by making overlays
of maps representing the suitability of significant land qualities. The suitability can
be expressed in different ways (ranking, scores, etc.) and according to the level of
measurement, these can be combined in different ways to obtain a synthetic
evaluation. Site and Microclimate

Besides the influence of the macro-climate the region, also topo- and micro-climate
affect the precise location of a settlement and the way it is constructed.
220 9 Building Blocks of the Landscape

In general, a compact building will have a relatively small exposed surface,

which can offer advantages for controlling insulation all year round. Site evaluation
considers the orientation to solar illumination and dominating winds during differ-
ent seasons. The position and size of openings, shelter facilities, the thickness of the
walls and building materials are choices the architect has to make accordingly.
Also, the surrounding settings are important such as courtyards, the screening with
walls, hedges and trees.
Following factors are important to consider:
• Insolation: daily and seasonal sun paths depend on the latitude of the site, and
define the duration of daylight and the heating of the site;
• Wind conditions: includes direction and speed of (dominant) winds, but is also
influenced by the presence of obstructions, such as vegetation screens or stone-
walls or cliffs;
• Vegetation (land cover): species and formations (size, shape and height), spatial
configuration (woods, tree screens and hedgerows);
• Lithology and soils: parent material and texture in relation to thermal properties,
soil drainage, run-off patterns;
• Water: water bodies (ponds, lakes, rivers), groundwater (depth).
Micro-climate is the generic term referring to local climatic conditions gener-
ally. Topo-climate refers to climatic conditions in a landscape determined by its
topography. It combines effects of insolation, wind, elevation, substrate and geo-
morphology. Topography is the main factor causing local winds such as the sea
breeze (day) and land breeze (night) along coasts. Similarly, mountain and valley
winds alternate daily. The air moves upwards during the day when slopes are
intensively heated by the sun, and downwards into the valley during the night
when the upper slopes rapidly cool. Another local wind type is the katabatic wind
where dense cold air flows from higher to lower regions under the influence of
gravity. Topography and exposure define also the regional winds such as the bora,
the mistral, the f€
ohn and Chinook.
In effects of topo-climate and micro-climate is clearly visible in traditional rural
buildings that are not yet altered by modern insulation techniques. Climate has
become an important topic in urban ecology and urban heat islands affect the health
of millions of people (Clergeau 2007; Gaston 2010; Niemelä et al. 2011; Figs. 9.7
and 9.8).

9.2.2 Principles for a Settlement Typology

Although each settlement is unique and the variety of settlement forms seems
infinite, they have many traits in common and can be grouped in generic typologies.
Essentially it comes to the spatial arrangement of spaces with different status such
as public or private, and functions such as dwelling (living or working) or travel-
ling. Edges (walls, fences, hedges, etc.) form enclosures and gates mark the
9.2 People Inhabit the Landscape: Settlements 221

Fig. 9.7 Slope exposition defines several qualities for a place to settle and to choose the
appropriate land use. An example: zone 1: protected from dominated northern winds, humid
during the day, dry during the night; zone 2: warm during the night, very warm during the day;
zone 3: cold and wet during the night, warm and dry during the day; zone 4: shadow slope: cooler
and more humid (After Roberts 1987)

Fig. 9.8 Orientation and architecture adapted to micro-climatic conditions: an example of a mas
in the Provence (France). (a) search for sunlight and warmth in winter, (b) searching shadows in
summer, (c) protection against cold winds, (d) protection against the dominant direction of rain
showers (After Massot 1975)

interfaces between these different spaces. They vary in size and shape and can be
hierarchically embedded. Figure 9.9 illustrates a generic model. Typical private
spaces to travel are turnpike roads or waterways with a regulated right of way. They
have a characteristic infrastructure such as tollgates, tree rows and towpaths.
Table 9.3 gives criteria commonly used in settlement typologies. Settlement
types can be ordered according to universal dimensions referring to spatial
222 9 Building Blocks of the Landscape

Fig. 9.9 Type of spaces

forming the building blocks
of a settlement

Table 9.3 Criteria used in settlement typologies

• Morphological criteria
– Scale: single household or farm, hamlet, village, town, city
– Ground plan: regular – irregular, simple – complex, clustered – dispersed
– Size and shape of the common public place(s) if any
– Shape and borders (edges) – access and gates
– Architecture
• Functional criteria
– Site of the settlement – adaptation to physical environment
– Geographical situation – position in settlement network
– Occupancy (temporal, permanent)
– Functions and services: social (juridical, political, religious, communication, military. . .) and
economic (market, industry) and associated characteristic elements
• Historical criteria
– Age, origin of settlement
– Settlement continuity – time layers
– Transformation phases
– Symbolic meanings
• Place names
9.2 People Inhabit the Landscape: Settlements 223

properties, morphology, functions and historicity. The relations between these

dimensions are fundamentally different for settlements with ancient historical
roots and more recent created settlements. This difference characterises the
distinction between traditional landscape and new landscapes as discussed in
Chap. 7. Settlements with Ancient Historical Roots

Most settlements were initially villages and hamlets with a local self-sufficient
economy based on a pastoral and/or agrarian lifestyle. However, they were also
embedded in a larger geographical space and formed networks with neighbouring
settlements. Trade and political power rapidly formed a hierarchical structure
where some places could benefit from their situation, growing and developing
more rapidly and getting control over vast areas. Towns and cities emerged and
benefited from the agglomerated economy of the nearness of diverse activities and
crafts, and so became centres of innovation. Settlement types not only vary in size
but also with cultural, social and economic factors and the historical period of their
origin and development. Similar types are often culture and time related, and
witness of the available technology.
Figure 9.10 gives an example of an ordering of settlement forms common in the
European countryside according to their spatial dispersion and degree of arrange-
ment and agglomeration, their morphology and historicity (age, origin). Figure 9.11
illustrates schematically some typical rural settlement forms and Fig. 9.12 shows
some typical defensive settlement forms in Europe. Similar typologies can be made
elsewhere in the world.
The first dimension for the classification is the spatial dispersion and degree of
the arrangement of the settlement elements. This is also related to their size, shape
and function. On one end of this dimension are the solitary settlement sites varying
from small individual farms to larger estates as manors and abbeys. The isolated
situation often demanded measures of defence. The site is carefully chosen con-
sidering the land qualities and opportunities, necessary architectonic elements are
added. On the other end are agglomerations at the size of a town and city, fully
adapted for specific functions of a central place (Fig. 9.13).
The main groups based on the morphology of the site are based on the shape of
the ground plan and the occurrence of typical elements. For most of the rural
settlements in the countryside, three main groups can be recognized: (1) places
with a common central space, such as a green or square, (2) linear settlements
arranged in rows along roads, waterways or edges, and (3) clustered settlements
where the elements are agglomerated in a varying degree of regularity. Each type is
associated with a period of origin, reflecting the then existing societal rules, the
cultural style, the dominant economic activities and the available technology. The
distinction between a hamlet, town and city is not only dependent on its population
size, but also on the existence of certain services, functions and rights. Changes
during history caused by growth and decline are also reflected in the morphology of
224 9 Building Blocks of the Landscape

Ancient historical roots Recent
Solitary or

moated site pseudo-farms

isolated manor estate neo-style manors
moat and bailey
settlements industrial sites
(water)mills, forges
enclosed mining sites
abbey, cloister, priory

irregular cluster
nebular settlement

hilltop settlement
cross-roads cluster
garden city/district
castle/manor-church-village association
new town
leisure resorts
border crossing tourist resorts
port compound

Row or lineair street village

settlements land reclamation settlement
peat extraction
marshland reclamation ribbon settlement
along dyke or embankement (caused by
forest or wasteland reclamation urban sprawl)

with central
place or space
simple green village
walled settlement ('burg', 'bastide')


composite polyfocal, multiple spaces
bastioned place

Fig. 9.10 Example of a classification of rural settlement types in Europe, using dimensions as
spatial dispersion, the degree of arrangement and historicity

the place (Knaepen and Antrop 2003). An example of such a historical evolution
typical in Europe is given for a green village in Fig. 9.14.
Highly specialised places such as defence sites, border settlements and econom-
ical gates have an adapted morphology reflecting the main activities and functions
of the place. They often have an important cultural and heritage meaning
(Fig. 9.12).
Settlements can also be classified according to their age of origin and the
continuity of their existence. Places with ancient roots often witness of their
9.2 People Inhabit the Landscape: Settlements 225

Fig. 9.11 Schematic representation of some typical rural settlement types in Europe. 1 garden/toft
and house, 2 road, street, 3 field border, 4 common open space (green, square), 5 pond, 6 tree row,
hedge, 7 church, chapel; cross, monument, 8 fortification, 9 dyke, 10 waterway, canal (c), river,
coast (After Antrop 2007; Roberts and Wrathmell 2000; Lebeau 1979; Uhlig and Lienau 1972)
226 9 Building Blocks of the Landscape

Fig. 9.12 Schematic

representation of some
typical defensive settlement
types. Individual farms or
households use terrain
qualities as high grounds
and the superficial water
table to dig ditches.
Eventually, artificial buttes,
like a ‘terp’ or ‘motte’, were
erected. Stockades and
enclosures completed the
defence. Also,
agglomerated settlements
looked for natural defence
conditions as hilltops, cliffs,
islands or riverbanks. When
these were missing,
artificial defences were
constructed such as
stonewalls around newly
designed villages or
compounds, or fortifications
around existing settlements,
such as the Vauban-
fortifications. (a. river, b.
natural high ground, c.
lowland, marshland with a
risk of flooding, d. natural or
artificial caves as shelter, e.
flood level)
9.2 People Inhabit the Landscape: Settlements 227

Fig. 9.13 Examples of

settlement forms with
ancient historical roots: (1)
an isolated forest
reclamation farm (G:
Einzehof): Ferme du Bestin
in the Ardennes (Belgium)
(© Photo Department of
Geography, Ghent
University 1970), (2) a
proto-industrial green. The
village of Doornzele (1753
in.) has an exceptional
green with a length of
1500 m (© Photo J. Semey
1988), (3) a cluster village
in a polje basin: Didima
(Argolid, Greece) (© Photo
M. Antrop 1970), (4) The
medieval harbour Damme
(near Bruges) became a
bastioned fortification
during the religious wars
(© Photo J. Semey 1989)
228 9 Building Blocks of the Landscape

Fig. 9.14 Examples of paths for the evolution of rural village plans in Europe. Stages: (A) remote
antecedent, initial settlement, (B) developed antecedent form, (C) mature traditional village form
before industrial changes and urbanisation, (D) preserved historic form at present, (E) shrinkage of
village, (F) abandonment and deserted or destructed village, (G) expanded village by
industrialisation and urbanisation. Common space: (a) green, (b) place-market, (c) moat and
bailey, castle, (d ) manor
9.2 People Inhabit the Landscape: Settlements 229

historicity in the morphology of the ground plan and the heritage of ancient
elements and typical architecture. However, abstract morphological types are not
reliant criteria for assessing historicity. Many neo-styles emerged since the nine-
teenth century and important changes occurred in the functions of the old settlement
elements. For example, former farms transformed into residential houses or places
with services for leisure and tourism. Settlements of an Urbanised Network Society

Since the Industrial Age and processes induced by urbanisation (see Chap. 8), many
historical settlements were profoundly transformed and new kinds of settlements
emerged in the countryside (Table 9.4).

Table 9.4 Common main types and characteristics of modern residential urbanisation
Type Characteristics
Streets in the rural a street formed by several dozen of houses located in the middle op
(open) agrarian land with no connection to a village centre; often a
uniform architecture
Ribbon-building Open residential housing in a large variety of architecture along
existing roads on small garden lots with no historic connection to the
rural land. Mainly the result of (unplanned or uncontrolled) urban
Suburbs in former Individual housing in a large variety of architecture in open garden
estate parks setting, embedded in tall trees and ponds of former estate park.
Suburbs on openfields Individual open housing in a regular arrangement in similar or varied
architecture and situated on former open field agricultural land. Often
bordered by green screens from the adjacent agricultural land.
Social housing Regular arrangement of similar small houses in open garden setting.
New allotment Small lots that can be rent temporarily for use of vegetable and fruit
gardens gardens. These new allotment gardens situated on unused land at the
edges of urban or industrial development zones or along transport
infrastructure. Although initially no permanent residences are allowed,
many developed into weekend houses and flower gardens were created.
Two groups can be recognised: The ones developed from individual
initiatives, and the ones planned by the landowner.
Garden district A block of residential houses with uniform architecture characterised
by a small front garden; conceived following the garden city ideal.
Urbanisation in natu- Individual large houses in open setting in dunes, woodland or along
ral areas beaches.
Enclosed compounds Enclosed compounds are a response to a demand for better safety
control of residential and leisure areas. They are sometimes vast areas
located in the rural countryside and offer various services within the
compound to mostly elitist inhabitants working in a nearby city. The
compounds are often disconnected from local services and social life.
Also, the infrastructure and architecture are completely different from
the surrounding traditional styles.
230 9 Building Blocks of the Landscape

One of the early measures to reduce the migration from the rural countryside to
the industrialising cities was to create cheap workers housing in the countryside.
These ‘streets in the rural’ are often spatially disconnected from existing villages
and appear as alien elements in the countryside (Fig. 9.15-1). Labour was found
locally in farms and local industries and sometimes transportation was guaranteed
by local rail or tramway connections.
When the individual mobility developed with the use of cars, urban sprawl
resulted in ribbon-building and the creation of suburbs is various forms. Ribbon-
building consists mainly of the open residential housing along existing roads on
small garden lots with no functional connection to the adjacent rural land (Fig. 9.15-
2). This makes this settlement form, as well as the type of streets in the rural,
different from the historic row settlements, which were typical for planned land
The early suburbs emerged in the countryside close to the expanding cities.
Estate parks became open residential suburbs, offering additional income for the
landowner of the estate (Fig. 9.15-4). In a next phase, suburbs developed on the
well-drained soils of the arable open field, creating hard interfaces between the
open and build landscape (Fig. 9.15-3). Garden-city style suburbs were created as
well (Fig. 9.15-5 and 6).
Historical settlements also expanded their residential area. Social housing near
the village centre helped to densify the population, providing new better quality
housing than the older existing houses (Fig. 9.15-7).
At the edges of the cities and industrial areas, new allotment gardens were
created. Here, the labour population could rent cheaply a small lot to produce
their own vegetables and fruits, which allowed to keep the wages low and prevented
the population from starvation. These are often located on unused and even derelict
land along road and railway infrastructure (Fig. 9.15-8).
At the end of the nineteenth century, Edenezer Howard published his urban
planning principles for an ideal garden city, which initiated the garden-city move-
ment. (Howard 1898, 1902). A garden city according to Howard should be planned
from scratch as a self-contained and economically self-sufficient place integrating
all necessary functions. Garden cities should be located at short commuting dis-
tances of a large city to which they are linked as a satellite but separated from it by a
greenbelt. It was a way to contain wild urban sprawl and to provide decent housing
for the fast growing urban population. The ideal plan of a garden city was circular,
with a concentric and radial pattern of streets, connecting open public green spaces
and parks.
The idea was adopted in England after the Second World War with the creation
of New Towns. The garden city philosophy was adapted in many ways. Many
garden suburbs were created, not necessarily compliant to Howard’s initial ideas.
The style became popular in the creation of cheap social housing suburbs. It
inspired also the concept of ecopolis and the lobe-city model (Tsjallingii 1995,
Fig. 9.15 Examples of housing related to urbanisation of the countryside and urban sprawl. (1) a
street settlement in the rural, (2) (open) ribbon-building, (3) the front of a suburb in an open field,
(4) a suburb allotment in an estate park of a manor, (5) a garden city for staff members of an factory
Herryville (Ghent), (6) workers housing in garden city-style (Belzele), (7) new social housing, (8)
new allotment gardens along the terminal of the Eurotunnel at the urban fringe of Calais (France)
(© Photos M.Antrop 2006)
232 9 Building Blocks of the Landscape

Fig. 9.16 Example of a new residential compound combined with a country club and golf. The
Belas Clube Campo (a) was created from scratch in 1991 in the rural countryside. The building
rules allow only modern architecture and any reference to traditional styles is prohibited. It covers
500 ha and is located 20 km north-west from Lisbon, and is larger than the historical centre of the
capital (b). (Google Earth 2007)

Enclosed compounds are a response to a demand for better safety control of

residential and leisure areas. They are sometimes vast areas located in the rural
countryside and offer various services within the compound to mostly elitist
inhabitants working in a nearby city (Fig. 9.16). The compounds are often discon-
nected from local services and social life. Also, the infrastructure and architecture
are completely different from the surrounding traditional styles.

9.3 People Use the Land

9.3.1 Organizing Territory and Landscape

Settlements are the control centres from where people use the surrounding land,
organize their territory and shape the landscape. This is intensively studied by
anthropologists, historical geographers and archaeologists. Historical geographers
9.3 People Use the Land 233

defined many of the basic concepts in the study of the development of the cultural
landscapes in Europe. Settlement and Land Use Zoning

Early agrarian settlements were land-bound and self-subsisting. In organizing the

land use around the settlement, distance was an important factor and related to the
intensity of the daily care for the land. When the land qualities around the settle-
ment were more or less homogeneous (isotropic), often a concentric zoning of the
land used was created. Land closest to the settlement demanded most care and was
intensively used; land far away was used more extensively. Around a village, a
fringe was created of gardens, tofts, orchards combined with grazing. Around this
fringe a zone of arable cropland was lead, very often with open fields divided in
tracts. Even further, near the periphery of the territory, land was more extensively
used as common grazing land, often extending in the surrounding wilderness or
forest. The Romans denoted these zones as hortus, ager, saltus and silva. Historical
geographers refer to infields (the intensively and most valuable land) and outfields
(temporary fields and grazing land) and inmark and utmark in Scandinavian lan-
guages. Part of the area may be called commons and indicate land where the local
community has specific use rights. Sometimes infields are also referred to as
townfields in English, in French one speaks of finage, in German of Flur.
The hortus was often also enclosed from the land outside by hedgerows or
stonewalls, or sheltered by banks planted with trees. In Mediterranean areas, the
hortus was often irrigated. The hortus causes villages to become ‘disguised in
green’ in the landscape.
The ager was made on the most fertile and easy workable soils. Initially the
whole community worked the land together and ley farming was a common
practice. The land was divided in separate tracts allowing crop rotation to avoid
soil deterioration. Land division and distribution of the fields was regulated in such
a manner that all farmers had land of equal qualities regarding soil, slope, and even
sun division (solskifte) was applied so all had the same exposure (Muir 2000).
In Europe, ley farming practices resulted in different crop rotation systems.
Under a two-field rotation, half the land was planted with crops while the other
half was left fallow or under pasture allowing the soil to recover. Under three-field
rotation, the land was divided into three sections: one was planted in the autumn
with winter wheat or rye, the next was planted with peas, lentils or beans, and the
third section was fallow. In the tenth century there was a transition from a two-field
to a three-field crop rotation to enable an increased food production for the fast-
growing population. Early the sixteenth century, a four-field rotation system was
introduced in Flanders (Belgium), and was popularized by the British agriculturist
Charles Townshend in the eighteenth century. It consisted of a rotation between
wheat, turnips, barley and clover. The introduction of a fodder and grazing crop in
the rotation allowed livestock to be bred year-round near the village, allowing
clearing the outfields for more cropland. The system became a key development in
234 9 Building Blocks of the Landscape

the British Agricultural Revolution and initiated the Enclosure Movement. The
crop rotation systems became obsolete with the introduction of new, exotic crops,
artificial fertilizers and mechanization during the Agricultural Revolution in the
second half of the nineteenth century.
Dryland farming is uniquely dependent on natural rainfall. Moisture is captured
until the crops can use it. A common practice is summer fallow rotation in which
one crop is grown during the wet season and the land is left fallow or used as pasture
during the dry season. The system is vulnerable to soil erosion by runoff and dust
storms especially when vegetation cover does not protect the soil. Measures
preventing soil erosion on hilly terrain is contour ploughing, building talus and
terracing fields.
Ley- and dry-farming resulted in forms of multifunctional land use. Already the
earliest Roman farmers planted a multi-storey canopy of olives, grapes, cereals, and
fodder crops referred to as cultura promiscua. It is a way of multifunctional land use
typical for the Mediterranean poly-culture. It includes also techniques such as
summer fallow rotation and using the fields then for grazing. These practices
resulted in typical traditional landscapes such as the montado and dehesa (Pinto-
Correia and Vos 2004).
Less fertile soils and land at a greater distance from the settlement formed the
saltus or outfields. It was more extensively used, mostly as common grazing land, or
temporary cropland in times of need. Extensive grazing located far from the
settlement could be common land to several settlements, which had special use
rights. This was the case with mountain summer pastures (alpages) and coastal
marshlands. In English, the saltus is often referred to as commons. Old Flemish
words for this type of land are veld and wastine (related to ‘wilderness’) and in
French p^ atures-sarts and p^ aturages. The saltus became cultivated when the
population grew and new cropland was needed. Often new settlements were
established in the periphery of the primal village territory. The reclamation of the
land could be planned systematically, creating new villages. Scattered farms of
different sizes could also reclaim tracts of land. Initiatives often came from
powerful landowners such as abbeys. This was the case for large land reclamation
projects. The saltus became enclosed land and new distinct landscape types were
In the concentric land-use zoning model, cultivated land and outfields were
surrounded by the silva; woods and forests, which initially formed a vast wilderness
between the settlements. The adjacent communities used these in various ways,
which is sometimes indicated by the names used to denote different parts of the
woodland (Rackham 1990, 2003). The silva communis was the part that could be
used by the villagers; the silva palaria was used for small wood production (French:
bois a pieux) and consisted of coppice woodland (French: taillis) (Larrère and
Nougarède 1993). The silva concide was the woodland used for charcoal produc-
tion where charcoal burners lived most of the year (in French: charbonniers,
faudeurs) (Fig. 9.34). The silva forestis indicated an area with a special status as
hunting reserve for the landlord (English: forest, French: forêt royale, Dutch:
foreest or warande, German: Forst). Later, the term forest became used without
9.3 People Use the Land 235

any judicial connotation for denoting vast woodlands. In Germanic languages large
extents of land covered with trees were referred to as woodland (German: Wald,
Dutch: woud). The Franks referred to small woodlands as bosk, which resulted in
the Dutch word bos, and was latinised into boscus, giving the French bois. These are
similar to the English grove, thicket or copse (Muir 2000; Rackham 1990).
Wood pasture is a historical land management system in which open woodland
provided shelter and forage for grazing animals, particularly sheep and cattle, as
well as timber for construction and coppiced stems for poles, wood for fuel and
charcoal making. The system is also known as silvopasture. Advantages of a
properly managed silvopasture are enhanced soil protection and increased long-
term income due to the simultaneous production of trees and husbandry. Evidence
of old wood pasture management systems can be detected in many of the ancient
woodlands (Rackham 1990, 2002; Emanuelson 2009, Fig. 9.17).
The concentric land-use zoning resulted not only from the intensity of work
needed and the distance to the fields, but also economic and financial factors could
be important. The German agriculturist Johann Heinrich von Thünen (1783–1850)
demonstrated this principle. He formulated a circular-concentric land-use zoning
model based on the prices of the crops and the costs to produced them. The model
was based on his observations in his estate Tellow in Mecklenburg. He imagined an
isolated town situated in a homogeneous (isotropic) fertile plain bordered by a
wilderness. He used the model to demonstrate how to maximize agricultural
production taking into account crop selection, transportation costs (related to
distance to the centre, i.e. the manor) and the land rent. The result was a land-use
zoning in concentric circles around the centre. He published his findings in

Fig. 9.17 Land-use zoning around early agrarian villages. (1) village, (2) hortus, (3) ager, (4)
saltus, (5) silva, (6) forest. (1) to (3) correspond to the infields and can be enclosed land; (4) the
outfields. The forest is a judicial status of the land as hunting ground for the landlord where
villagers have well-defined use rights. A forest can cover woodland, wood-pasture, outfields and
infields. Natural conditions and social rules can cause distortions from this concentric ring shaped
zoning model
236 9 Building Blocks of the Landscape

Fig. 9.18 The model of von Thünen: 1. central place, 2. horticulture, orchards, 3. woods, 4.
intensive agriculture, 5. crop rotation with fallow (long cycle), 6. crop rotation with fallow (short
cycle), 7. extensive grazing land (outfields), 8. waste land, wilderness. (a) the concentric land-use
zoning in a uniform plain around a central place (town, city), dependent on distance (transportation
cost) and land rent; (b) pattern resulting from the existing of a second smaller centre (9); (c)
influence of transportation routes (i.e. river, road) disclosing the territory; (d) distortions due to
differences in land qualities (8b rocky outcrops, 4b fertile soils) (Adapted from von Thünen 1826;
Haggett 2001)

a three-volume book “Der Isolierte Staat in Beziehung auf Landwirtschaft und

National€okonomie” (von Thünen 1826) and his model is considered being the first
model in location theory (Fig. 9.18). Field Systems

Land is subdivided in fields or parcels according to land use and land ownership.
Fields are characterised by their size and shape, and by the way they are bordered.
Fields are grouped and ordered in field systems. These are important to understand
the development of landscapes, in particular concerning the agrarian history. The
distinction between land use parcels and ownership parcels is essential. Land use
parcels are the ones we see in the landscape or on aerial photographs and satellite
imagery. They are characterised by uniform land utilization. The patterns we see
can be very different from the ones represented on cadastral maps, which represent
the land ownership parcels (Fig. 9.19).
Fields vary in size, shape and pattern, and the names used for the different types
vary between languages and disciplines as well. Human geographers attempted to
make comparative classifications of field systems (Dussart 1957, 1961; Uhlig and
Lienau 1972; Lebeau 1979; Roberts 1987; Muir 2000, 2004). Figure 9.20 and
Table 9.5 give a summary of these attempts (Fig. 9.21).
Field systems can be very old and are very informative for the landscape history.
Typical examples are remnants of Celtic fields and the Roman centuriation field
system. Size, shape and spatial organization of fields reflect technology,
9.3 People Use the Land 237

Fig. 9.19 Land-use parcels and landowner parcels. The fields we see on aerial photographs
represent tracts of land with uniform land use. The parcels defined by land ownership as
represented on cadastral maps can be very different

understanding of the natural environment and social rules. Old practices are some-
times retained in specific terminology. In French, un journal refers to an area of
cropland that could be worked by one land worker (un journalier) in one day,
varying between regions and relief conditions, but on average half a hectare. The
term oxgang or bovate is an old land measurement formerly used in England and
Scotland, but the term is also known in other languages and refers to the area of land
that could be tilled using a single ox in a season. The area varied from place to
place, but was on average 15 acres or 6 ha. Other terms were used as well, not only
varying from region to region or between languages, but also according to the kind
of work. For example, the French journal refers to arable land, but for haying the
term faux or fauchée was used, and in vineyards une ouvrée. Similar contiguous
parcels were grouped into larger blocks that often had a proper field name. Larger
units could be bordered as well with tree rows, hedges, stonewalls, ditches or earth
and wood banks.
The difference between land use parcels and ownership parcels is fundamental
for reading the landscape. Generally speaking, land use parcels group adjacent
ownership parcels. The field system as represented on cadastral maps shows the
ownership parcels and often a severe fragmentation of the land can be noted, which
is the result of century-long subdividing in processes of inheritance. Recent land
consolidations overcome this extreme land fragmentation by grouping these parcels
238 9 Building Blocks of the Landscape

Fig. 9.20 Principles of field system classification. For rectangular fields, the length/width ratio
(L/W) is used to denote the different types (see text). Special forms of micro topography: (a) bank
at field edge, (b) ridge and furrow, (c) level ditch pattern, (d ) raised field. Slope characteristics: (e)
the threshold between gentle and steep slopes is 7 or 12% according to FAO, the shape of the
slopes can be straight (s), convex (cx) or concave (cc). Terrace types: ( f ) lynchet or talud, (g)
bench terrace, (h) walled bench terrace, (i) level bench terrace
9.3 People Use the Land 239

Table 9.5 Comparison of names of different field types in four languages

Type English French German Dutch
L/B < 2.5/1 Block Parcelle Block Blok(perceel)
L/B > 2.5/1 Strip lanière, Streifen Repel(perceel)
Long sides parallel parcelle
Smallest unit Field, Parcelle, Parzelle Perceel, kavel
parcel champ
Group of similar Furlong, Quartier, Parzellenverband Kwartier
adjacent fields bundle groupe
Adjacent groups with Complex Complexe Parzellencomplex Perceelscomplex
similar pattern of. . . de. . .
Whole of cultivated land ‘open finage Flur ‘land’
field’ agricole,

Fig. 9.21 Cadastral plans show parcels as registered by landownership. Often, also field-names
(F, toponyms) and minor land elements are indicated (as small terraces or lynchets in this case, T ).
Size and shape of the parcels are different and they form larger structures. Main shapes in this case
are strips (a), blocks (b) and irregular polygons (c). Strips are grouped into furlongs (Q, called
quartier in this region), which often are denoted by a proper name. Roads that are concordant (1)
with the field system indicate they were simultaneously created. Roads (2) that cut field systems
are probably more recent (Example from the cadastral plan of Bleid (Lorraine, Belgium), after
Dussart 1961)
240 9 Building Blocks of the Landscape

in larger new fields. Thus, only old cadastral maps still contain the historically
relevant information of the ancient field systems. Sometimes, fragments of former
field systems and related roads and settlements are visible as soil or crop marks on
aerial photographs. Territorial Models

The size and shape of territories varies a lot, as well as the location of the settlement
within. Different models of territories can be defined (Fig. 9.22). The hexagonal
model describes settlement patterns in a uniform, isotropic environment, which can
support a large population (Fig. 9.23). Most of the land is infield and in agricultural
use, and concentric land-use zoning is probable. The island-model shows settlement
along the coast and stretching their territories out inland, using water divides and
mountain ranges as borders. The sea should be considered a common territory as
well. The basin- or oasis-model describes the territories of settlements located at the
fringe of a closed basin of fertile land surrounded by a vast wilderness. This is for
example the case in tectonic basins and poljes in karst regions (Fig. 9.24, Table 9.6).
Following theoretical presumptions are used in analysing the size and shape of
• for agrarian settlements the assumption is the more fertile the soil is, the smaller
the territory can be needed to sustain its population;
• the larger the population of a settlement is, the larger the territory must be;
• as an efficient control centre, the location of the settlement is near the spatial
centre of gravity of the territory and the shape of the territory must be as compact
as possible.
Deviations from these assumptions have reasons and these help understanding
the actual size and shape and their historical development.
The systematic analysis of site location and territories demands first the defining
and delimiting of the territory related historically to the settlement that organized
it. Often these are townships, but accurate borders are often missing. Sometimes
they overlap with parishes. The establishment of municipalities often included an
administrative reorganization where the direct link between the settlement and its

Fig. 9.22 Territory models: (a) hexagonal-concentric, (b) island-model, (c) basin or oasis-model
9.3 People Use the Land 241

Fig. 9.23 An example of a landscape with a hexagonal territorial model: Haspengouw (Hesbaye)
in Belgium. The fertile loamy soils (alfisols; in light grey) are used entirely as cropland and allow
sustaining small village communities. The settlement density was high and the competition for
expanding the territory was severe. Consequently, the municipal territories are small and their
boundaries are sharp; the borders are angular and follow the field borders. The average number of
neighbours (the contact number) is six. The shapes of the territories are compressed hexagons
ordered along the axis of the valleys (alluvium in fine dotted dark grey; colluvial soils white and
dotted) (After Antrop 2007)

initial territory became fuzzy or was lost. In many countries administrative reforms
in the late twentieth century resulted in merging many settlements and older
structures were lost entirely.
A theoretical basis for estimating the initial territories can be based upon a
subdivision of space around each settlement using Thiessen-polygons. Figure 9.25
shows some concepts and used in the analysis of the location of settlements in their
territory such as centroid, centrality index, contact number and shape index.

9.3.2 Small Elements Give a Characteristic Touch Traditional Rural Architecture

The architecture of traditional rural buildings strongly relates to the landscape. The
building materials are often found locally and the style reflects the time of building,
the technology that was available and also the status of inhabitant. As the landscape
reflects the character of the community that shaped the land, so does the traditional
242 9 Building Blocks of the Landscape

Fig. 9.24 Example of a basin-model territory: the polje of Exo Lakkonia (Crete, Greece) (© Photo
M. Antrop 1997)

Table 9.6 Lifestyle, area of land needed and population supported

Lifestyle-farming type Population supported
Hunter-gatherers 1 person/km2
Herding and shifting cultivation 20 people/km2
Farming with permanent manured fields 50 people/km2
Advanced organic farming 200 people/km2
Modern farming with artificial fertilizers 3000 people/km2
After Emanuelsson (2009)

rural architecture. By typical and similar characteristics it enhances the identity and
unity of the landscape (Lake and Edwards 2006; Bell 1999; Zonneveld 1985;
Pinchemel 1970).
Most of what we consider today as traditional rural architecture, is often not
older that the mid-nineteenth century. Sources of information for older building
styles are found in paintings and drawings, which show that most common houses
were timber-framed and wattle and daub buildings (F: colombage, D: vakwerk, G:
Fachwerk) (Muir 2000; Yorke 2001). Stone buildings were rather exceptional,
expressing status, wealth and (strategic) importance. Regional differences are not
only due to the building materials used, but different styles can be recognized based
on cultural factors and landownership.
As for settlements, site and situation are also important for individual buildings,
in particular when they are dispersed. Traditional houses sought sheltered and
9.3 People Use the Land 243

Fig. 9.25 Variables used in the analysis of settlement sites and territories (After Van Eetvelde and
Antrop 2005)

protected sites. Constructions with other functions demanded other sites, such as
windmills, which as in the case of Holland, became icons of the identity of the
polder landscape. Trees and Hedgerows

Fields and tracts of land can be bordered by fences, stonewalls, hedgerows or trees,
creating an enclosed landscape with a distinct character, referred to as hedgerow
landscapes and bocage landscapes. Tree production is often combined with other
land uses, creating really multifunctional landscapes. Throughout history, trees
have been carefully managed as a valuable natural resource. Oliver Rackham
(2003) recognizes six forms of management:
• woodland: land on which trees have arisen natural and managed in a sustainable
• wood-pasture: multifunctional land use combining tree production and grazing;
a special form is the montado-dehesa combing trees and crop production and
• non-woodland: trees in fields and hedgerows;
• orchards;
• trees of gardens and streets;
• plantation: often as a monoculture.
Many types of vegetation borders exist as a result of management practices.
Similar management practices get distinct names, which vary according to the
language. In following examples the English names are compared to the French
244 9 Building Blocks of the Landscape

(F:), Dutch (D:) and German (G:) names, and Fig. 9.26 gives graphical represen-
tations of the different types and Fig. 9.27 some photographic examples:
• Hedgerows (F: haie, D: haag, heg, G: Hecke)
• Tree rows (F: rangée/rideau d’arbres, D: bomenrij, G: Baumreihe)
• Coppice (F: taillis, D: hakhout, G: Niederwald)
• Coppice (wood) (F: taillis (sur têtards), D: hakhout(bos), G: Niederwald
• Coppice with standards (F: taillis sous futaie, D: spaartelgenbos, G: Mittelwald)
• Coppice with pollards (F: taillis sur têtards, hakhout met knotbomen, G:
Niederwald mit Kopholz)
• Pollard, bolling (tree) (F: arbre écimé/étêté, D: knotboom, G: pollard)
• Shredded tree (F: ragosse, D gekandelaberde boom)
• Avenue (F: allée, D: dreef, G: Allee)
Hedges, coppices and pollards can be planted on talus, terraces or (wood)banks
(F: terrasse, rideaux, D: graft, talud, G: Terrasse).
Small vegetation elements have significant ecological effects on adjacent fields
with respect to microclimate, soil fertility and agricultural production. Figure 9.29
summarizes some of the microclimatic effects of a hedgerow on a field in the lee
side of the wind. Tall hedges, tree rows and thickets of coppice have a filtering
effect on the visual landscape. They create a gradual and textural enclosure of the
landscape scenery with increasing viewing distance. On level land, where no
overviews are possible, the successive screens create a particular staged landscape

Fig. 9.26 Examples of coppicing and pollarding practices. (1) suckers (cloning from roots), (2)
coppice, (s) stool can be above or below the ground, (3) coppice hedgerow, (4) woven hedgerow,
(5) stub on talus or lynchet (t), or woodbank (w), (6) coppice on stem, (7) low pollard, (8) normal
pollard, (b ¼ bolling), (9) high pollard (têtard), (10) candelabra tree, (11) stag-headed pollard, (12)
shredded tree (After Rackham 1990; Tack et al. 1993; Burel and Baudry 2000; Muir 2005)
9.3 People Use the Land 245

Fig. 9.27 (1) Trimmed hedgerows combined with pollards are characteristic for the bocage-
landscape of the Avesnois (France), (2) Low pollards combined with a hedge formed by suckers
enclosing a pasture, (3) Hedgerows and coppices on earth banks formed by removing stones from
the field and played with a combination of a hedge and pollards; an example from Normandy
(© Photo Antrop 2004, 2005)
246 9 Building Blocks of the Landscape

Fig. 9.28 An example of the coulisse effect of successive rows of pollard trees filtering the
landscape transparency (Klaverdries, Flanders) (© Photo Antrop 2007)

Fig. 9.29 Microclimatic effects on the lee side of a coppice or hedgerow in function of the
distance h (h ¼ height of hedgerow and expressed in relative positive or negative impact: (a) soil
moisture, (b) air temperature (day), (c) soil temperature (day), (d ) relative humidity, (e) air
temperature (night), ( f ) evaporation, (g) wind speed (After Forman and Baudry 1984)

giving a coulisse effect (Fig. 9.28). This was typical in many areas of Flandres and
described succinctly by the French geographer Raoul Blanchard (1906).
In landscape ecology these small vegetation elements received special attention,
in particular as small biotopes and their role in landscape connectivity and frag-
mentation, their functioning as stepping stones and corridors and their importance
in preserving biodiversity (Agger and Brandt 1984; Forman and Baudry 1984;
Forman and Godron 1986; Schreiber 1987; Mander et al. 1995; Burel and Baudry
2003; Jongman and Pungetti 2004; Jongman 2007). Ecological networks became
important in landscape planning and restoration. They are also referred to as green
networks, greenways, and in French as tr^ ame verte et bleue (Fig. 9.29). Terraces, Earthworks and Soil Marks

Many micro relief forms exist and have natural or anthropogenic origins. Most
typical are terraces on steep slopes. Natural terraces are created by rivers and
9.3 People Use the Land 247

characterize valley slopes over long distances. They indicate temporary base levels
in the river erosion. Anthropogenic terraces and talus have been built to prevent soil
creep and gully erosion and to flatten stepwise the land making it easier to cultivate
and gain more arable soil. In areas with rocky soils, stones are amassed and used to
build field walls and terrace walls. In loose soils on slopes, other earth banks are
common and witness from practices of clearing and working the land. Examples are
strip lynchets as they are called in English, rideaux or talus in French, graften or
taluds in Dutch and Abhang or B€ oschung in German. Some mark as woodbanks the
edges of woods and were often planted with hedges and coppices, to prevent
animals going on the fields. When forest was cleared, these hedgebanks mark the
former wood limits in the fields (Fig. 9.30).
Common are hollow roads, which in loamy substrates, develop from intensive
use of dirt roads connecting the valley and the plateau causing gully erosion on the
steep, convex top slopes (Fig. 9.31). Many hollow roads disappeared with the
up-scaling and mechanisation of agriculture. Many became waste dumps and
were filled. Others were paved to prevent further erosion. The remaining ones
became often protected because of their ecological value (Fig. 9.32).
Many other features witness from former practices, sometimes in an ephemeral
way. Soil marks are recognizable by colour differences on the bare soil or patterns
formed by growth differences in the vegetation and are then called cropmarks
(Figs. 9.33, 9.34, and 9.35, see also Figs. 3.7, 4.4 in Chaps. 3 and 4). They express
small variations in soil conditions, which can be caused by natural factors such as

Fig. 9.30 Cultivation terraces in west Crete. This example shows old, partially abandoned and
eroded terraces and new ones recently built (Prases, © Photo Antrop 2010)
248 9 Building Blocks of the Landscape

Fig. 9.31 Hollow roads develop from intensive use of dirt roads connecting the valley (a) and the
fields on the plateau (b), causing gully erosion on the steep, convex top slopes. They are
characterised by a S-shaped profile. A: initial topography, B: profile hollow road, C: evolution
of the erosion

Fig. 9.32 (A) The paving of the hollow road prevents further erosion. (B) talus and terraces were
created to reduce soil erosion (© Photo Antrop 2005)
9.3 People Use the Land 249

Fig. 9.33 Charcoal burning pans form dark circular patches in the soil and can still be seen in
cropland; they indicate the extent of the former forest (© Photo Antrop 2008)

Fig. 9.34 Charcoal burners cleared parts of the forest to produce charcoal in furnaces made of cone-
shaped stacks of timber. This practice continued as long as wood was available and cheaper than coal,
as illustrated by this photograph from Brittany in the beginning of the twentieth century

Fig. 9.35 Soil en crop marks. A: substrate, B: normal top soil layer, C: shallow top soil with buried
stony foundation remnants, D: filled ditch or pit, E: lithological outcrop, F: colluvium from soil
creep covering the top soil. Crop a caused a negative growth over the shallow soils, and a positive
above the filled ditches (b). Crop c has an opposite response
250 9 Building Blocks of the Landscape

soil erosion, or have anthropogenic origins, such as filled ditches of buried foun-
dations. They are informative traces from the past and significant for landscape
archaeological research.

9.3.3 People Name Landscapes: Toponyms

Toponymy (derived from the Greek t opos-place and onoma-name) is the study of
place names (toponyms) in a broad sense, including field names, names of rivers
and landforms, etc. The study covers the etymology, the origins (time period),
meanings, use and typology. Toponyms contain valuable information about the
historical land use. Examples of the meaning and interpretation of place and field
names in the context of landscape research are mainly found in studies in historical
geography, in historical ecology (Rackham 1986, 1990, 2003) and in archaeology
(see for example Muir 2004, 2001).
A typology of toponyms can use following dimensions:
• Age, time period: can be older than written records
• Language: indicate successive cultures that lived in the area
• Topography: high – low
• Relation to water
• Specific landforms
• Specific vegetation or land use
• Border, edge
• Religion
The most important sources of toponyms are land registry and topographical
maps, where the link can be made between the location, the land use or cover and
the linguistic meaning of the toponym. However, place names on maps are also
defined by cartographic rules. In particular, in their selection and location is
determined by the readability of the map. Modern developments wipe away the
former landscape, which may result in a shift of the place name or its disappear-
ance. Figure 9.36 gives an example of a shift due to urbanisation, breaking the link
between the original meaning of the toponym and the landscape.

9.4 Mosaics and Borders: Interactions Between Places

9.4.1 Settlement Patterns: Clustered or Scattered?

Farmsteads can cluster together in hamlets or villages, or occur scattered in space.

The spatial pattern can be random or regular. Spatial analysis allows quantifying the
density of the pattern and techniques as nearest neighbour analysis allows testing if
9.4 Mosaics and Borders: Interactions Between Places 251

Fig. 9.36 Example of a shifting toponym “Buchten” on successive topographical maps (© IGN
topographical maps from 1910, 1979 and 1999, region of Ghent, Belgium). Originally the
toponym referred to wetlands and shifted due to urban sprawl. Today it refers to a bridge and
has no relation to the land use anymore

the pattern differs significantly from randomness. The fundamental question is why
people sometimes settle together and why in other cases they prefer isolation.
Several factors of attraction have been formulated: protection from a common
threat, need for cooperation, and the availability of a common natural resource,
such as a water well (Fig. 9.37). The conditions for a scattered distribution are the
opposite: peaceful times and abundance of natural resources. These factors may
change and so may the settlement pattern as well (Fig. 9.38). Often clustered
settlements are older and the scattered farmsteads filled the territorial periphery in
between (Table 9.7).
252 9 Building Blocks of the Landscape

Fig. 9.37 Spatial distribution types of individual settlements (at the scale of farms and hamlets)
and possible forces causing the pattern: (a) irregular agglomeration, (b) linear or row settlement,
(c) regular agglomerated, planned, (t) regular pattern in tessellation (hexagon). The transitional
form dispersed + clustered may result from the growing of a place into a village, or from the
infilling of the space around a village with new holdings

Fig. 9.38 An imaginary initial village settlement consisting of similar, homogenous units ran-
domly scattered (1), becomes more dense with the founding of new settlements (2), and diversifies
with some places growing (a) and others declining (b), move (c) or even disappear (x), resulting in
a more clustered pattern (3)

9.4.2 The Multiple Estate

Often places were not completely autonomous or independent, but had bonds with a
larger estate belonging to one landowner, or with a larger urban centre as a town or
city. Aston (2004) described a model of the complexity of the relationship in such a
situation of a multiple estate (Fig. 9.39).

9.4.3 Territorial Mosaics

Figure 9.40 gives an example of a settlement-territory analysis of some villages in

the Provence region (France). This case illustrates the hexagonal model for the
9.4 Mosaics and Borders: Interactions Between Places 253

Table 9.7 A key model for understanding the spatial distribution of rural settlements
distribution Economic factors Social factors Political factors
Nucleation Major goals Locational regulators Decision making
Principle: • survival – wealth Access to resources • Internal (local)
communality: accumulation • natural • external (colonial)
• of effort • change – stability • cultural
• to economise
• of enforcement Economic base Size regulators Decision enforcement
or defence • agriculture • physical environment • internal (local)
• crafts / industry • supply (water, food, • external (colonial)
Dispersion • fishing fiber, etc.)
Principle: • husbandry / • social factors
individuality: herding • density of population
• Freedom/status • forestry
• cash income • extraction Population regulators
... • demographic profile
• immigration / emigra-
• hazards (disease, war,
famine. . .)
• social frameworks
After Roberts and Wrathmell (2000)

initial territories around hilltop villages in a heterogeneous natural environment.

The landscape is a compartmental mosaic of forested limestone plateaus and basins
with sediments from the Cenozoic period, offering fertile soils in which the
settlements are located near the cultivated land. Note that the spatial pattern of
the landscape units, which are defined by geology, landform, land use and settle-
ments, is incongruous with the spatial pattern of the territorial units indicated here
by the township borders of the old municipalities. Most villages are clustered
hilltop settlements and originated between the eleventh and thirteenth centuries as
defensive sites. From the church tower, a visual control over the most valuable
arable territory was possible. The township borders, however, bind most of the
viewsheds (Sevenant and Antrop 2007). In later times, dispersed settlements were
added in the valley parts closer to the fields.
Territorial shape analysis reveals several characteristics. All territories belong to
the hexagonal model type and the average number varies between four and seven,
close to the expected value of six in a hexagonal tessellation. All territories cover
the largest diversity in natural land resource possible, i.e. both the fertile basin and
the forested upland, which was used for grazing as well. Shape analysis shows two
254 9 Building Blocks of the Landscape

Fig. 9.39 The multiple estate model (adapted after Aston 2004). A big landowner (lord, abbey,...)
possessed land and settlements scattered over a vast area, not necessarily continuous and adjacent
to each other. The regional centre (A) exerted the control over the entire domain by power, rights
and services, including the spatially separate land (B). It has the main church and a castle exerting
the military power in the region and controlled the traffic across the river. The smaller settlements
specialised considering the land qualities in their territory (in the figure indicated by their fictitious
name) and particular village form. Each settlement has infields and outfields, but also have rights
on the common land they share (forest, marshland, heathland). (C) is an independent domain of a
monastery controlling water power and the crossing of the river. Relations are for example: (1)
central power (political, judicial, religious), (2) taxes and services, (3) access to forest land (wood,
charcoal), (4) access to the uplands (summer pastures, heathland, stone), (5) access to marshland
(peat, salt). (6) the border of the domain. Landmarks such as solitary trees, crosses and chapels
mark the boundary of the domain

different patterns. In the southwest, the hexagonal territories are very compact,
while in the northeast they are compressed to oblongs caused by terrain accessibil-
ity to the infields.
Figure 9.41 shows some examples of the village territories in four landscape
regions in Belgium, illustrating the effect of the natural conditions on size and
shapes of the territories.
1. The Kempenland lies in the sandy region in the north of Belgium and has poor
natural resources for early settlers. On the sandy soils with low fertility
(spodosols), only a low density of settlements was possible and extensive
territories were needed to sustain the communities. Arable land is limited and
large areas of heathland are extensive common grazing grounds. The shapes of
the territories are irregular and more elongated. Borders can be straight over long
distances. The village sites are often eccentric in the territory.
9.4 Mosaics and Borders: Interactions Between Places 255

Fig. 9.40 Example of an analysis of settlement sites and their territories in the Provence (Var,
France). Most villages are clustered hilltop settlements (1) with some dispersed farms and hamlets
(2) around. (3) the centre gravity of the arable land (ager), and (4) the centre of gravity of the
territory. (5) border of the territory (ancient township, now municipal border), (8) agricultural land
(ager), (9) forest, (10) water. The site of the village is located closer to the centre of gravity of the
ager (3) than to the centre of gravity of the territory (4). This is expressed numerically by the
centrality indices (in %): (7a) for the ager, (7b) for the territory. The shape index of the territories
256 9 Building Blocks of the Landscape

2. Haspengouw (Dutch) or Hesbaye (French) is a fertile agricultural region on the

middle of Belgium. The landscape consists of gentle plateaus with fertile loamy
soils (alfisols), which allowed a dense settlement of villages specialized in
agriculture. Small territories were possible and the high population pressure
resulted in compressed hexagonal territories with angular borders. Most of the
villages are located near the centre of gravity of the territory. The site location in
the valley and the territories covering slopes and plateau grounds indicate
searching for the largest diversity of land qualities possible.
3. The third example shows the transition of the natural regions of the Calestienne
(C) and Ardennes (A) (marked by a thick dotted line). The natural conditions are
completely different between both. The Calestienne is a limestone plateau at
250–300 m with shallow inceptisols and some eolian loess deposits. The
Ardennes is an upland plateau rising to 400 m with long steep, north facing
slopes. Soils are shallow and the substrate consists of hard rocks (regosols). The
initial settlements occupied the Calestienne bordering the forested, steep slopes
of the Ardennes. The high density of agrarian settlements resulted in compact
hexagonal territories with the village near the centre. Later expansion of the
territory was only possible towards the wilderness in the south and resulted
finally in long oblong territories (After Antrop 2007). A detailed analysis of the
case of the village Chanly (shaded in Fig. 9.42) is given in next section.

9.4.4 Evolution of Settlement Patterns and Territories

The actual settlement pattern is the result of a long evolution. Changes in the pattern
are mainly controlled by population dynamics, lifestyles, technology and econom-
ics. If there is ample space available and no neighbours, territories can expand when
the population grows. However, in most cases constraints exist and expansion is
only possible by reclaiming unsettled wilderness. This was the case at the border-
land between the Calestienne and Ardennes regions, as illustrated in Fig. 9.41. Once
the territories were fixed and recognized, no territorial expansion was possible
anymore. With still growing populations and no option of migration, the only
solution is increasing the productivity of the land. Successive phases of land use
intensification and abandonment, related to population growth and decline, can be
read from current land use patterns.

Fig. 9.40 (continued) is given by coefficient of variation of the radii from the centre of gravity to
the border: (7c) for the ager, (7d) for the municipality. The contact number (7C) gives the number
of neighbours for each settlement; in the case of hexagonal territories the theoretical average of six
is expected. The viewsheds (6) from the hilltop sites cover the ager area maximally and are bound
within the municipal territories (After Antrop 1989)
9.4 Mosaics and Borders: Interactions Between Places 257

Fig. 9.41 Size and shape of municipal territories in different landscape regions in Belgium
Lines represent administrative borders of townships, approximated by the ancient municipal
258 9 Building Blocks of the Landscape

Figure 9.42 describes this evolution the land use in the territory of the village of
Chanly (shaded in Fig. 9.41, case 3). The forest on the northern slopes of the
Ardennes part was cleared, first to make outfields (pat^ ures-sarts) and towards the
end of the eighteenth century most of the territory became arable land (1). In the
nineteenth century, a large part of the rural population migrated towards the
industrializing and coal production areas of Wallonia. The declining population
caused abandonment of the outfields and rewilding. The law on land reclamation of
1847 forced the landowners and municipalities to make all wasteland productive.
The were reforested (2) and planted with conifers, which wood was highly
demanded in the coalmines (3). Only few outfields remained as arable and grazing
land (i), but with the set aside policy in the second half of the twentieth century
many parts were planted with Christmas trees.
As a result of catastrophic changes, the whole structure of the landscape, with
settlements and territorial patterns can be wiped away. Severe famines or epi-
demics, wars and massive emigration resulted in shrinking of the size of settlements
and places becoming completely deserted (Runnels and van Andel 1987).
Settlement continuity is a basic issue in archaeology and historical geography
(Butlin 1993; Roberts 1987). For towns and cities evidence is more readily avail-
able, but this is seldom the case for smaller rural settlements. For example, Roger
Agache (1978) demonstrated a clear discontinuity in the settlement and a funda-
mental change in the pattern between the Roman and medieval period in Picardie
(France). After many years of surveying the region using oblique aerial photogra-
phy, he was able to reconstruct the settlement pattern during the Roman period,
which consisted of scattered villas all over the land. Using aerial photography,
deserted and lost villages have been documented abundantly in particular in Great
Britain and Ireland (Fig. 9.43).
Changes in land use and territories can be indicative for population dynamics as
well. The first settlers were free to choose the best land; fertile soils, easy to work, well
situated and abundant and diverse natural resources nearby. Early agrarian settlements
subsisted from the land and they could as long as they managed the land in a
sustainable way. Many techniques were invented to achieve this, e.g. ways to fertilize
the soil, crop rotation, fencing, pollarding, draining and clearing the land from stones.
Increasing population demanded more land to be cleared and the newcomers had
to find new places to settle. Gradually, also marginal land became settled, or new
land was created as by draining marshlands and creating polders.
The different practices and phases are imprinted in the landscape and are
partially visible on (historical) maps and aerial photographs.

Fig. 9.41 (continued) borders. Village centres indicated by black dot, current build-up area shaded
grey. The maps are at the same scale
1. Territories in the Kempenland
2. Territories in Haspengouw/Hesbaye
3. Territories at the transition of the Calestienne (C) and Ardennes (A) natural regions (marked by
a thick dotted line)
9.4 Mosaics and Borders: Interactions Between Places 259

Fig. 9.42 Expanding and shrinking of a territory, the case of the village of Chanly (Wallonia,
Belgium). C: village of Chanly at a site bridging the Lesse river, H: village of Halma, R: village of
Resteigne, a: largest extension of the arable land at the end of the eighteenth century, b: township
border of Chanly (former communal border), c: border between the historical provinces of Namur
and Luxemburg, d: motorway dating from the end of the twentieth century, e: geological limit
between the Calestienne (north) and the Ardennes (south), f: crest lines, g: coniferous plantation of
the former outfields, h: reforestation with deciduous forest on the former outfields, i: still
remaining outfields, with systematic field strips ( j), k: initial deciduous beach and oak forest, x:
areas of recent Christmas tree plantation. 1: largest expansion of the arable territory, 2–3 retreat
phases of the cultivated territory and forest plantation. (white: arable land, light grey: deciduous
forest, dark grey: coniferous forest)

9.4.5 Hierarchy of Settlements

Settlements vary in size and importance. Most settlement typologies use the
population size to define the classes and vary in class limits and terminology.
Many terms used to describe settlements types have no legal definition and are
not comparable between states. Using census data of population size are often
260 9 Building Blocks of the Landscape

Fig. 9.43 Settlement continuity in Picardie (France). The actual villages and hamlets (black) are
clusters of farms and were founded in the medieval period. There is no relation with the scattered
pattern of the villas (grey circles and rectangles) during Roman times, which was reconstructed
after aerial surveying of soil- and crop-marks (Extract from Agache 1978)

misleading as the size and boundaries of the census unit is seldom significant for the
spatial distribution of the people.
An example of a classification combining population size with other character-
istics was proposed by the Greek city planner Constantinos Doxiadis. He called
settlements ekistic units and his hierarchy defined eight classes (Doxiadis 1976,
• Isolated settlement: an isolated dwelling or farm (one family)
• Hamlet: a group of few houses, having a tiny population (<100 inhabitants) and
no or very few services
• Village: a group of houses with a population between 100 and 1000 inhabitants,
forming a community. The place offers elementary services, such as a small
shop or post office, and a church. Villages often have a proper territory and land
use rights on it.
• Town: a place with a population between 1000 and 20,000 or even up to 100,000
inhabitants. Towns often possess historical privileges. Its spatial structure is
diversified in neighbourhoods. Towns offer most of the essential services and act
as a central place on the surrounding countryside, which forms its hinterland.
• City: the population is over 100,000 people, but less than one million for large
cities (Doxiadis called this a polis). Services are abundant and residential
References 261

functions are combined with economic activities and administrative services.

Cities form communication and trade nodes in a broader urban network. The city
centre often has ancient historical roots.
• Conurbation: is often a chaotic spatial complex grouping smaller existing
historical settlements that expanded by industrialisation and urban sprawl. The
population is very dense and varies from the size of a city up to more than a
million inhabitants.
• Metropolis: a large urban place consisting of multiple cities and towns including
suburbs. The population is usually one to three million.
• Megalopolis: an urban place of more than ten million people. The United
Nations speaks of megacities when the population becomes larger than ten
million inhabitants.
The larger the settlements are in size, the fewer in numbers they will be. Hence,
there are many small settlements and few large cities. Smaller settlements are close
together and have relatively small hinterlands, while large cities have large
The larger a settlement is, the larger its sphere of influence becomes and the
more important it acts as a ‘central place’. The German geographer Walter
Christaller (1933/1966), based on previous work by Edgar Kant (Buttimer 2005),
formulated a theory of central places that explains the number, size and location of
human settlements in a hierarchical urban system. Each central place provides some
specific services for its own hinterland.
Christaller’s theory resulted in an orderly and hierarchical subdivision of space
according to the spheres of influence, resulting in hexagonal tessellations, which
properties depend upon the dominating nature of services the centres provide
expressed as three organising principles: market, traffic-transportation and admin-
istration. Although the theory uses simplifying assumptions, such as an unbounded
isotropic, homogeneous space, equidistant settlements, evenly distributed popula-
tion and natural resources, hierarchical have been identified, which enhanced the
popularity of the theory. It has even been applied in planning new landscapes such
as the Noord-Oostpolder and Flevoland in the Netherlands.
Other spatial models of hierarchical settlement patterns have been devised based
on economic and demographic principles (Haggett et al. 1977). Nevertheless,
empirical observed hexagonal territorial patterns exist also without the theoretical
assumptions of Christaller’s theory, as illustrated by examples in Figs. 9.40, 9.41
and 9.42.


Agache, R. (1978). La Somme Pré-Romaine et Romaine (515 pp). Société des Antiquaires de
Picardie: Amiens.
Agger, P., & Brandt, J. (1984). Registration methods for studying the development of small scale
biotope structures in rural Denmark (pp. 61–72). In Proceedings of the First International
262 9 Building Blocks of the Landscape

Seminar on Methodology in Landscape Ecological research and Planning, theme II, Roskilde
University Centre, IALE.
Antrop, M. (2007). Perspectieven op het landschap (322 pp). Gent: Academia Press.
Antrop, M., & Van Eetvelde, V. (2000). Holistic aspects of suburban landscapes: Visual image
interpretation and landscape metrics. Landscape and Urban Planning, 50(1–3), 43–58.
Aston, M. (2004). Interpreting the landscape. Landscape archaeology and local history (168 pp).
London: Routledge.
Aston, M., & Rowley, T. (1974). Landscape archaeology: An introduction to fieldwork techniques
on post-roman landscapes. Newton Abbot: David & Charles Ltd.
Bell, S. (1999). Landscape. In Pattern, perception and process (344p). Group: Taylor & Francis.
Blanchard, R. (1906). La Flandre. E´tude géographique de la Plaine Flamande en France,
Belgique et Hollande (pp. 530). Dunkerke: Société Dunkerquoise pour l’avancement des
lettres, des sciences et des arts.
Bloemers, J. H. F., Kars, H., van der Valk, A., & Wijnen, M. (Eds.). (2010). The Cul- tural
landscape heritage paradox. Protection and development of the Dutch archaeological–histor-
ical landscape and its European dimension. Amsterdam: Amsterdam University Press.
Burel, F., & Baudry, J. (2000). Ecologie du paysage. Concepts, méthodes et applications (359 pp).
Technique et documentation.
Burel, F., & Baudry, J. (2003). Landscape ecology: Concepts, methods and applications (362 pp).
Enfield: Science Publishers.
Butlin, R. A. (1993). Historical geography. Through the gates of space and time (306 pp). London:
E. Arnold.
Buttimer, A. (2005). Edgar Kant (1902–1978): A Baltic Pioneer, Geografiska Annaler. Series B,
Human Geography, 87(3). Special Issue: The Heritage of Edgar Kant and J. G. Gran€ o
(pp. 175–192).
Chisholm, M. (1962). Rural settlement and land use. London: Hutchinson University Library.
Christaller, W. (1933/1966). Die zentralen Orte in Süddeutschland. Jena, Gustav Fischer
(C. W. Baskin, Trans.). Central Places in Southern Germany. Prentice Hall, 1966.
Claval, P. (2004). The languages of rural landscapes. In: H. Palang, H. Sooväli, M. Antrop &
S. Setten (Eds.), European rural landscapes: Persistence and change in a globalising envi-
ronment (pp. 11–40). Kluwer Academic Publishers.
Clergeau, Ph. (2007). Une écologie du paysage urbain (pp 136). Rennes, Édition Apogée.
Doxiadis, C. A. (1968). Ecumenopolis: Tomorrow’s city. Britannica book of the year 1968.
Encyclopedia Brittanica,’s
Doxiadis, C. A. (1976). ACTION for human settlements. Center of Ekistics: Athens.
Dussart, F. (1957). Geographie der länderlichen Siedlungsformen in Belgien und Luxemburg.
Geographische Rundschau, 1, 12–18.
Dussart, F. (1961). Les types de dessin parcellaire et leur répartition en Belgique (p. 33). Bulletin
de la Societé belge d’études géographiques.
Emanuelsson, U. (2009). The rural landscapes of Europe. How man has shaped European nature
(pp. 383). The Swedish Research Council Formas.
Forman, R., & Baudry, J. (1984). Hedgerows and hedgerow networks in landscape ecology.
Environmental Management, 8–6.
Forman, R., & Godron, M. (1986). Landscape ecology (619 pp). New York: Wiley.
Gaston, K. J. (2010). Urban ecology. New York: Cambridge University Press.
Haggett, P. (2001). Geography: A global synthesis (672 pp). New York: Prentice-Hall.
Haggett, P., Frey, A. E., & Cliff, A. D. (1977). Locational analysis in human geography (p. 605).
New York: John Wiley & Sons.
Howard, E. (1898). To-morrow: A peaceful path to real reform. London: Swan Sonnenschein &
Howard, E. (1902). Garden cities of to-morrow. London: Swan Sonnenschein & Co. http://soils.
References 263

Jongman, R. H. G. (2007). Ecological networks, from concept to implementation. In S.-K. Hong,

N. Nakagoshi, B. Fu, & Y. Morimoto (Eds.), Landscape ecological applications in
man-influenced areas – Linking man and nature systems (p. 536). Dordrecht: Springer
Jongman, R. H. G., & Pungetti, G. (2004). Ecological Networks and Greenways: Concept, Design,
Implementation (p. 345). Cambridge: Cambridge University Press.
Knaepen, R., & Antrop, M. (2003). The functional evolution of common place-settlements in the
rural Kempenland of Belgium and the Netherlands. In T. Unwin & T. Spek (Eds.), European
landscapes: from mountain to sea. Proceedings of the 19th Session of the Permanent European
Conference for the Study of the Rural Landscape (PECSRL) at London and Aberystwyth
10–17th September 2000 (pp. 54–62). Tallinn: Huma Publishers.
Lake, J., & Edwards, B. (2006). Farmsteads and landscape: Towards an integrated view. Land-
scape, 7(1), 1–36.
Larrère, R., & Nougarède, O. (1993). Des hommes et des forêts (p. 128). Paris: Découvertes
Gallimard Traditions.
Lebeau, R. (1979). Les grands types de structures agraires dans le monde (162 pp) Paris: Masson,
troisième édition.
Mander, Ü., Palang, H., & Jagomägi, J. (1995). Ecological networks in Estonia. Landscape, 12(3),
Massot, J.-L. (1975). Maisons rurales et vie paysanne en Provence. Aix-en-Provence, Ed. SERG.
Muir, R. (1999). Approaches to landscape (310 pp). London: MacMillan Press.
Muir, R. (2000). The NEW reading the landscape. Fieldwork in landscape history (256 pp).
Exeter: University of Exeter Press.
Muir, R. (2001). Landscape detective: The landscape of Ripley. Macclesfield: Windgather Press.
Muir, R. (2004). Landscape Encyclopaedia. A reference guide to the historic landscape (297 pp).
Bollington: Windgather Press.
Muir, R. (2005). Ancient Trees – Living Landscapes (p. 256). Stroud: Tempus Publ.
Niemelä, J., Breuste, J. H., Guntenspergen, G., McIntyre, N. E., Elmqvist, T., & Philip James,
P. (Eds.). (2011). Urban ecology: Patterns, processes, and applications (p. 392). Oxford: OUP.
Pinchemel, P. (1970). Géographie de la France. In Tome II : les milieux : campagnes, industries et
villes (295 pp). Paris: Librairie A.Colin.
Pinto-Correia, T., & Vos, W. (2004). Multifunctionality in Mediterranean landscapes – Past and
future. In R. Jongman (Ed.), The new dimensions of the European landscape, Wageningen EU
Frontis Series. Wageningen: Springer.
Rackham, O. (1986). The history of the countryside (445 pp). London: Dent.
Rackham, O. (1990). Trees & Woodland in the British landscape. The complete history of Britain’s
trees woods & hedgerows (p. 234). London: Dent.
Rackham, O. (2003). Ancient woodland: Its history, vegetation, uses in England. Colvend:
Castlepoint Press.
Rackham, O., & Moody, J. (2012). Drivers of change and the landscape history of Cavo Sidero. In
T. Papayannis & P. Howard (Eds.), Reclaiming the Greek landscape (pp. 219–232). Athens:
Roberts, B. (1987). The making of the English village (237 pp). London: Longman.
Roberts, B. K., & Wrathmell, S. (2000). An atlas of the rural settlement in England (77 pp).
London: English Heritage.
Runnels, C. N., & van Andel, T. H. (1987, July–September). The evolution of settlement in the
southern argolid, Greece: An economic explanation. Hesperia: The Journal of the American
School of Classical Studies at Athens, 56(3), 303–334.
Schreiber, K. F. (Ed.). (1987). Connectivity in landscape ecology. In: Proceedings of the 2nd
international seminar of the international association of landscape ecology. Münster,
Münsterische Geographische Arbeiten 29, Sch€ oning, Paderborn.
Sevenant, M., & Antrop, M. (2007). Settlement models, land use and visibility in rural landscapes:
Two case studies in Greece. Landscape and Urban Planning, 80(4), 362–374.
264 9 Building Blocks of the Landscape

Tack, G., Van Den Bremt, P., & Hermy, M. (1993). Bossen van Vlaanderen. Een historische
ecologie. Leuven: Davidsfonds.
Tsjallingii, S. (1995). Ecopolis, strategies for ecologically sound urban development (p. 108).
Leiden: Backhuys Publishers.
Tsjallingii, S. (1996). Ecological conditions. Strategies and structures in environmental planning
(320 pp). Wageningen: IBN Scientific Contributions 2. IBN-DLO, (ISBN 90-801112-3-6).
Uhlig, H., & Lienau, C. (1972). Die Siedlungen des ländlichen Raumes Rural Settlements –
L’Habitat Rural. Materialien zur Terminologie der Agrarlandschaft, Vol.II, Internationale
Arbeitsgruppe für die geographische Terminologie in der Agrarlandschaft. Giessen: Lenz
USDA (2010). Keys to soil taxonomy (11th edn). United States Department of agriculture –
Natural resources conservation service.
Van Eetvelde, V., & Antrop, M. (2005). The significance of landscape relic zones in relation of soil
conditions, settlement pattern and territories in Flanders. Landscape and Urban Planning., 70
(1–2), 127–141.
Von Thünen, J. H. (1826). Der Isolierte Staat in Beziehung auf Landwirtschaft. Hamburg:
Frederich Perthes.
Yorke, T. (2001). Tracing the history of villages (p. 192). Newbury: Countryside Books.
Zonneveld, J. I. S. (1985). Levend land. De geografie van het Nederlandse landschap. Scheltema
& Holkema, Utrecht/Antwerpen: Bohn.
Zonneveld, I. S. (1995). Land ecology (199 pp). Amsterdam: SPB Academic Publishing bv.
Chapter 10
Identifying, Mapping and Assessing

Abstract The complex, varied and continuous landscape can be understood better
when classified in types and spatial units. The classification process comprises three
phases: identification, assessment and monitoring. Several methods for identifying
and mapping landscape types and regions have been developed for different needs.
Typologies classify landscapes that are distinct from each other based on a set of
differentiating attributes, irrespective their location or geographical context. A
chorology defines and delineates spatial landscape units and orders them in a
hierarchy according to scale. Two approaches are possible in landscape classifica-
tion: by subdivision or descending and by aggregation or ascending. Subdivision
often used a holistic method similar to photo-interpretation. Aggregation follows a
parametric method where first basic units are defined using a selection of attributes,
which are subsequently clustered. Combined methods are possible and are more
flexible in combining diverse data sources in a multi-scale context using holistic
and parametric methods alternatively. The focus of the inventory can be on natural
landscape regions, functional-historical characteristics of cultural landscapes and
mapping the landscape scenery and mindscapes. This demands appropriate
methods and different ways of representing the results, such as landscape atlases
and catalogues. Assessment involves assigning values with a specific goal in mind.
Landscape character assessment has become a common approach in landscape
policy. As landscapes are dynamic, changes need to be monitored and different
methods have been developed for different purposes.

Keywords Inventory • Land evaluation • Landscape atlas • Landscape character

assessment • Landscape observatory • Monitoring

10.1 From Identification to Monitoring

Article 6 of the European Landscape Convention (Council of Europe 2000) lists the
specific measures to be taken to realise the aims of the convention. The third one
describes a huge task of identification and assessment “with a view to improving
knowledge of its landscapes”:

© Springer Science+Business Media B.V. 2017 265

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_10
266 10 Identifying, Mapping and Assessing Landscapes

a. (i): to identify its own landscapes throughout its territory;

(ii): to analyse their characteristics and the forces and pressures transforming them;
(iii): to take note of changes;
b. to assess the landscapes thus identified, taking into account the particular values
assigned to them by the interested parties and the population concerned.

This implies defining and describing landscape types and spatial units. It means
inventorying and mapping their characteristic in a dynamical way so the processes
and forces changing the landscapes can be analysed and monitored. Next, each of
the landscape units thus identified, need to be assessed in a participatory process
involving all who are concerned.
This implies a series of methodological very different tasks:
1. Identification (inventorying and classification)
2. Assessing (including atlases)
3. Monitoring (including observatories)
Inventorying means essentially data collection about all aspects of the landscape
and from any kind of source. Classification consists of analysing and ordering all
collected information in a descriptive way (typology) and by mapping landscape
units (chorology). Together these steps result in the identification of the landscapes
that will be assessed and for which specific landscape quality objectives can be
formulated. Monitoring covers both the ‘natural’ changes that go on as the effects of
landscape planning and management.

10.1.1 Making Spatial Units

The geographical space is a continuum in which a variety of more or less well-

delineated areas can be recognised. Land classification denotes the methods to order
these spatial units for better understanding the landscape. Several classification
systems have been developed according to the purpose the will serve and according
to the approaches and methods used (Fig. 10.1).
Landscape classifications are made with a specific goal in mind. This can be for
purely scientific purpose in order to obtain a deeper understanding of the different
aspects of the landscape. Several thematic classifications will be necessary to cover
the whole complexity of the landscape. Examples are the classifications of the
geology, soils, landform, land cover and settlement systems. The classifications of
landscape types and regions aim to make a synthesis. The land units or landscape
units are then considered as holons and are hierarchically ordered. Classifications
with a functional purpose serve specific applications. Typical are landscape assess-
ments and classifications of land suitability. These can be policy supportive, such as
Landscape Character Assessments.
An easy way to combine spatial, thematic and temporal observations is using a
geographical data matrix (Fig. 10.2). Conventionally, spatial units form the rows
10.1 From Identification to Monitoring 267

Fig. 10.1 Approaches and methods used in subdividing space

and their attributes columns. This allows a multivariate analysis using both the
inter-variable covariance and the inter-object relationships (Davis 1986). Analysing
the correspondence between the variables (rows) is similar to the approach in factor
analysis, assuming that the variables are correlated with unknown underlying and
uncorrelated factors in the data set and which can be extracted. In the context of
landscape classification, one could say landscape types are like factors. The attri-
butes and variables define also a thematic ‘profile’ for each spatial unit. The
similarity between the ‘profiles’ allows grouping and defining spatial landscape
units as regions. This can be done in hierarchical steps at different scale levels. Both
ways of analysing the geographical data matrix allow two ways of classifying
landscape units: typology and chorology. Comparing data on the time dimension
allows defining landscape trajectories and paths as discussed in Chap. 7.

10.1.2 Typology and Chorology

Typology and chorology are two fundamentally different forms of landscape

classification (Fig. 10.3). Typology is a classification of items into groups or types
that are clearly distinct from each other according to the attributes used. It is a kind
of generalisation as the groups or types refer to common characteristics. They can
268 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.2 The geographical data matrix combines three dimensions: space, time and feature-
themes (feature space). Observed data are ordered for each time slice in a matrix combining spatial
observations, samples, objects or spatial units with their characteristics measured by attributes
(qualitative, categorical) and variables (quantitative). Comparing variables in columns allows to
define relationships and types, comparing rows allows to define similarities between places and to
define spatial units at different scales

Fig. 10.3 Difference between typology and chorology. A, B, C, D, E, F are landscape types. I, II,
III, IV are regions. The same landscape types can occur in different spatial context. Chorology is
defined by the combination of types and spatial configuration. Regions I, II and IV are composed
of the same landscape types A, B, C and D and have the same typology. However, their spatial
configurations are different, which defines a different chorology. Regions I and III have the same
spatial configuration, but they are composed of different landscape types, both regions are
chorologically different

be considered generic ‘idealised types’ or models of the items at a higher level in

the classification. Zonneveld (1995) speaks of ‘typification’ or non-spatial
10.1 From Identification to Monitoring 269

Typology works differently for discrete objects and continuous features. A

typology of discrete objects is sometimes called categorization and often uses
qualitative attributes. For continuous features, such as slope, water tables, etc.,
there is a fuzzy or gradual transition from one ‘type’ to another (e.g. from a gentle to
a steep slope). In order to map the different types as spatial entities, differentiating
attributes must be used and classes must be defined. This can be done in different
ways and use different methods according to the properties of the attributes used.
For example, the results will be different when the slope degree is measured on an
ordinal scale using categories as ‘flat’, ‘gentle’, ‘moderate’ and ‘steep’, or measured
in degrees or percentage. The number of classes and the methods of assigning
values to classes must be chosen. Thus, different classifications are possible for one
item and also for landscape types in one area.
Typologies are often hierarchically structured into as system defined by the level
of generalisation or scale. Often criteria are used successively, meaning they get a
different significance or weight. Main and subtypes can be defined, as well as
variants, complexes and associations.
Following groups of landscape typologies are common:
1. a geomorphological typology of ‘natural regions’, based on landform, geomor-
phological, geological and climatic processes and of different age;
2. a functional-historical typology of the cultural landscape: landscape types are
defined by the history of occupancy by humans and the ways people organised
the landscape. The functional aspect relates both to ecological, social and
economic aspects;
3. a visual-spatial/physiognomic typology makes types based on perceptual, phys-
iognomic and preferential properties. Criteria as openness, naturalness scale,
variation, stewardship, order, tranquillity, etc. are used.
Chorology considers the geographical context the different landscape types are
spatially arranged in and also relates to scale in a hierarchical topology. Combining
different adjacent typological entities allow defining larger spatial units composed
of complexes or associations at a higher scale-level. These are called chorological
units (from Greek khōros, ‘place, space’). The German geographer Ferdinand von
Richthofen (1833–1905) regarded chorology as a specialisation within geography,
i.e. regional geography, and his pupil Alfred Hettner (1859–1941) considered the
study of regions as the main field of geography. However, ancient and humanist
scholars already used the concept.
The difference between typology and chorology can be understood easily with
the following example. Polder is a general generic term of a landscape type
characterised as a low-lying tract of land enclosed by embankments (dykes) that
forms an artificial hydrological entity. It is a landscape entirely created by humans
and managed by water boards (Dutch ‘waterschap’ or ‘polderschap’). Different
types can be recognised according to the period of land reclamation and the
technology used. Examples are polders formed by land reclamation along coasts
or rivers, others are the result of draining lakes and another type is the result of
landscape restoration after peat extraction. The earliest polders date from the
270 10 Identifying, Mapping and Assessing Landscapes

eleventh century and were land reclamations in the Scheldt estuary in Flanders and
The Netherlands. The technique rapidly spread and today polders are found all over
the world. During that long history, technology improved and different types of
polders emerged. Inundations also caused restoration and the creation of new
polders. Consequently, different spatial arrangements of polders developed,
forming unique landscapes, which are chorological distinct.
In the 1960s–1970s, the interest for regional geography and chorology declined,
but other disciplines took over the chorological classification (see Chap. 2). First, it
was used in land evaluation in developing countries and based on aerial photo
interpretation (Mabbutt 1968; Mitchell 1971, 1973; Howard and Mitchell 1980;
Zonneveld 1972; Zonneveld 1995; Pedroli 1983). Different methods were devel-
oped according to the research domain and goals: e.g. in geological engineering
(Aitchison and Grant 1968), military terrain assessment (Webster and Beckett
1970), biophysical land surveying (Mitchell 1973), assessing land suitability, soil
surveying and land use planning (FAO 1976) and even in economic assessment
(Rossiter 1995). In this approach, mainly physical landscape components were used
to define the spatial units; hence terminology as land evaluation or classification,
terrain classification and land surface classification instead of landscape evaluation
or classification. Nevertheless, the methods can be used for landscape classification
as well.

10.1.3 Basic Methodological Approaches

Basically, two basic approaches are possible in land(scape) classification: by

subdivision and by aggregation (Mabbutt 1968; Mitchell 1973; Zonneveld 1972,
1995, 2005). Zonneveld (1995) refers to descending and ascending classification.
The approach of subdividing the study area follows a holistic method and originated
from the observation of the landscape on aerial photographs (Troll 1939). The
aggregation approach starts by defining basic units that are described by a
predefined set of attributes, which are used to cluster them into groups that can be
represented on a map. Descending: The Holistic Landscape Classification –

Subdividing the Whole into the Parts

With the holistic method, geographical space is subdivided by visually distinct units
(Fig. 10.4). This is the approach used in air photo interpretation and is based on the
Gestalt-capability of our perception to recognise complex patterns. The successive
subdivisions create a hierarchical structure of land units. Heuristic methods of
photo-interpretation can be used to reduce subjectivity by using keys and decision
rules. Nowadays also semi-automatic or hybrid forms are possible where the digital
image sources being processed by segmentation and pattern recognition software
10.1 From Identification to Monitoring 271

Fig. 10.4 The method of the holistic landscape classification using documents that give a synoptic
view of the landscape such as aerial photographs or detailed topographical maps. Visual photo-
interpretation allows delineating major landscape units which may be subdivided subsequently
into smaller units (1–4) and described systematically in more detail and organised in a geograph-
ical data matrix

for automatic delineation of landscape units. The holistic landscape classification

results in a zoning map with sharp boundaries. Gradients and fuzzy borders are not
indicated in the delineation but can be represented as transition zones or described
in the properties of the landscape units.
As the holistic method essentially based upon the knowledge and expertise of the
interpreter, a certain degree of subjectivity is involved. Another criticism is that the
resulting holistic landscape units are somewhat like Pandora boxes (Zonneveld
2005). However, the method has many advantages. Basically, it starts from a
comprehensive observation of the landscape itself, indirectly by the aerial image
and supported by field observations for building the interpretation key and eventu-
ally subsidiary maps. The method is fast to realise a classification covering the
whole area of interest and allows dealing with inconsistencies, defects and all kinds
of technical shortcomings in the data sources available for the classification. Once
the spatial units are defined, a refinement of their delineation, content description is
continuously possible.
272 10 Identifying, Mapping and Assessing Landscapes Ascending: The ‘Parametric’ Landscape Classification –

Aggregating Basic Units

The ascending approach first defines basic units using a set of attributes or, called in
the old terminology of land evaluation, parameters (hence the ‘method of paramet-
ric land classification). These basic units can be used for typification (non-spatial
typology) or can be aggregated to larger spatial units. There are different ways to
define the basic mapping units. Most elementary, samples of point observations of
different attributes are used to define types, from which spatial units are aggregated
(Kilchenmann 1971, 1973). Somewhat similar is sampling raster cells in a regular
grid overlaying theme maps or using grid cells or pixels in digital raster maps or
images to define the types (Van Eetvelde and Antrop 2009).
However, most often, map overlaying is used to combine different thematic
theme maps into a synthetic composite map, where the resulting spatial units (grid
cells or polygons) are characterised by the joint properties of each theme used
(Fig. 10.5). When thematic maps are available in digital format, this is easily done
in GIS-overlay. In raster maps, each cell is described by a series of attributes in the
linked attribute table and statistical analysis is further needed to define landscape
types and to build land units. With vector maps, overlaying always results in a
composite map composed of polygons that can be regarded as land units defined by
the theme attributes used. However, also many meaningless sliver polygons are
created as well (Delafontaine et al. 2009).
In many cases, existing thematic maps are combined, which means using data
sources that are already the result of a complex and often not well-known processes
of data collecting, interpretation and mapping procedures. Most often themes are
chosen referring to the substrate (geology, parent material, soil conditions), land-
forms (elevation, slope degree and orientation), vegetation, land cover or land use.
Some thematic maps can represent different attributes in a combined way: e.g. soils
series represent unique combinations of attributes as parent material, texture,
drainage and profile development.
Although the parametric method may seem more objective than the holistic
method, a lot of subjective input and uncertainty is involved. The thematic base
maps used as sources for the classification are already the result of a long
map-making process. Differences in scale, accuracy, resolution, and time differ-
ences can complicate comparison and combination. Also, the map units shown do
rarely define real terrain ‘objects’ or crisp borders but are the result of interpolation
and cartographic trade-offs. Combining thematic vector maps in GIS also result in
spatial units with crisp borders and creates many sliver polygons of which many are
not relevant. A careful (visual) inspection is needed to see if they represent map
errors and inaccuracies of corresponding to real gradients on the terrain. Anyhow,
the result of automatic map overlaying always demands careful and critical
10.1 From Identification to Monitoring 273

Fig. 10.5 The method of parametric landscape classification using basic units that are the result of
overlaying thematic maps and then aggregated according to their properties. In this example, three
basic map themes are used: the substrate, landform and land cover. On each theme, each map unit
is described by an attribute code. After map overlaying, composite land units are obtained,
characterised by the joint properties of the themes. These are clustered into types and adjacent
spatial units can be aggregated to form larger spatial units Combining Holistic and Parametric Methods

Van Eetvelde and Antrop (2009) used a combination of holistic and parametric
approaches to make a landscape characterisation of Belgium at two scale levels
(Fig. 10.6). The method also used square cells to sample a series of base map of
varying scales, properties and qualities, a method introduced by Kilchenmann
(1973). The use of raster cells (as vector polygons) not only allows integrating
map themes with different forms of representation but allows also to integrate
different scales. This was essential in making a landscape characterisation at the
national level in a federal state as Belgium, as the number of datasets at the federal
level is limited and most landscape classifications are based on maps having an
extent at the regional level, as landscape planning and protection is the competence
of the regions. The methods allowed inter-regional integration to realise a national
overview (Van Hecke et al. 2010).
274 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.6 The stepwise method of landscape classification at two scale levels using cells as basic
units to typify landscapes and combining parametric and holistic methods (Van Eetvelde and
Antrop 2009)

10.1.4 Adding the Third Dimension

Terrain elevation and landform are essential variables in any kind of landscape
classification. Today, Digital Elevation Models contain the basic altimetric
10.1 From Identification to Monitoring 275

information, which can be used to derive various properties describing the land-
form, such as slope degree and shape. Burrough (1986) defined a Digital Terrain
Model (DTM) or Digital Elevation Model (DEM) as “any digital representation of
the continuous variation of the relief over space”. To create a DEM, different
solutions are possible according to the data properties, the data format and structure
in the GIS and the objectives of the study. Basically, three formats of digital models
are used and each has different characteristics in data storage efficiency and
accuracy in representing the relief. Contour lines can be interpolated from point
data but are the least efficient to store and manipulate data. However, they give the
smoothest representation of the landform. Triangulated Irregular Networks (TINs)
represent the surface by a set of contiguous and non-overlapping triangles
connecting the original data points. The TIN model fits triangular polygons to
land facets and calculates automatically slope angle and slope orientation for
each triangle. They have quality properties between GDEM and contour lines
(Ramirez 2006). Grid Digital Elevation Models (GDEMs) represent the relief as
a two-dimensional regular grid whereby each grid cell contains one elevation value.
This format is the simplest and most efficient in terms of storage. GDEMs allow
easy manipulation of the resolution and a broad range of spatial analysis procedures
are available. However, they are relatively poor in visualising the relief.
Since the 1980s Airborne LiDAR technology is used to create detailed digital
elevation models (Ackermann 1999; Drosos and Farmakis 2006; Liu 2008). LiDAR
is an active remote sensing technique for laser altimetry. It registers x-, y- and
z-coordinates of sub-randomly distributed terrain point data. An advantage of
LiDAR altimetry is that it also penetrates vegetation and allows acquiring ground
level points. The raw LiDAR data give the terrain elevation when the laser beam
reaches the ground, but also any object such as vegetation and buildings. This is
called the envelope elevation model of the actual terrain. Various filtering methods
are used to separate vegetation and obstacles points and ground points, so also a
DEM of the ground level points can be created, although some filtering artefacts
remain (Crutchley 2009; Liu 2008).
Because only coordinates are provided and no object information is given, the
raw data have to be interpreted as an aerial photograph. However, the exceptional
planimetric accuracy at centimetre level and the very dense scan pattern allows
identifying small landscape elements. LiDAR-DEMs are used in a broad range of
applications, such as geomorphological research (Kasai et al. 2009), coastal zone
monitoring (Fernandes da Silva and Cripps 2011; Lohr 1998; Mutlu et al. 2008;
Saye et al. 2005; Utkin et al. 2002; Zhou and Xie 2009), forest fire management
(Mutlu et al. 2008; Utkin et al. 2002), infrastructural and environmental projects
(Challis et al. 2008; Wehr and Lohr 1999) and in landscape archaeology (Bewley
et al. 2005; Challis 2006; Devereux et al. 2008; English Heritage 2009, Gallagher
and Josephs 2008; Harmon et al. 2006; Powlesland et al. 2006; Gheyle et al. 2014).
The first example (Fig. 10.7) shows a LiDAR DEM overlaid with a historical
topographical map of 1909. The contour lines of the map were drawn from
triangulated elevation points during the terrain survey. The result is a smoothed
topography depending on the equidistance of one meter between the contour lines.
276 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.7 LiDAR-GDEM overlaid with a historical topographical map of 1909. The LiDAR data
have an average sampling density of 1 point/2 m2, resulting in an altimetric accuracy varying
between 7 and 20 cm and a planimetric accuracy of 50 cm. The GDEM shows how fields are
related to the micro-topography, revealing the artificial convex field topography resulting from
digging drainage ditches at the borders and concentric ploughing, Artefacts from filtering out
recent buildings are indicated by arrows (After Werbrouck et al. 2011, topographical background:
historical Belgian topographical base map at scale 1:20,000, Dépot de la Guerre 1909, National
Geographical Insitute)

The LiDAR point data were filtered to remove vertical objects as trees and
buildings to obtain the ground surface with the height accuracy of 7–20 cm. The
field borders and hedgerows of the map correspond in detail with the micro-
topography, illustrating the accurate adaptation of the land use.
Analytical hill shading is a common technique to enhance micro-topography and
small elements on planar LiDAR DEMs. The second example (Fig. 10.8) shows a
LIDAR DEM map used to survey a battlefield from the First World War visualised
with Sky-view factor technique based on diffuse illumination so all directions are
enhanced equally (Zakšek et al. 2011; Kokalj et al. 2011). This allowed detecting
buried field structures and small above-ground features such as shell holes, traces of
10.1 From Identification to Monitoring 277

Fig. 10.8 A hill-shaded LiDAR-DEM of the battlefield of the First World War north of Antwerp
(Belgium). Diffuse hill shading reveals old buried field structures, above-ground features such as
shell holes, remains of narrow-gauge railways and trenches that are not visible on the aerial
photographs. A: anti-tank ditch, B: defence line 1917, (a) tranch, (b) bunker, (c) shell hole, (d)
trace of a narrow-gauge railway (After Gheyle et al. 2014)

narrow-gauge railways and trenches that are invisible on the aerial images taken
during the Great War (Gheyle et al. 2014).

10.1.5 Dealing with Borders

The borders between landscape features can be crisp or fuzzy gradients, tangible of
immaterial as many administrative divisions. Overlaying thematic maps generates
many sliver polygons of which many have no significance. Some borders or
278 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.9 Geographical regions of Estonia with transitional borders as defined by thematic map
overlaying. Borders become transitional zones when the differentiating themes (I–IV) do not
overlap within a width of 10 km. When three or four themes change, the border zone (V ) gives
a stronger differentiation between de adjacent areas than when only two themes change (VI)
(Gran€o 1929, 1952; Gran€o and Paasi 1997)

transitions are important, others are not. Tracing borders of map units is always a
risky decision. Delineating geographical regions is a problem from the beginnings
of landscape science. Already in 1929 Johannes Gran€o (1929, 1952, Gran€o and
Paasi 1997) formulated a method to delineate geographical regions by overlaying
thematic maps, using the sliver polygons to define core areas and transitional
borders (Fig. 10.9). Gran€o called the themes elements and the spatial units regions.
The area where all theme units overlap forms the core of a ‘pure’ region. He defined
also transitional zones when the width of the ‘slivers’ formed by the borders of the
themes exceeds 10 km.

10.1.6 A Hierarchy of Landscape Units

As landscape elements and spatial units vary in size, there is an optimal scale to
view them in a comprehensive way. Zooming in and out is a common practice to
obtain a better understanding of the landscape structure. Forman and Godron (1986)
referred to the technique of shuttle analysis. The classic method in geography to
represent chorological classifications is using different map scales. This follows
10.1 From Identification to Monitoring 279

entirely the idea of a hierarchical structure of embedded holons (Naveh and

Liebermann 1994) (see Chap. 5).
A great variety of hierarchical land classification systems has been developed in
regional geography and later in land evaluation. Terminology and map scales for
representing the hierarchical units varied widely. Several extensive overviews have
been given (Mabbutt 1968; Mitchell 1973, 1991; Zonneveld 1972, 1995, 2005;
Stewart 1968; Fernandes da Silva and Cripps 2011). The spatial units that are
represented on maps are called land units according to many methods of land
classification or land evaluation. They are considered as holistic ‘black boxes’
(Zonneveld 2005). In geography, these units are referred to as landscape units,
which is more appropriate considering the descriptions at various scale levels.
Jongman and Bunce (2000) considered the traditional landscape classifications
qualitative and intuitive and suggested they could be improved by using quantita-
tive classification methods from ecology.
Table 10.1 gives a synopsis of the most common units and their logic to structure
them hierarchically. Note that all scale limits are approximated. Table 10.2 gives a
brief overview of classification systems that were common in the early days of land
The smallest unit is called ecotope, geotope, site or land element. Essentially this
is the smallest area that will be observed and its size also depends on the measure-
ment to be made, e.g. a slope degree. In soil science, this corresponds to the concept
of a pedon, which corresponds to 1 m2 and a depth of 1–2 m. For land cover, this
corresponds to a field or area with a homogeneous land cover type.
A land facet is usually defined as a homogeneous spatial unit characterised by
slope, soils and land cover. Examples are meadows on flat alluvial soils, or
woodland on steep, rocky slopes. Contiguous land facets form series can be
arranged by topography and form a toposequence. Such a larger unit is called a
land catena. In general, land facets and land catenas are not named but
characterised by a code built from the attributes used to describe them (Fig. 10.10).
Contiguous land facets or catenas linked by horizontal relations define the
functioning of a larger unit, called a land system. Examples are a river valley, a
dune belt and a village territory. Landscape systems are often called according to
the characterising landscape type or place. The term system is somewhat mislead-
ing since landscape units at all levels can be considered systems.
A land region is a unique geographical association of land systems. It has a
distinct character and identity and often has a proper name.
Contiguous land regions that share a common genesis or history may form a land
province. Many historical landscapes and counties belong to the level of a land
Land divisions are still larger grouping mostly corresponding to major geolog-
ical structures as a tectonic basin or mountain belt.
Finally, land zones refer to large climatic and vegetation zones.
280 10 Identifying, Mapping and Assessing Landscapes

Table 10.1 A common hierarchy of land units

Scale of map
Land unit Main characteristic representation
Geotope, Smallest tract of land allowing a description > ¼ 1:10,000
Land facet Unique combination of slope, substrate (soil) and land 1:10,000–1:50,000
cover/land use
Land catena Toposequence of land facets
Land Spatial and functional associations of land facet and/or 1:50,000–1:100,000
system -catenas
Land region Unique spatial and geographical association of land
Land Contiguous land regions with a common genesis or 1:100,000–1:1,000,000
province history
Land Large geographical units based on landform and
division geology
Land zone Climatic and vegetation zones < 1,000,000

10.2 Landscape Identification

10.2.1 Atlases, Catalogues, Observatories

Identifying the landscapes is based on inventories of all data available about the
landscapes. The data consist of descriptions, field observations, maps, illustrations,
photographs and narratives. This information is ordered systematically in a land-
scape inventory, which is conceived for a specific scale and planning goals.
Inventories use methods of landscape classification and exist in different forms.
Most of the information about landscapes at the national and international level
is collected top-down by experts. To cover vast areas in a consistent way, inven-
tories are based on available data that use a systematic description, i.e. maps with a
comparable legend. As an example, we can mention the European landscape map
(LANMAP2) (Mücher et al. 2003). General data about the physical environment
such as geology, soils, terrain elevation and land cover are often available on a
continental scale and allow transborder consistency in their legends. Land cover
data are widely available with remote sensing imagery, allowing making regular
updates. Often land cover is used as a proxy of the cultural features of the
landscapes. This is the case for example with the European CORINE Land Cover
data that are used and reused in various ways in agriculture, forestry and urban
planning as well as for making climate scenarios. The complementary LUCAS-
programme (Palmieri et al. 2011), based on stratified sampling, makes a clear
distinction between land cover and land use to monitor changes between the main
categories. Land cover is also a common theme that allows studying changes over
vast areas. An example is the international Land Use and Land-Cover Change

Table 10.2 Comparison between selected hierarchical land classification systems

System > ¼ 1:10,000 1:10,000–1:50,000 1:50,000–1:100,000 1:100,000–1:1,000,000 < 1:1000 000
CSIRO 1957 Site Land unit Land system Complex land system Compound land
DOS 1962 – – Land facet Land system Land region, province,
division, zone
Troll 1963 Ökotop Fliesengefüge Naturräumliche Gruppen Naturräumliche –
Landscape Identification

Haupteinheit Haupteinheiten
Vinogradov et al. 1962, Facies Urochishche Mestnost Mestnost Landscape
Vinogradov 1966, 1976
Nakano 1962 Landform type Series Association Section Province
MEXE 1965 Land element Land facet ¼Land system Land system land region, province,
division, zone
Neef 1967 Ökotop Microchore Mesochore Macrochore Megachore
PUCE 1970 Terrain component ¼ Terrain unit Terrain pattern Province
Zonneveld 1972 Ecotoop, site Land facet Land system Landscape –
Howard and Mitchell 1980 Land element, land Land facet, land Land system Land region Land province,
clump catena division, zone
Fernandes da Silva-Cripps Land element; Land Land facet Land catena Land system Land region; Land
2011 subfacet; Land clump province; Land zone
After Vinogradov et al. (1962), Vinogradov (1966, 1976), Christian and Stewart (1968), Aitchison and Grant (1968), Beckett and Webster (1969), Howard and
Mitchell (1973), Mitchell (1973, 1991), Pedroli (1983), Zonneveld (1995)
CSIRO Commonwealth Scientific and Industrial Research Organization (Christian and Stewart 1968)
DOS (Commonwealth) Directory of Overseas Surveys
MEXE Military Evaluation Experiment (Beckett and Webster 1969)
PUCE Pattern Unit Component Evaluation (Aitchison and Grant 1968)
282 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.10 A land catena consists of a series of land facets (1–9) organised by topography. Each
land facet is defined by a unique vertical association of attributes, i.e. soil, slope and land use.
A arable land, M meadow, B build land, L scrubland, D deciduous woodland, C coniferous
woodland, b colluvium, c inceptisol deep, d inceptisol stony, e inceptisol shallow and very
stony, a alluvium (Example Le Roptai catena in the Calestienne landscape region, Belgium, ©

(LULCC) programme in the framework of Global Change studies (Lambin and

Geist 2006).
The spatial resolution of inventories on international (continental) scales is
insufficient to represent individual elements of the landscape and data is often
aggregated in abstract spatial units such as square kilometre cells. The goal is to
compare and integrate countries and the results are used for international planning
policies. The whole procedure is top-down and does not include public participation
and the results are often difficult to interpret and for laypeople. At regional and local
levels, landscape inventories need to be more detailed and concrete, complying
with following:
• conceive public’s participation from the beginning
• collect already existing, but partial and scattered information of all kinds already
familiar to the local actors
• update data where necessary
• define a baseline, which can be used to compare different scenarios
• define a spatial reference frame to integrate all information and link compre-
hensive maps and visualisations
• clearly make the distinction between descriptive information and the valuation
Many methods for landscape identification and assessment have been developed
in a great diversity of approaches, formats, styles and names. Antrop (2004) gave an
overview of landscape research in thirteen European countries involving making
landscape inventories. Groom et al. (2006) made a methodological overview of
existing classifications in Europe and how they used the approaches of Landscape
Character Assessment. The conference Living Landscapes (UNISCAPE 2010)
gave the state of the art 10 years after the European Landscape Convention was
10.2 Landscape Identification 283

introduced. Most common is the Landscape Character Assessment (LCA), often

based on the examples in England and Scotland (Swanwick 2004). Warnock and
Griffiths (2015) describe how the process of landscape characterisation developed
in LCA and has been refined in the Living Landscape Project. In non-English
speaking countries concepts as landscape ‘observatories’ (observatoire du paysage)
and landscape ‘catalogues’ are used, which mean more than just an inventory, but
refer to a whole system of landscape assessment and monitoring changes. The
Landscape Catalogues of Catalonia is a good example (Nogué et al. 2016). In the
Netherlands, the concept of a landscape biography (Landschapsbiografie or
Biografie van het Landschap) was introduced, in particular in applied interdisci-
plinary research for heritage management (Kolen et al. 2015).
Three main types of landscape inventories can be recognised: the landscape
atlas, the landscape biography or catalogue and landscape photography databases:
• Landscape atlases present all information in a series of thematic maps, often in a
GIS-database and web-environment and are becoming more and more
• Landscape biographies and catalogues have the format of a regional mono-
graph, integrating text, maps, photographs, figures, tables and bibliographic
• Landscape photographic inventories are databases of geolocalised photographs
of selected spots to represent the different landscape types or regions. Often
historical archive photographs are used and rephotographed to illustrate land-
scape changes. This can be the base for a photographic landscape observatory
and to set up a monitoring system.
Probably the first example of a holistic landscape atlas is the Atlas of the Irish
Rural Landscape (Aalen et al. 1997). It consists of a series of thematic maps
combined with explanatory essays. In France, Yves Luginbühl (Luginbühl et al.
1994) proposed a method for the elaboration of landscape atlases on a regional
basis. The approach is holistic and integrated and includes using map-based
landscape classification, identification and characterisation, but also history and
dynamics. Important is also the use of iconographic representations and participa-
tion of the local public. The method was promoted by the French Ministry of
Ecology and Sustainable Development and applied in several French regions. It
was also adapted for the landscape atlases of Wallonia in Belgium (see CPDT
2007–2012 and Catalonia in Spain (Nogué and Sala 2006;
Nogué et al. 2016).
Various types of landscape atlases can be recognised. In the simplest form, they
consist of a synthetic landscape classification map with an extensive description of
the landscape units. Sometimes different ‘layers’ of the landscape are discussed
separately. This type is often included as a theme in National Atlases. These
national or regional classifications used different data sets, methods and landscape
types, which make merging adjacent maps difficult or impossible, as was shown
with the European Landscape Character Assessment Initiative ELCAI (Pérez-Soba
and Wascher 2005). Exemplary is the case of Belgium, which almost disappeared in
284 10 Identifying, Mapping and Assessing Landscapes

international overviews since it became a federal state and each region adopted its
own landscape policy (Van Eetvelde and Antrop 2009). In most European coun-
tries, landscape policy became a responsibility of the regions, so the concepts and
methods of landscape atlases vary accordingly.
Photographic inventories are collections of very diverse iconographic docu-
ments, which can be geotagged in various ways. They are often intended to study
the history of a landscape by its visual representations or to build a base for
monitoring landscape change. Specific methods and guidelines make that photo-
graphs are taken in a systematic manner, including metadata and eventually addi-
tional field information. Examples of elaborated methods are the Norwegian system
(Puschmann and Dramstad 2003) and the French photographic landscape observa-
tory (Observatoire Photographique Transfrontalier des Paysages 2008). Often, new
photographic recordings are made on the location of historical archive photographs,
such as postcards. Examples of such rephotographing are Recollecting Landscapes
in Belgium (Uyttenhove 2006), the Visual Monitoring of Finnish Landscapes
(Heikkilä 2007) and the photographic observatories of transborder landscapes in
river basins, as for the Semois river between France and Belgium (Lobet et al. 2006)
and the Transborder Observatory of the Regional Landscape Park of Hainaut. In
France, a specific methodology was developed based on nineteen landscape itiner-
aries in the whole country (MEEDDAT 2008).

10.2.2 Generic Traditional Landscape Types

Traditional landscapes refer to the pre-industrial and pre-urbanised landscapes,

which have been shaped during centuries mainly by agrarian and pastoral commu-
nities (Antrop 1997). Their spatial organisation shows a great ecological coherence
and strong relations exist between substrate, soil, landform, land cover and land use.
This makes that classifications of traditional landscapes closely follow the physical
and natural qualities of the land, modified by cultural variations as differences in
settlement patterns and field systems (see also Chap. 9). Traditional landscapes
express regional character, culture and history, and different types can be
During the long history, landscapes were remodelled and adapted to changing
natural conditions and societal needs. The landscape that we see can be compared to
a palimpsest where older layers partially show through more recent ones (Claval
2005) (see also Chap. 7). Vos and Meekes (1999) proposed following succession:
• Natural/prehistoric landscape (from Palaeolithic till ancient Greek times)
• Antique landscape (from ancient Greek times till early Mediaeval times)
• Mediaeval landscape (from early Mediaeval times till Renaissance)
• Traditional agricultural landscape (from Renaissance till the nineteenth century,
sometimes till today)
10.2 Landscape Identification 285

Table 10.3 Main groups of common historic-genetic traditional landscapes in Europe

Landscapes of agricultural land reclamation and development
• Open field landscapes, champion landscapes
• Polder landscapes, land reclamation on sea and rivers
• Landscapes of peat extraction
• Hedgerow/bocage landscapes, enclosures; ancient landscape
• Landscapes of forest clearing
• Mosaic landscapes (mixture fields, pasture and woodlots)
Landscapes of (animal) husbandry
• Common pastoral landscapes – heathlands, wood-pastures
• Enclosure-landscapes, enclosed common land
• Mountain pastures of transhumance
Landscapes of Mediterranean polyculture
• The selva castanile and chataigneraie
• Montado – dehesa
• Huerta
• Irrigated/drained delta-landscapes
• Cultura promiscua
• Estates, parks, gardens, . . .
• The landscape settings of certain heritage
• Monuments

• Industrial landscapes (mostly from mid-eighteenth till mid-twentieth century, in

many places till today)
Particular in Europe, several historic-genetic landscape types have been identi-
fied (Table 10.3). Emanuelsson (2009), Butlin (1993), Lebeau (1979), Rackham
(1986) and many others made typological classifications. For example, Lebeau
(1979) recognised following rural landscape types in Europe (excluding the Soviet
• Landscapes of enclosures and dispersed settlement, grasslands dominate
• Ancient open field landscapes with grouped settlement; evolved to dispersed
settlement, field re-allotment and enclosure
• Landscapes of open fields with grouped settlement and large areas of cropland
• Open field landscapes transformed by collectivisation
• Reclamation landscapes of forest, heathland and marshland with linear settle-
ment and stretched fieldstrips
• Open Mediterranean cropland, eventually with arboriculture, small patches of
huertas and clustered and dispersed settlement
• Landscapes of coltura promiscua
• Estate landscape of montado and dehesa
286 10 Identifying, Mapping and Assessing Landscapes

Generic landscape types are holistic and can only be described fully by all
composing characteristics including their relationship and development. For exam-
ple, an ‘open field landscape’ consists of arable infields surrounded by extensively
used outfields. Both are common lands for the community, which has specific
property and use rights. Often the population lives in a clustered settlement in the
centre of the territory, which was often a green-village type (see Chap. 9). The
infields formed blocks of open cropland, i.e. without physical borders of hedges or
ditches. The outfields were used as common grazing land. This type of landscape
originated in the Middle Ages as a collective agricultural system based crop
rotation around the centrally localised village. The oldest form was a two-field
system, which used half of the land to grow crops, while the other half was left
fallow to recover. To increase the production a three-field system was introduced.
The land was divided into three fields: one-third lay fallow, one-third was sown in
autumn with crops as wheat, and one-third was planted in spring with oats and
legumes. The system became common in regions where there was sufficient rainfall
during summer. The ownership of the fields was distributed equally over the
infields. In continental Europe, the system of inheritance caused a continuous
land subdivision of the properties resulting in narrow field strips arranged in blocks.
Oliver Rackham called these landscapes ‘champion’ landscapes (Rackham 1986).
In France, they are called ‘campagne’. Land consolidation, up-scaling and
mechanising of agriculture only kept the major structure of this type and many
characteristic elements vanished or were transformed. In some areas of
collectivisation, these historic field structures have been wiped away completely
(Lipsky 1995; Palang et al. 2006), but the village structure remained.
Each place and landscape unit have a unique composition, configuration and
history. For each, a unique narrative can be told. Generic landscape types aim to
define the most common characteristics of landscape units that developed in a
similar way and time period.

10.2.3 Mapping New Landscapes

It is not obvious to make a classification of the new landscapes that emerged with
the industrialisation and urbanisation and which transformed, overlaid and eventu-
ally replaced the existing traditional landscapes. The resulting landscapes are
eclectic complexes with mixed characteristics and many new elements and struc-
tures. The typical problem is demonstrated by the attempts of defining the land-
scapes of the urban fringe and ‘rurban’ landscapes (see Chap. 7). Vos and Meekes
(1999) recognised following postmodern landscape types:
• the industrial production landscapes: landscape as an industry
• the overstressed multifunctional landscapes: landscape as a supermarket
• the archaic traditional landscapes: landscape as a historical museum
• the marginalised vanishing landscapes: landscape as a ruin
• the natural relict landscapes and new nature: landscape as a wilderness
10.2 Landscape Identification 287

Castells (2000) argued that our society develops as a network in which the
relations are becoming less place-bound. Accordingly, Hidding and Teunissen
(2002) recognised four interlaced networks that define the contemporary
• The economic network in which accessibility is the most important factor. This
network controls the development of the urban and industrial corridors.
• the transport network as the complex of all kinds transport and utility networks.
Also referred to as ‘grey’ network
• the ecological network corresponds to the corridors greenways and open
non-built land or ‘green’ network
• the water network or ‘blue’ network.

10.2.4 Mapping the Visual Landscape

Most landscape classifications result in maps, i.e. represent the landscape in a

bird’s-eye perspective. When dealing with properties that are specific for the
horizontal, ‘everyday’ perspective of most people, the map representation becomes
more abstract and alternatives have to be found. Besides mapping the visual
landscape, also various forms of landscape visualisation developed, in particular
for the purpose of planning, design and management (Nijhuis et al. 2011). Mapping the Visual Landscape: The Beginnings

Gran€o (1929) gave the earliest examples of mapping perceptual properties of

landscapes. De Veer and Burrough (1978) gave the first overview of methods
used in The Netherlands for mapping the visual landscape or landscape physiog-
nomy. They defined three approaches to the problem (Fig. 10.11):
1. The compartment approach makes a landscape classification using holistic
methods and based on model 4 (see Chap. 8) for the analysis of landscape
patterns, i.e. reducing the landscape to three primitives: masses, spaces and
screens. This approach can start from field observations or analysis of topo-
graphical maps and aerial photographs. Space compartments are described by
their size (area, defining the openness) and shape and the nature of their borders
with masses (transparency).
2. The field of view approach is essentially field-based. The field of view is
determined at regular or selected positions in the landscapes. The methods use
sight lines to describe visual properties such as visibility, viewshed and
288 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.11 Three mapping approaches for mapping visual landscape properties: (a) the compart-
ment method, (b) the field of view method: the dots s are the viewpoints, (c) the grid cell method.
(1) open spaces, (2) built zones, (3) vegetation masses

3. The grid cell approach samples the landscape by an overlay of a tessellation of

cells for which variables that are significant for physiognomic properties. Most
often a square grid is used overlaid on a detailed topographical map. Variables
used are indicators of landscape diversity and complexity and expressed as
landscape metrics.

Most of the early methods for mapping landscape physiognomy used field
observations and analogue data sources as topographical maps and were very
time-consuming and labour intensive. Using aerial photographs and terrain photo-
graphs (Palmer and Lankhorst 1998) allowed improving the efficiency through
indirect analysis of visual features in the landscape. The development of GIS and
spatial analysis rapidly allowed automated analyses of sight lines and viewsheds
(Burrough et al. 1982; Burrough 1986; Tomlin 1990). However, detailed and
reliable digital data were often lacking to obtain acceptable results. The Impact of Digital Terrain Models and GIS-Analysis

New algorithms and better digital data, detailed digital elevation models (DEM), in
particular, caused an exponential growth of the GIS-based analysis of the visual
landscape and landscape visualisations (Nijhuis et al. 2011). New methods and
concepts were introduced, e.g. using the space syntax introduced by Hillier and
Hanson (1984). It refers to theory and methods for the analysis of spatial configura-
tions and originally conceived for architectural design and analysing urban environ-
ments. Concepts related to spatial syntax are indicated with an asterisk* in following
list, which gives a selection of the most important concepts in the GIS-based analysis
of the visual landscape (Table 10.4). Note that some are not always exactly similar to
the concepts used in perception studies as discussed in Chap. 6.
GIS allows the calculation of viewsheds or isovists for any point in a digital
terrain model (Fig. 10.12). When more than one viewpoint is used, also the term
10.2 Landscape Identification 289

Table 10.4 Concepts related to the GIS-based analysis of the visual landscape
Axial line* The longest sight line
Axial map* Topological graph of a street or road network representing the longest and
fewest axial lines
Central line of The line of vision that bisects the view
Convex space A space where all positions (‘points’) are visible for each other.
Isovist field Cumulative field of isovists derived from multiple viewpoints
Isovist* Sight field polygon representing the (eye-level) panoptical view from an
observer at a certain viewpoint.
Obstacle or Non-transparent volume or mass that limits the view.
Point depth Delineating the degree of visibility from every point in the (public) space
Sight line The direct line of vision between the observer’s eye and a target object or
point in space.
Transparency Degree by which sight lines can penetrate obstructions.
View The (composed) areas of the landscape covered by a horizontal, binocular
field of vision with a given angle (usually varying between 20 and 60
Viewshed The (GIS-)representation of areas of the landscape visible from one or more
Virtual 3D landscape representations created by computer modelling and
3D-landscapes rendering.
Visibilitime View defined by a mobile observer and representing The time that points
are visible.
Visibility The geographical extent an object or an observation position can be seen.
Visible form The visual appearance of the three-dimensional form (objects, volumes) in
the landscapes and its relationship with the context space, expressed as the
structural organisation.
Visualscape The spatial representation of any visual property generated by, or associated
with, a spatial configuration (Llobera 2003)

cumulative viewshed is used. The total viewshed refers to a cumulative viewshed

using all points from the Digital Elevation Model (Llobera 2003). When viewsheds
are combined by a logical function, the term multiple viewshed is used (Ruggles
et al. 1993). Joly et al. (2009) make the distinction between the active and passive
viewsheds. The active viewshed refers to the area that can be seen from a given
observation position and corresponds to the prospect from that position; the passive
viewshed corresponds to the area from where a given target can be seen and
corresponds to the visibility of the target.
Properties of the viewshed and the viewpoint position can be used for a land-
scape character analysis reflecting the visual properties of the landscape. However,
the reliability and validity of the outcome rely basically on the quality of the digital
terrain model used.
290 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.12 The view and

isovist concept: (1) view or
isovist from viewpoint VP,
(2) obstacle limiting the
view, (3) part of the space
not seen (view shadow)
(After Llobera (2003))

Fig. 10.13 gives an example of the use of cumulative viewsheds calculated from
the villages on the island of Paros (Greece). Note that the municipal boundaries
follow approximately the watersheds and the high density of settlements on the
eastern slopes of the island. Generally, about one-third of the municipal area falls in
the viewshed of the village, but viewsheds extend over two to five other territories
in the eastern part of the island. Nevertheless, the intervisibility amongst the
villages is rather limited. Analysing and Characterising the Visual Landscape

Ode et al. (2008) proposed visual landscape indicators in assessing the landscape
visual character, which they define as “The visual expression of the spatial ele-
ments, structure and pattern in the landscape.” (Ode et al. 2008, p. 90). They
identified nine visual concepts that characterise the visual landscape and which
are related to theories of landscape preference and experience: complexity, coher-
ence, disturbance, stewardship, imageability, visual scale, naturalness, historicity
and ephemera (Fig. 10.14) (see also Chap. 6). These concepts all refer to holistic
aspects of the landscape structure. They are interrelated and some even overlap,
while others are opposites. Similar concepts have different names in the literature
(Tveit et al. 2006). For each concept, several indicators can be formulated using
different kinds of data sets, dimensions of analysis and landscape attributes (Tveit
et al. 2006). Ode et al. (2008) warn that the indicators should be used with great care
as they make a reduction of the holistic experience observer’s have of the land-
scape. They suggest an indicator selection (‘filtering’) based on criteria that identify
indicators that are suitable for a specific project and landscape context. The filtering
follows a sequence reducing the selection in each step: indicators should have a
clear theoretical base, be transferable between landscapes, be quantifiable, mappa-
ble, available and relevant. Fry et al. (2009) demonstrated the common ground
10.2 Landscape Identification 291

Fig. 10.13 Cumulative viewsheds from villages on the island of Paros (Greece): (a) village
centre, (b) municipal boundaries. The visible area is shaded according to the number of times it
is seen from one or more villages (After Sevenant and Antrop (2007))

Fig. 10.14 Concepts

characterising the visual
landscape and their
interrelations (After Ode
et al. (2008))
292 10 Identifying, Mapping and Assessing Landscapes

between these visual landscape indicators and ecological landscape indicators.

Tveit (2009) showed how two indicators of visual scale (percentage open land in
the view and the size of the landscape rooms) predict the landscape preferences.
Nohl (2001) defined four aesthetically oriented prototypes of future landscapes:
the traditional cultural landscape (based on the aesthetic category of ‘the beauti-
ful’), the spontaneous landscape (based on the category ‘the (new) sublime’), the
urban-industrial landscape (based on the category ‘the interesting’) and the rural
functional landscape (based on the category ‘the plain’) (see also Chap. 6).

10.2.5 Mapping the Mindscape

In Chap. 8, we argued that the map is not the landscape but a representation affected
by many choices made by the mapmaker. Maps are tools to visualise the landscape
in a bird’s-eye perspective, to make immaterial phenomena and structures visible,
and to communicate with others about the landscape. Maps represent material and
intangible landscape features and use a specific graphical language. With geograph-
ical information systems and computer tools, mapmaking became easy for every-
one. Non-cartographers often forget that some rules apply in good mapmaking.
Abundant map producing making all possible combinations and using the default
settings is not a good practice and does not help better understanding the landscape.
It is easy to lie with maps, also by ignorance.
Mental mapping and cognitive mapping refer to methods of representing
mindscapes in map form (Soini 2001; Gould and White 1974). From this perspec-
tive, the landscape is a map full of local knowledge and social values. This
information is subjectively experienced, is different between insiders and outsiders,
between lay people and experts. Social knowledge and values about the landscape
are contextual, culturally-bound, place-related and vary spatially (Plets 2013; Tuan
1977). To capture this knowledge, active participation of the local inhabitants is
essential (Sauer 1925). Basically, two approaches are used in mental mapping
(Gould and White 1974). In one approach the participants draw map sketches of
their landscape. The second approach is based on surveys using questionnaires to
capture the cognition of the respondents in such way it can be located. Then,
statistical trend surface analysis is used to create density maps of that knowledge.
Besides using cognitive mapping in fundamental research on environmental per-
ception and preference, it proved also useful in urban planning and design (Lynch
1960). The first approach led to the development of participatory mapping methods.
With the help of GIS-functionality, it became easy to link local (mostly qualitative)
and expert (quantitative) knowledge (Craig et al. 2002). Thus, participatory geo-
graphical information systems (PGIS) and public participation GIS (PPGIS) devel-
oped (Brown and Reed 2009) (see Chap. 12). Successful applications in
transdisciplinary assessment of natural resources in rural communities (Fagerholm
and Käyhk€ o 2009) and archaeological surveying for heritage protection (Plets
10.2 Landscape Identification 293

Corbett (2009) gave an overview of the methods of participative mapping,

illustrated by cases of good practice. Following methods are ranked according to
increasing IT-skills needed:
1. Ground mapping: a basic mapping method involving community members to
draw maps on the ground using any available materials.
2. Sketch mapping: freehand drawings are made on large pieces of paper and from
memory. They show the key features significant for the community and size
and position reflects their importance. Hence, they do not rely on exact mea-
surements, don’t have a consistent scale and are not geo-referenced.
3. Transect mapping: a spatial cross-section of the community territory depicting
geographic features along the transect (e.g. infrastructure, local markets,
schools, land use and vegetation. Transect mapping can be done while walking
and discussing with the locals.
4. Scale mapping using existing maps: a topographical map is used as a base map
and elements are drawn on it as accurately as possible, allowing
geo-referencing. Orthophoto maps, geo-referenced aerial photographs and
detailed satellite imagery can be used for the same purpose.
5. Using aerial and remote sensing images: oblique aerial photographs and
satellite imagery are not scale-consistent and geometrically distorted. Using
these as reference base demands some help to interpret and precaution in
using them.
6. Scale mapping using survey techniques: when no topographical maps are
available, rudimentary maps can be made from scratch using surveying tools
and GPS.
7. Participatory 3-D modelling (P3DM): these are stand-alone three-dimensional
relief models at a reduced scale. They are constructed from the contour lines on
a topographic map and modelled using cardboard and paper-maché or foamed
plastic. The model is finished with adding small elements and paint. The
participation consists in making the model by the local community;
8. GPS mapping: a GPS receiver is used to capture the exact position of the
features of interest. Coordinates are stored in a digital format to be used in
9. Multimedia mapping: the base maps are shown on a computer screen and
interactively features can be added or additional information can be linked
and consulted such as video, photos, text and theme maps (soils, geology, etc.).
10. Participatory geographic information systems (PGIS) are computer-based sys-
tems that capture, manage, analyse, store and visualise geo-referenced spatial
information from a variety of sources. Specially trained PGIS practitioners
from outside the community are needed to help the local community using the
11. Internet-based mapping is similar to PGIS but uses web-based applications
(e.g. Google Maps and Google Earth). The participation is less based on
interactive communication with the local community (Figs. 10.15 and 10.16).
294 10 Identifying, Mapping and Assessing Landscapes

Fig. 10.15 Example of a mental map based on interviews using method 2 described above

10.3 Landscape Assessment

10.3.1 Attributes, Variables, Indicators and Criteria

Attributes describing properties of an item (element or spatial unit) are used to

assign that specific item to a class. Formally this is done using techniques for
clustering and ordination. Some landscape components are discrete objects; others
are continuously varying (as landforms) (see Chap. 8). Some attributes describing
spatial entities may have no spatial dimension. For example, architectural style,
building height, age, function and condition may be significant to classify different
buildings as discrete objects but have no spatial dimension. Vink (1980) and
Kwakernaak (1984) defined different groups of attributes according to the ability
to map spatial units and according to their purpose:
• Differentiating attributes are the ones used to define and delineate spatial units;
• Purely descriptive attributes are linked to previously defined spatial units and
have no effect on their delineation;
• Diagnostic attributes are attributes used, eventually in combination, to formulate
an indicator describing indirectly some characteristic. For example, age,
10.3 Landscape Assessment 295

Fig. 10.16 Example of a collective participatory map identifying landscape elements important
for the local identity, using methods 4 and 4 discussed above (Based on Loupa Ramos et al. 2017)

function and condition of a building can be used as an indicator to describe the

degree of decay of a settlement.
The term attributes refers to qualities and used here in a general sense, covering
all scales of measurement. Sometimes, it refers only to qualitative descriptions (the
nominal or Boolean scale of measurement), while the term variables is used for
quantitative attributes (measured on ordinal, interval and ratio scales). The term
criterion is used to refer to attributes and variables that were selected for a
multivariate analysis such as in multi-criteria assessment or clustering. Combined
variables are referred to as factors or indicators.
The description of spatial entities with attributes is done systematically using a
geographical data matrix (Fig. 10.2), which is consistent with database structures in
GIS and matrices for statistical analysis: the spatial entities, places or objects form
the cases (rows), and all attributes the variables (columns). These terms have
different meaning according to the context they are used, in particular when
combining applications in natural and social sciences, which is often the case in
transdisciplinary landscape studies.
Attributes, variables, indicators and criteria are all used to describe landscape
properties, characteristics and qualities. In landscape assessment, the English word
quality is used in a neutral purely descriptive sense but may cause confusion when
translated in other landscapes where it often gets a connotation of value (better-
worse). Attributes and variables are sometimes used as synonyms, but also to
296 10 Identifying, Mapping and Assessing Landscapes

Table 10.5 Levels of measurement

Level of
measurement Properties Examples/comments
1. Nominal Assigning qualitative attributes to Land use categories, landscape types,
groups, determining equality of cate- LCA and HLC types and units, styles.
gory. Counting frequency of items.
2. Ordinal Ranking of classes using operators ¼, Slope classes: flat, gentle, steep
>, <; counting frequencies of classes
3. Interval Measuring quantitative variables using Sometimes also referred to as scores.
an arbitrary zero and a scale with a Negative values are possible,
constant interval: addition and subtrac- e.g. altitude, temperatures in  C or  F.
tion arithmetic operators are applicable the psychometric 5–point Likert-scale
is considered as an interval level
4. Ratio Measuring quantitative variables using Distances, areas, temperature
an absolute zero and a scale with a expressed in Kelvin
constant ratio between the values: all
arithmetic operators allowed.

differentiate between qualitative and quantitative descriptions. An attribute is often

intuitive and defined in a generic way. A variable needs to be operationalized by
defining it formally and keeping in mind the further data processing, which mainly
depends on the scale of measurement. The four levels of measurement define the
types of mathematical and statistical analysis and the methods of combining
measurements that can be made (Table 10.5).
In common meaning, also the term parameter is used as a synonym of variable,
which is confusing in a mathematical or statistical context where variable and
parameter have clearly a different meaning.
The concept indicator comes from environmental sciences and refers to a
numerical value that helps to understand the state of the environment, ecosystems
or human health. Indicators are quantitative variables measured over time to show
the change of state of the environment at a predefined geographical scale. Often,
series of indicators are needed to address the complexity of the environment. Also,
an indicator can be constructed from a combination of standardised variables and is
then called an index. Tracking over time made indicators much wanted instruments
for policy analysis (Parris 2004) Ecological indicators are a specific subset of
environmental indicators, as are agricultural and landscape indicators (Dramstad
and Sogge 2003). Often, indicators are abstract and only partial expressions of
complex holistic phenomena. To be useful for policy, indicators must be sensitive
to detect significant changes, allow comparison between times and regions and link
different scales. These conditions are rarely met and a lot of uncertainty remains in
their interpretation.
Criteria are used in decision-making analysis, in particular in multi-criteria
assessment or evaluation. A criterion gives some measure of a quality significant
for the decision-making. Criteria describe properties and qualities in function of a
selection or classification in a process of decision-making. Typically, multiple
10.3 Landscape Assessment 297

Fig. 10.17 The difference between accuracy and precision. The true value lies in the centre at the
crossing of the lines. Dots represent measurements from a sample. The result (a) is accurate and
precise, the result (b) is not accurate but has a high precision and result (c) is not accurate and not

conflicting criteria need to be compared and trade-offs need to be made, in particular

when also multiple objectives or stakeholders are involved. Multi-criteria evaluation
became a common technique in land evaluation and land use planning using GIS
(Eastman et al. 1993). Often selected attributes and variables are transformed to
make them useful in multi-criteria assessment procedures. Examples are
reclassification and reordering values, transformation to Boolean variables, the use
of thresholds and applying weighting procedures and error assessment (Fig. 10.17).

10.3.2 Assigning Values: What Is Significant and Important?

During the assessment phase, humans assign values to landscapes or elements of the
landscape they consider important at the moment. Landscape evaluation methods
have been initially based on land evaluation (Zonneveld 1995) but expanded when
the participation of the public became important (see Chaps. 10 and 12).
When it comes to policy and planning, the expression of the values needs to meet
particular requirements. First, the values are based on criteria that are a priori
defined in policy, often as legal requirements. Criteria are variables that describe
properties and qualities, which can be used for selection, classification or evalua-
tion. Criteria or combinations of criteria can be indicative of the landscape proper-
ties that are not directly ‘measurable’, such as landscape character and other holistic
properties. Some criteria are qualitative descriptions than can be used as guidelines,
other are quantitative expressions using various levels of measurement, and some
are expressed in monetary terms. Some criteria have a more technical meaning
(e.g. class borders), some have a legal status, others describe specific landscape
Table 10.6 summarises commonly used criteria for the designation of sites and
landscapes for heritage protection. Most often the criteria are generic and values
expressed as qualities. The criteria used by UNESCO for nomination as World
Heritage are an example of purely qualitative criteria (see also Chap. 12).
298 10 Identifying, Mapping and Assessing Landscapes

Table 10.6 Criteria used in Value of the site, landscape or landscape element
the evaluation for heritage • Intrinsic values
protection – Natural
– Historical
– Cultural
– Aesthetical
– Symbolic
• Context values
– Site conditions
Scale of meaning or importance of the selected sites
• Local
• Regional
• National
• Universal
Functional meaning
• Actual use
• Potential use
• Restrictions and limitations in using the site
• Management goals for maintaining the values
• Spatial quality
Value expression
• Non-monetary, qualitative
• Monetary (e.g. value of the property)
Antrop (2004)

A second requirement is to combine related criteria in few synthetic and

quantitative indicators. This is necessary to see if the policy decisions are effective
and progress is made toward the intended goals (positive), or on the contrary
negative effects can be detected. Table 10.7 gives an example of the requirements
for policy-relevant environmental indicators formulated by the Organisation for
Economic Co-operation and Development (OECD).

10.3.3 Landscape Character Assessment

The concept landscape character clearly refers to a holistic synthesis – a Gestalt –

of all constituting components the landscape expresses in its appearance. Character
is used here in a similar sense as a person’s character. Landscape character became
popular by the emphasis by the European Landscape Convention puts on the
concept. A method for Landscape Character Assessment (LCA) developed in the
UK (Swanwick 2002) as a framework for local and regional local stakeholders and
authorities involved in a multi-scale and integrated planning, management and
conservation in the context of sustainable development with a focus upon the
10.3 Landscape Assessment 299

Table 10.7 Requirements of environmental indicators

An environmental indicator must:
1. Provide a representative image of environmental conditions, the pressure on the environment
and the social response;
2. Be simple, easy to interpret and able to show trends in time;
3. Be sensitive to changes in the environment and interrelated human activities;
4. Provide a basis for international comparison;
5. Be useable at both a national level and in issues of regional interest;
6. Be associated with a threshold or value of reference so the user can rapidly assess the
determined level.
Analytical soundness
An environmental indicator must:
1. Be well defined from a theoretical point of view and in technical terms;
2. Be based on international standards and be validated at an international level;
3. Be ready for interfacing with economic models and territorial IT systems.
The data necessary for the construction of the indicator must be:
1. Already available or obtainable at a reasonable cost/benefit;
2. Suitably documented and of a certifiable quality;
3. Revised at regular intervals in accordance with validation procedures.
After Dramstad and Sogge (2003)

countryside. Basically, LCA describes what is typical for a landscape rather than
evaluating its qualities and no judgements are made in terms of suitability or utility.
This is summarised in following four key concepts (Swanwick 2002):
• the emphasis is placed on landscape character;
• there is a clear division between the process of characterisation and the making
of judgements to inform decisions;
• both objectivity and subjectivity occur in the process;
• there is a potential for application at different scales.
Basic concepts get specific definitions which are somewhat different from the
terminology used in other systems of land(scape) classification and evaluation:
• Character: a distinct, recognisable and consistent pattern of elements in the
landscape that makes one landscape different from another, rather than better or
• Characteristics: elements, or combinations of elements, which make a particular
contribution to a distinctive character.
• Characterisation: the process of identifying areas of similar character, classify-
ing and mapping them and describing their character.
• Elements: individual components which make up the landscape, such as trees
and hedges.
300 10 Identifying, Mapping and Assessing Landscapes

• Features: particularly prominent or eye-catching elements, like tree clumps,

church towers, or wooded skylines.
The LCA distinguishes between landscape character types and landscape char-
acter areas. Landscape character types are distinct types of landscape that are
relatively homogeneous in character. They share broadly similar combinations of
geology, topography, drainage patterns, vegetation, land use and settlement pat-
terns. However, also historic, aesthetic and perceptual aspects are included as well
in the assessment. Landscape character areas are defined as single unique areas and
are the discrete geographical areas of a particular landscape type. Each area has its
own individual character and identity, even though it shares the same generic
characteristics with other areas. This distinction is reflected in the naming of
types and areas: landscape character types have generic names such as moorland
plateau and river valley, but landscape character areas take on the names of specific
places (Swanwick 2004). The character types and areas are defined at different,
hierarchical scale levels, comparable with the hierarchical approach of the holistic
classification method, but the main difference is that the scale levels are not defined
by cartographic mapping scale, but rather by administrative and decision-making
levels, ranging from local to national.
Although the LCA intended an integrated approach, another similar framework
developed in parallel. English Heritage took the initiative for a Historic Landscape
Characterisation (HLC) , with a focus on landscape characterisation for archaeo-
logical and historical aspects (Aldred and Fairclough 2002; Rippon 2012).
The work on landscape characterisation in the UK certainly inspired other
countries in their implementation of the European Landscape Convention, in
particular for inventorying their landscapes. However, most applications in land-
scape character assessment use specific methods developed at a national and even
regional level. Consequently, assessments of landscapes between contiguous polit-
ical regions rarely fit (Van Eetvelde and Antrop 2009; Mücher et al. 2003).

10.4 Landscape Monitoring

The landscape is dynamic and change is one of its fundamental characteristics.

Concerning the handling of landscape change, article 6C of the European Land-
scape Convention refers to identification and assessment of the landscapes, includ-
ing the analysis of “their characteristics and the forces and pressures transforming
them” and “to take note of changes”. Bunce et al. (2008) make a clear distinction
between surveillance and monitoring. Surveillance is the recording of features at a
specific location in one time frame, monitoring involves repeated observations
allowing to detect changes. Hence, surveillance in this sense belongs to identifying
the state of the landscape. Monitoring is one of the possibilities “to take note of
changes”. Monitoring landscapes is different from the study of their historical
development, from the analysis of landscape paths and trajectories (see Chap. 7),
10.4 Landscape Monitoring 301

although the past evolution helps to understand the processes and is significant for
assessing the actual qualities. The purpose of monitoring is different and primarily
aims to improve management decisions. Hutto and Belote (2013) used following
practical and operational definition of long-term monitoring: “repeated field-based
empirical measurements are collected continuously and then analysed for at least
10 years.”
Many monitoring systems have been elaborated and most of them have a specific
focus. A general distinction is made between passive and active monitoring.
Passive monitoring aims to register any kind of changes. This is often the case
when monitoring landscape changes. Active monitoring uses an experimental
design to detect the effects caused by specific activities or decisions. This is the
case of most ecological and land use monitoring systems.
The earliest monitoring programs related to the evolution of the biodiversity and
focus on ecological processes and human impacts, which are often considered being
negative. The list is long and no attempt will be made to give a complete overview.
More interesting is the experienced gained so far. Long-term monitoring programs
proved to have difficulties to endure. Lindenmayer and Likens (2009) list the
deficiencies of long-term ecological research and monitoring programs suffer
from so they fail or are ineffective:
• Mindless, lacking questions
• Poor experimental design
• Monitoring too many things poorly rather than fewer things well
• Failure to agree on what entities to monitor
• Flawed assumption that all monitoring programs can be the same
• Scientific disengagement from monitoring programs
• Poor data management
• Loss of integrity of the long-term data record
• Lack of funding
• Loss of key personnel
• Unexpected major event
Nichols and Williams (2006) made the distinction between targeted monitoring
and surveillance monitoring, which, however, has a contradictory meaning com-
pared to the definition of Bunce et al. (2008). Targeted monitoring comprises a
priori hypotheses about system responses, which is not the case in surveillance
monitoring. Targeted monitoring corresponds to what Hutto and Belote (2013) call
question-driven and active monitoring, while surveillance belongs to passive mon-
itoring. Here the purpose is to link the observed changes to particular activities
using indicators describing their association. Many forms of monitoring land cover
changes in agriculture and forestry attempt to do this in order to assess the
effectiveness of policy decisions and measures (Parris 2003).
Adaptive monitoring has been proposed to improve the monitoring system in a
changing context, allowing incorporating new questions, using new technology and
data while maintaining the integrity of the core measures on the long-term
302 10 Identifying, Mapping and Assessing Landscapes

(Lindenmayer and Likens 2009). Hutto and Belote (2013) saw the improvement of
monitoring programs in better formulating the basic questions and design.
The advances in landscape ecology made ecological monitoring to become
‘monitoring at the landscape-scale’. This means a multi-scale approach in species
and habitat monitoring and taking into consideration landscape characteristics as
diversity, fragmentation and heterogeneity described by landscape metrics (see
Chap. 8) (Jones 2011; O’Neill et al. 1997). Most common is using a stratified
random sampling of test sites (Bunce et al. 2005; Brandt et al. 2002; Bunce 1984).
The fast and profound changes in our environment make that inventories always
lack behind and it is not feasible the update them in real time. Hence, the necessity
that monitoring systems should be fast in collecting, analysing and reporting data.
The use of aerial photographs and remote sensing are straightforward since the
resolution became satisfactory at the field/object level.
However, ‘monitoring at the landscape-scale’ is not a landscape monitoring. The
landscape is much more than just a geographical matrix for species and habitats or
economic production in agriculture and forestry. Landscape qualities also cover
cultural and societal needs. When ecological monitoring provided solid methods in
representative sampling and statistical analysis (Howard et al. 2004), monitoring
the landscape as composed of holistic units demands more diversified approaches.
This is well illustrated by the comparison of the strategic monitoring systems used
in the Nordic countries, facing rather similar environments (Groom and Reed
2001). Also, the visual aspects of the landscape need to be included in landscape
monitoring (Fry et al. 2009).
Photo-monitoring of the landscape uses terrestrial photography from fixed
positions allowing making repeated registrations. Standardised methods have
been developed to guarantee repetitive observations (Webb et al. 2010; O’Connor
and Bond 2007; Dramstad et al. 2001). Terrain photographs also allow associating
the view with landscape preferences (Dramstad et al. 2006). Concerning sampling
the observation positions, several methods are used. Using fixed photo-points
linked to stratified random sampling sites allows more comprehensive description
of the test sites (Bunce et al. 2008; Dramstad et al. 2002). Alternatively, the fixed
positions can be chosen based on a previous landscape classification or selected
landscapes types or at frequently visited places, e.g. transects along routes and
special viewpoints with special aesthetical qualities. Finally, also the positions can
be chosen from historical photographs, such as old postcards, allowing ‘repeat
photography’ (see Sect. 10.2.1) (Heikkilä 2007; Uyttenhove 2006).


Aalen, F. H. A., Whelan, K., & Stout, M. (Eds.). (1997). Atlas of the Irish rural landscape
(352 pp). Cork: Cork University Press.
Ackermann, F. (1999). Airborne laser scanning – Present status and future expectations. ISPRS
Journal of Photogrammetry and Remote Sensing, 54, 64–67.
References 303

Aitchison, C. D., & Grant, K. (1968). Terrain evaluation for engineering. In G. A. Stewart (Ed.),
Land evaluation (pp. 125–147). Papers from the CSIRO- Symposium in co-operation with
UNESCO. Canberra: Macmillan of Australia.
Aldred O., & Fairclough, G. (2002). Historic landscape characterisation – Taking stock of the
method. The National HLC Method Review 2002, English Heritage & Somerset County
Antrop, M. (1997). The concept of traditional landscapes as a base for landscape evaluation and
planning. The example of Flanders Region. Landscape and Urban Planning, 38, 105–117.
Antrop, M. (2004). Landscape research in Europe: Editorial. Belgeo, 2–3, 199–207.
Beckett, P. H. T., & Webster, R. (1969). Military engineering experimental esthablishment
(MEXE) test report no.1123. A review of studies on terrain evaluation 1960–1969. Oxford:
Department of Agricultural Science.
Bewley, R. H., Crutchley, S. R., & Shell, C. A. (2005). New light on an ancient landscape; lidar
survey in the Stonehenge World Heritage Site. Antiquity, 79, 636–647.
Brandt, J. J. E., Bunce, R. G. H., Howard, D. C., & Petit, S. (2002). General principles of
monitoring land cover change based on two case studies in Britain and Denmark. Landscape
and Urban Planning, 62, 37–51.
Brown, G., & Reed, P. (2009). Public participation GIS: A new method for national forest
planning. Forest Science, 55, 166–182.
Bunce, R. G. H. (1984). The use of simple data in the production of strategic sampling systems.
Roskilde University Centre, IALE, Proceedings of the First International Seminar on Method-
ology in Landscape Ecological research and Planning, theme IV (pp. 45–56).
Bunce, R. G. H., Groom, G. B., Jongman, R. H. G., & Padoa-Schioppa, E. (Eds.). (2005).
Handbook for surveillance and monitoring of European habitats, Alterra-rapport 1219
(107 pp). Wageningen: Alterra.
Bunce, R. G. H., Metzger, J., Jongman, R.H.G., Brandt, J., de Blust, G., Elena-Rossello, R.,
Groom, G.B., Halada, L., Hofer, G., Howard, D.C., Kovář, P., Mücher, C.A., Padoa-Schioppa,
E., ., Paelinx, D., Palo, A., Perez-Soba, M., Ramos, I.L., Roche, P., Skånes, H., Wrbka,
T. (2008). A standardized procedure for surveillance and monitoring European habitats and
provision of spatial data, Landscape Ecology, 23, 11-25.
Burrough, P. A. (1986). Principles of geographical information systems for land resources
assessment (193pp). Oxford: Oxford University Press.
Burrough, P. A., Buitenhuis, A., & de Veer, A. A. (1982). Het informatiesysteem
Landschapsbeeld. Wageningen: Pudoc.
Butlin, R. A. (1993). Historical Geography. Through the gates of space and time (306 pp).
London: E. Arnold.
Castells, M. (2000). The rise of the network society. Oxford: Blackwell.
Challis, K. (2006). Airborne laser altimetry in alluviated landscapes. Archaeological Prospection,
13, 103–127.
Challis, K., Kokalj, Z., Kincey, M., Moscrop, D., & Howard, A. J. (2008). Airhorne lidar and
historic environment. Antiquity, 82, 1055–1064.
Christian, C. S., & Stewart, G. A. (1968). Methodology of integrated surveys. In Proceedings of the
conference on aerial surveys and integrated studies (pp. 233–280). Toulouse: UNESCO.
Claval, P. (2005). Reading the rural landscapes. Landscape and Urban Planning, 70(1–2), 9–19.
Corbett, J. (2009). Good practices in participatory mapping (p. 55). Rome: International Fund for
Agricultural Development (IFAD).
Council of Europe. (2000). European landscape convention. Firenze, 20.10.2000 http://
Craig, W. J., Harris, T. M., & Weiner, D. (2002). Community participation and geographic
information systems. In W. J. Craig, T. M. Harris, & D. Weiner (Eds.), Community participa-
tion and geographic information systems (pp. 3–16). London: Taylor and Francis.
Crutchley, S. (2009). Shedding light on the past: Using airborne LIDAR to understand ancient
landscapes. In O. Bender, N. Evelpidou, A. Krek, & A. Vassilopoulos (Eds.), Geoinformation
304 10 Identifying, Mapping and Assessing Landscapes

technologies for geocultural landscapes, European perspectives (pp. 87–106). London: CRC
Davis, J. C. (1986). Statistics and data analysis in geology. New York: John Wiley & Sons.
De Veer, A., & Burrough, P. (1978). Physiographic landscape mapping in the Netherlands.
Landscape Planning, 5, 45–62.
Delafontaine, M., Nolf, G., Van de Weghe, N., Antrop, M., et al. (2009). Assessment of sliver
polygons in geographical vector data. International Journal of Geographical Information
Science, 23, 719.
Devereux, B., Amable, G., & Crow, P. (2008). Visualization of LiDAR terrain models for
archaeological feature detection. Antiquity, 82, 470–479.
Dramstad, W., & Sogge, C. (Eds.) (2003, October 7–9). Agricultural impacts on landscapes:
Developing indicators for policy analysis. In Proceedings from the NIJOS/OECD Expert
Meeting on Agricultural Landscape Indicators in Oslo, Norway. NIJOS Report 7, 350 pp.
Dramstad, W. E., Fry, G., Fjellstad, W. J., Skar, B., Helliksen, W., Sollund, M.-L. B., Tveit, M. S.,
Geelmuyden, A. K., & Framstad, E. (2001). Integrating landscape-based values – Norwegian
monitoring of agricultural landscapes. Landscape and Urban Planning, 57(3–4), 257–268.
Dramstad, W. E., Fjellstad, W. J., Strand, G.-H., Mathiesen, H. F., Engan, G., & Stokland, J. N.
(2002). Development and implementation of the Norwegian monitoring programme for agri-
cultural landscapes. Journal of Environmental Management, 64, 49–63.
Dramstad, W. E., Tveit, M. S., Fjellstad, W. J., & Fry, G. L. A. (2006). Relationships between
visual landscape preferences and map-based indicators of landscape structure. Landscape and
Urban Planning, 78(4), 465–474.
Drosos, V., & Farmakis, D. (2006). Airborne laser scanning and DTM generation. In: E. Manolas
(Ed.), Naxos international conference on sustainable management and development of moun-
tainous and Island areas (pp. 206–218). Greece.
Eastman, J. R., Kyem, P. A. K., Toledano, J., & Jin, W. (1993). GIS and decision making. UNITAR
explorations in GIS technology (vol. 4, 112 pp). Geneva: UNITAR.
Emanuelsson, U. (2009). The rural landscapes of Europe. How man has shaped European nature
(p. 383). The Swedish Research Council Formas.
English Heritage. (2009). Aerial survey and investigation English heritage. Available at http:// (1-09-2009).
Fagerholm, N., & Käyhk€ o, N. (2009). Participatory mapping and geographical patterns of the
social landscape values of rural communities in Zanzibar. Tanzania, Fennia, 187(1), 43–60.
FAO. (1976). A framework for land evaluation (Vol. 32). Rome: FAO Soils Bulletin.
Fernandes da Silva, P. C., & Cripps J. C. (2011). Geo-environmental Terrain assessments based on
remote sensing tools: A review of applications to hazard mapping and control. In
E. Broniewicz (Ed.), Environmental management in practice. InTech, ISBN: 978-953-307-
358-3. Available from:
Forman, R., & Godron, M. (1986). Landscape ecology (p. 619). New York: Wiley.
Fry, G., Tveit, M. S., Ode, A., & Velarde, M. D. (2009). The ecology of visual landscapes:
Exploring the conceptual common ground of visual and ecological landscape indicators.
Ecological Indicators, 9(5), 933–947.
Gallagher, J., & Josephs, R. (2008). Using LiDAR to detect cultural resources in a forested
environment: An example from Isle ROYALE National Park, Michigan, USA. Archaeological
Prospection, 15, 187–206.
Gheyle, W., Dossche, R., Bourgeois, J., Stichelbaut, B., & Van Eetvelde, V. (2014). Integrating
archaeology and landscape analysis for the cultural heritage management of a world war I
militarised landscape. The German field defences in Antwerp. Landscape Research, 39(5),
Gould, P., & White, R. (1974). Mental maps. Pelican geography and environmental studies
(p. 204). New York: Penguin Books.
References 305

Gran€o, J. G. (1929). Reine Geographie. Eine methodologische Studie beleuchtet mit Beispielen
aus Finnland und Estland. Helsinki, Helsingfors, Acta Geographica, 2(2), 202 pp.
Gran€o, J. G. (1952). Régions géographiques et une méthode pour les délimiter. Compte Rendu du
XVIè Congrès International de Géographie. Lisbonne 1949, 11 pp.
Gran€o, O., & Paasi, A. (Eds.). (1997). Pure geography (J. G. Gran€ o, M. Hicks, Trans.) (19e1 pp).
Baltimore/London: The John Hopkins University Press.
Groom, G., & Reed, T. (Eds.). (2001). Strategic landscape monitoring for the Nordic countries,
TemaNord. Nordic Council of Ministers: Copenhagen.
Groom, G., Wascher, D., Potschin, M., & Haines-Young, R. (2006). Landscape character assess-
ments and fellow travellers across. In: R. G. H. Bunce & R. H. G. Jongman (Eds.), Landscape
ecology in the mediterranean: Inside and outside approaches. Proceedings of the European
IALE Conference, Faro, Portugal, 29 March–2 April 2005 (pp. 221–231). IALE Publication
Series 3.
Harmon, J. M., Leone, M. P., Prince, S. D., & Snyder, M. (2006). Lidar for archaeological
landscape analysis: A case study of two Eighteenth-century Maryland plantation sites. Amer-
ican Antiquity, 71, 649–670.
Heikkilä, T. (2007). Visual monitoring of Finnish landscapes, Photographic documentation of
changes in cultural landscapes (157 p). Helsinki: University of Art and Design.
Hidding, M. C., & Teunissen, A. T. J. (2002). Beyond fragmentation: New concepts for urban-
rural development. Landscape and Urban Planning, 58(2–4), 297–308.
Hillier, B., & Hanson, J. (1984). The social logic of space. Cambridge: Cambridge University
Howard, J., & Mitchell, C. (1980). Phyto-geomorphic classification of the landscape. Geoforum,
11(2), 85–106.
Howard, D. C., Petit, S., & Bunce, R. G. H. (2004). Monitoring multi-functional landscapes at a
national scale – Guidelines drawn up from the Countryside Survey of Great Britain. In
J. Brandt & H. Vejre (Eds.), Multifunctional landscapes. Theory, values and history (Vol.
Vol. II, pp. 3–18). Southampton: WIT Press.
Hutto, R. L., & Belote, R. T. (2013). Distinguishing four types of monitoring based on the
questions they address. Forest Ecology and Management, 289, 183–189.
Joly, D., Brossard, T., Cavailhes, J., Hilal, M., Tourneux, F. P., Tritz, C., & Wavresky, P. (2009). A
quantitative approach to the visual evaluation of landscape. Annals of the Association of
American Geographers, 99(2), 292–308.
Jones, J. P. G. (2011). Monitoring species abundance and distribution at the landscape scale.
Journal of Applied Ecology, 48, 9–13.
Jongman, R. H. G., & Bunce, R. G. H. (2000). Landscape classification, scales and biodiversity in
Europe. In U. Mander & R. H. G. Jongman (Eds.), Consequences of land use changes
(pp. 11–38). Southampton: Wit Press.
Kasai, M., Ikeda, M., Asahina, T., & Fujisawa, K. (2009). LiDAR-derived DEM evaluation of
deep-seated landslides in a steep and rocky region of Japan, LiDAR-derived DEM evaluation
of deep-seated landslides in a steep and rocky region of Japan. Geomorphology, 113, 57–69.
Kilchenmann, A. (1971). Statistich-analytische Landschaftsforschung. Geoforum, 7, 39–52.
Kilchenmann, A. (1973). Die Merkmalanayse für Nominaldaten: eine Methode zur Analyse von
Qualitativen geographischen Daten. Geoforum, 15, 33–45.
Kokalj, A., Zaksek, K., & Ostir, K. (2011). Application of sky-view factor for the visualisation of
historic landscape features in lidar-derived relief models. Antiquity, 85(2011), 263–273.
Kolen, J., Renes, H., & Hermans, R. (Eds.). (2015). Landscape biographies. Geographical,
historical and archaeological perspectives on the production and transmission of landscapes.
Amsterdam: Amsterdam University Press, Landscape and Heritage Studies.
Kwakernaak, C. (1984). Information applied in ecological land classification. In J. Brandt &
P. Agger (Eds.), Proceedings of the first international IALE seminar on methodology in
landscape ecological research and planning. theme III: Methodology of data analysis
(pp. 59–66). Roskilde: Roskilde University Centre.
306 10 Identifying, Mapping and Assessing Landscapes

Lambin, E. F., & Geist, H. J. (Eds.). (2006). Local processes and global impacts, Global change –
The IGBP series (222 pp). Berlin: Springer.
Lebeau, R. (1979). Les grands types de structures agraires dans le monde (162 pp). Paris: Masson/
troisième édition.
Lindenmayer, D. B., & Likens, G. E. (2009). Adaptive monitoring: A new paradigm for long-term
research and monitoring. Trends in Ecology & Evolution, 24(9), 482–486.
Lipsky, Z. (1995). The changing face of the Czech rural landscape. Landscape and Urban
Planning, 31, 39–45.
Liu, X. (2008). Airborne LiDAR for DEM generation: Some critical issues. Progress in Physical
Geography, 32, 31–49.
Llobera, M. (2003). Extending GIS-based visual analysis: The concept of visualscapes.
InterntionalInternational Journal of Geographical Information Science, 17(1), 25–48.
Lobet, J., Nederlandt, N., & Rosillon, F. (2006). Mise en place d’un observatoire du paysage dans
le bassin transfrontalier Semois-Semoy (67 pp). Arlon: Ulg – Contrat de Rivière Semois.
Lohr, U. (1998). Digital elevation models by laser scanning. The Photogrammetric Record, 16,
Loupa Ramos, O., Bianchi, P., Bernardo, F., & Van Eetvelde, V. (2017). What does really matter
to people? Exploring contents of landscape identity at local scale. Landscape Research, in
Luginbühl, Y., Bontron, J.-C., & Cros, Z. (1994). Méthode pour des atlas de paysages.
Identfication et qualification. Paris: Strates/CNRS-SEGESA, Ministère de l’Aménagement
du Territoire, de l’Equipement et des Transports, Direction de l’Architecture et de
Lynch, K. (1960). The image of the city (194 pp). Cambridge, MA/London: The MIT Press.
Mabbutt, J. (1968). Review of concepts of land classification. In A. Stewart (Ed.), Land evaluation
(pp. 11–28). Melbourne: Macmillan of Australia.
MEEDDAT. (2008). Méthode de l’Observatoire photographique du paysage. Paris: Ministère de
l’Écologie, de l’Énergie, du Développement durable et de l’Aménagement du territoire. http://
Mitchell, C. (1971). An appraisal of a hierarchy of desert land units. Geoforum, 7, 69–79.
Mitchell, C. (1973). Land evaluation. London: Longman.
Mitchell, C. W. (1991). Terrain evaluation (2nd ed.497p). Essex: Longman.
Mücher, C. A., Bunce, R. G. H., Jongman, R. H. G., Klijn, J. A., Koomen, A. J. M., Metzger, M. J.,
& Wascher, D. M. (2003). Identification and characterisation of environments and landscapes
in Europe, Alterra-rapport 832 (119 pp). Wageningen: Alterra.
Mutlu, M., Popescu, S. C., & Zhao, K. (2008). Sensitivity analysis of fire behavior modeling with
LIDAR-derived surface fuel maps. Forest Ecology and Management, 256, 289–294.
Nakano, T. (1962). Landform type analysis on aerial photographs, its principle and technique.
Archives Internationales de Photogrammetrie, 14, 149–152.
Naveh, Z., & Lieberman, A. (1994). Landscape ecology: Theory and application (360 pp).
New York: Springer Verlag.
Neef, E. (1967). Die theoretische Grundlagen der Landschaftslehre. Haack Gotha: V.E.B.
Nichols, J. D., & Williams, B. K. (2006). Monitoring for conservation. Trends in Ecology &
Evolution, 21, 668.
Nijhuis, S., van Lammeren, R., & van der Hoeven, F. (2011). Exploring the visual landscape,
Resarch in Urbanism Series, vol. 2 (p. 334). Amsterdam: IOS Press.
Nogué, J., & Sala, P. (2006). Prototype Landscape Catalogue. Olot/Barcelona: Observatori del
Paisatge, 31 p.
Nogué, J. Sala, P. & Grau, J. (2016). The Landscape Catalogues of Catalonia. Methodology. Olot:
Landscape Observatory of Catalonia, pp 142.
Nohl, W. (2001). Sustainable landscape use and aesthetic perception – Preliminary reflections on
future landscape aesthetics. Landscape and Urban Planning, 54, 223–237.
References 307

O’Connor, P. J., & Bond, A. (2007). Maximising the effectiveness of photopoint monitoring for
ecological management and restoration. Ecological Management and Restoration, 8(3),
Observatoire Photographique Transfrontalier des Paysages. (2008). http://www.observatoire-¼index
Ode, Å., Tveit, M. S., & Fry, G. (2008). Capturing landscape visual character using indicators:
Touching base with landscape aesthetic theory. Landscape Research, 33(1), 89–117.
O’Neill, R. V., Hunsaker, C. T., Jones, K. B., Riitters, K. H., Wickham, J. D., Schwartz, P. M.,
Goodman, I. A., Jackson, B. L., & Baillargeon, W. S. (1997). Monitoring environmental
quality at the landscape level, using landscape indicators to assess biotic diversity, watershed
integrity, and landscape stability. Bioscience, 47(8), 513–519.
Palang, H., Printsmann, A., Gyuró, E. K., Urbanc, M., Skowronek, E., & Woloszyn, W. (2006).
The forgotten rural landscapes of Central and Eastern Europe. Landscape Ecology, 21(3),
Palmer, J. F., Lankhorst, J. R. (1998). Evaluating visible spatial diversity in the landscape.
Landscape and Urban Planning, 43(1–3), 65–78.
Palmieri, A., Martineo, L., Dominici, P., & Kasando, M. (2011). Land cover and land use diversity
indicators in LUCAS 2009 data.
Parris, K. (2003). Agricultural landscape indicators in the context of the OECD work on agri-
environmental indicators. In W. Dramstad & C. Sogge (Eds.), Agricultural impacts on land-
scapes, developing indicators for policy analysis. Proceedings from the NIJOS/OECD Expert
Meeting on Agricultural Landscape Indicators in Oslo, Norway, 7–9 October 2002, 10.18.
Parris, K. (2004). Measuring changes in agricultural landscapes as a tool for policy makers. In
J. Brandt & H. Vejre (Eds.), Multifunctional landscapes. Theory, values and history (Vol.
Vol. I, pp. 193–218). Southampton: WIT Press.
Pedroli, B. (1983). Landscape concept and landscape and rangeland surveys in the Soviet Union.
ITC Journal, 4, 307–321.
Pérez-Soba, M., & Wascher, D. (2005). Landscape charater areas. Places for building a sustain-
able Europe. Policy Brochure as deliverable from the EU’s Accompanyning Measure project
European Landscape Character Assessment Initiative (ELCAI), funded under the 5th Frame-
work Programme on Energy, Environmont and Sustainable Development (4.2.2).
Plets, G. (2013). Heritages in the Making – Social embodiment of cultural heritage objects and
places in the multicultural Altai Republic (p. 360). Ghent: Ghent University/Faculty of Arts
and Philosophy.
Powlesland, D., Lyall, J., Hopkinson, G., Donoghue, D., Beck, M., Harte, A., & Stott, D. (2006).
Beneath the sand – Remote sensing, archaeology, aggregates and sustainability: A case study
from Heslerton, the Vale of Pickering, North Yorkshire, UK. Archaeological Prospection, 13,
Puschmann, O., & Dramstad, W. (2003). Documenting landscape change through fixed angle
photography, In W. Dramstad & C. Sogge (Eds.), Agricultural impacts on landscapes, Devel-
oping indicators for policy analysis. Proceedings from the NIJOS/OECD Expert Meeting on
Agricultural Landscape Indicators in Oslo, Norway, 7–9 October 2002, pp. 258–268.
Rackham, O. (1986). The history of the countryside. London: Penguin Books.
Rackham, O. (2000). Prospects for landscape history and historical ecology. Landscapes, 2, 3–15.
Ramirez, J. R. (2006). A new approach to relief representation. Surveying and Land Information
Science, 66, 19–25.
Rippon, S. (2012). Historic landscape analysis. Deciphering the countryside, Practical handbooks
in archaeology n 16 (166 pp). York: CBA Publications.
Rossiter, D. G. (1995). Economic land evaluation: Why and how. Soil Use and Management, 11,
Ruggles, C. L. N., Medyckyj-Scott, D. J., & Gruffydd, A. (1993). Multiple viewsheds analysis
using GIS and its archaeological application: A case study in northern Mull. In J. Adresen,
308 10 Identifying, Mapping and Assessing Landscapes

T. Madsen, & I. Scollar (Eds.), Computing the past: Computer applications and quantitative
methods in archaeology (pp. 125–132). Aarhus: Aarhus University Press.
Sauer, C. O. (1925). The morphology of landscape. University of California Publications in
Geography, 2; reprinted in Leighly, J. (Ed.). (1974). Land and life (pp. 315–350). Berkely:
University of California Press.
Saye, S. E., Wal, D. V. D., Pye, K., & Blott, S. J. (2005). Beach–dune morphological relationships
and erosion/accretion: An investigation at five sites in England and Wales using LIDAR data.
Geomorphology, 72, 128–155.
Sevenant, M., & Antrop, M. (2007). Settlement models, land use and visibility in rural landscapes:
Two case studies in Greece. Landscape and Urban Planning, 80, 362–374.
Soini, K. (2001). Exploring human dimensions of multifunctional landscapes through mapping
and map-making. Landscape and Urban Planning, 57(3–4), 225–239.
Stewart, G. A. (Ed.). (1968). Land evaluation. Papers from the CSIRO- symposium in
co-operation with UNESCO. Canberra: Macmillan of Australia.
Swanwick, C. (2002). Landscape character assessment. Guidance for England and Scotland.
Wetherby: The Countryside Agency/Scottisch Natural Heritage.
Swanwick, C. (2004). The assessment of countryside and landscape character in England: An
overview. In: K. Bishop & A. Phillips (Eds.), Countryside planning. New approaches to
management and conservation (pp. 109–124). London: Earthscan.
Tomlin, C. (1990). Geographic Information Systems and Cartographic Modelling (249 pp).
Prentice-hall: Englewood Cliffs.
Troll, C. (1939). Luftbildforschung und Landeskundige Forschung. Erdkundliches Wissen.
Schriftenreihe für Forschung und Praxis, Heft 12. Wiesbaden: F.Steiner Verlag.
Troll, C. (1963). Landschafts€ okologie als geographisch-synoptische Naturbetrachtung. In

Okologische Landschaftsforschung und vergleichende Hochgebirgsforschung. Erdkundliches
Wissen, Schriftenreihe f€ ur Forschung und Praxis, Heft 11. Wiesbaden: F.Steiner Verlag.
Tuan, Y.-F. (1977). Space and place: The perspective of experience (496 pp). Minneapolis:
University of Minnesota Press.
Tveit, M. S. (2009). Indicators of visual scale as predictors of landscape preference; a comparison
between groups. Journal of Environmental Management, 90, 2882–2888.
Tveit, M. S., Ode, Å., & And Fry, G. (2006). Key concepts in a framework for analysing visual
landscape character. Landscape Research, 31(3), 229–225.
UNISCAPE. (2010). Living landscape. The European landscape convention in research perspec-
tive, 2 vols. Firenze: UNISCAPE.
Utkin, A. B., Lavrov, A. V., Costa, L., Simoes, F., & Vilar, R. (2002). Detection of small forest
fires by LIDAR. Applied Physics, 74, 77–83.
Uyttenhove, P. (Ed.). (2006). Recollecting Landscapes. Herfotografie, geheugen en transformatie
1904–1980-2004 (p. 383). Gent: A&S/Books.
Van Eetvelde, V., & Antrop, M. (2009). A stepwise multi-scaled landscape typology and charac-
terisation for trans-regional integration, applied on the federal state of Belgium. Landscape and
Urban Planning, 91(3), 160–170.
Van Hecke, E., Antrop, M., Schmitz, S., Sevenant, M., & Van Eetvelde, V. (2010). Atlas van
België: 2 Landschappen, platteland en landbouw (74 pp). Gent: Academia Press.
Vink, A. P. A. (1980). Landschapsecologie en landgebruik (151pp). Utrecht: Bohn, Scheltema &
Vinogradov, B. V. (1966). Les nouvelles méthodes d’interprétation des photographies aériennes
pour la cartographie de la couverture végétale. Technip, Paris, 2, 93–96.
Vinogradov, B. V. (1976). Hierarchy of topological units of vegetation cover. Biogeog. and Soil
Geog. (Moscow), 4, 133–137.
Vinogradov, B. V., Gerenschule, A. G., Isachenko, K. G., Raman, K. G., & Tseselchuk, Y. N. (1962).
Basic principles of landscape mapping. Soviet Geography: Review and Translatyion, 3(6), 15–21.
Vos, W., & Meekes, H. (1999). Trends in European cultural landscape development: Perspectives
for a sustainable future. Landscape and Urban Planning, 46(1–3), 3–14.
References 309

Warnock, S., & Griffiths, G. (2015). Landscape characterisation: The living landscape approach in
the UK. Landscape Research, 40(3), 261–278.
Webb, R. H., Boyer, D. E., & Turner, R. M. (2010). Repeat photography methods and applications
in the natural sciences. Island Press, p. 392. E-BOOK FORMAT, September 2012, $74.99,
ISBN: 9781610910064.
Webster, R., & Beckett, P. (1970). Terrain classification and evaluation using air photography: A
review of recent work at Oxford. Photogrammetria, 26, 51–75.
Wehr, U., & Lohr, U. (1999). Airborne laser scanning – An introduction and overview. ISPRS
Journal of Photogrammetry and Remote Sensing, 54, 68–82.
Werbrouck, I., Antrop, M., Van Eetvelde, V., Stal, C., De Maeyer, P., Bats, M., Bourgeois, J.,
Court-Picon, M., Crombe, P., De Reu, J., De Smedt, P., Finke, P. A., Van Meirvenne, M.,
Verniers, J., & Zwertvaegher, A. (2011). Digital elevation model generation for historical
landscape analysis based on LiDAR data, a case study in Flanders (Belgium). Expert Systems
with Applications, 38, 8178–8185.
Zakšek, K., Ostir, K., & Kokalj, Z. (2011). Sky-View factor as a relief visualisation technique.
Remote Sensing, 3, 398–415.
Zhou, G., & Xie, M. (2009). Coastal 3-D morphological change analysis using LiDAR series data:
A case study of assateague Island national seashore. Journal of Coastal Research, 25, 435–447.
Zonneveld, I. S. (1972). Land evaluation and Land(scape) science (Vol. VII-4). Enschede:
Zonneveld, I. S. (1995). Land ecology (199 pp). Amsterdam: SPB Academic Publishing bv.
Zonneveld, I. S. (2005). The land unit as a black box: A Pandora’s box? In: J. A. Wiens & M. R.
Moss (Eds.), (1999). Issues in landscape ecology. Studies in landscape ecology (pp. 331–345).
Cambridge: Cambridge University Press.
Chapter 11
The Artist’s Landscape

Abstract The landscape is a source of inspiration in literature, visual arts and

garden design and by extension of the creation of new landscapes. Visiting
unknown landscapes brought new ‘incitements’ and new technologies brought
new ways of seeing. Landscape art changed the concept of landscape. This is
illustrated with the history of landscape painting and landscape architecture.
While in China, landscape painting was from the ancient beginnings regarded as
an important art form, landscape painting in Europe became only a genre on its own
in the Renaissance with a popularity and style that varied from region to region.
Landscape paintings are imagined constructions reflecting philosophy, culture and
social ideology of a period and thus, express the relationship humans had with the
environment. Landscape paintings illustrate also the evolution in the craftsmanship
of depicting the three-dimensional world on a two-dimensional plane not only for
artistic but also for documentary reasons. The development of all kinds of pictorials
and perspectives laid a basis for mapping, design and planning as well. Landscape
painting evolved simultaneously with garden design, which up-scaled in landscape
architecture. Gardens and parks are creations of new landscapes expressing spiritual
visions and ideology as well as craftsmanship characteristic for a time. Also, the
evolution differs between the East and the West and a succession of styles reflects
the cultural evolution of society. Case studies illustrate the development of different
styles and models. Designed landscapes are nowadays recognised as a distinct
category of heritage.

Keywords Landscape painting • Landscaping • Landscape architecture • Design •


11.1 Landscape as a Source of Artistic Inspiration

More or less realistic descriptions and representations of the landscape began with
the great explorations, involving describing and mapping new lands and cultures.
Geographers and naturalists illustrated their vivid descriptions with drawings and
paintings. New perspectives changed landscape preferences and values and created
new ways of interacting and shaping the environment. Landscape prose and poems
illustrated philosophical ideas and landscape descriptions became popular as travel

© Springer Science+Business Media B.V. 2017 311

M. Antrop, V. Van Eetvelde, Landscape Perspectives, Landscape Series 23,
DOI 10.1007/978-94-024-1183-6_11
312 11 The Artist’s Landscape

guides. For gardeners and landscape architects, the landscape was the prima
materia of their art.
Alexander von Humboldt aimed to describe nature ‘as a whole’ in the ‘opus of
my life’, Cosmos, making an ultimate synthesis between scientific explorations of
the ‘external world’ and of the impressions it ‘produces on the feelings’ of the inner
world (von Humboldt 1858). The title ‘Cosmos’, he explained, came from the
Greek ‘kosmos’ (k oσμoς), meaning ‘beauty’ and ‘order’ and referred also to the
‘Order of the Universe’ (Wulf 2015; von Humboldt 1849). The second volume of
Cosmos is devoted to the “incitements to the study of nature”, which he formulated
as follows: “we now pass from the domain of objects to that of sensations” and he
defined three ‘kinds of incitement’: “(1) the aesthetic treatment of natural scenery
by vivid and graphical descriptions of the vegetable and animal world, which is a
very modern branch of literature; (2) landscape painting, so far as it portrays the
characteristic aspect of vegetation; and (3) the more extended cultivation of tropical
plants and the assemblage of contrasted exotic forms.” (von Humboldt 1849).
All three ‘incitements’ result in artistic representations. The eye, von Humboldt
wrote, was the “Organ der Weltanschauung”, “the medium through which we may
contemplate the universe” (Wulf 2015), and “the comprehension of a natural whole,
the feeling of the unity and harmony of the Cosmos, will become at once more vivid
and more generally diffused, with the multiplication of all modes of bringing the
phenomena of nature generally before the contemplation of the eye and of the
mind” (von Humboldt 1849).
In Landscape and the European sense of sight – eyeing nature, Denis Cosgrove
(2002) demonstrates how the connections “between the material and the imagina-
tive in landscape have evolved historically in close connection with changing
technologies of seeing and representing space.” The usage of the word ‘landscape’
refers to the tangible and measurable ensemble of material elements in a given
geographical area, as well as to all forms of representing these in a wide variety of
media, such as text, paintings, photographs, maps and computer visualisations. The
concept landscape relates also to memories, to the imagination of the experience of
the outside world with all senses.
The landscape became indeed a source of inspiration in literature and painting,
and by extension in garden design. As such it has been studied in cultural geography
(Daniels 2008). One of the definitions of the word landscape in the dictionary is a
genre in painting, drawing and photography, depicting natural or rural scenery. It
refers nowadays to a format of printing as well. The verb landscaping means to
improve the appearance of an area of land and infrastructure, as by planting trees,
shrubs, grass, or altering the terrain surface.
11.2 Landscape Painting – The Imagined Landscape 313

11.2 Landscape Painting – The Imagined Landscape

11.2.1 Landscapes That Are Not the Landscape

Landscapes in graphical arts vary from entirely imaginary to realistic views and the
subjects vary from details of plants and rocks to vast panoramas. They seem to have
a naturalistic look but are in fact a patchwork of smaller scenes that have a more
coherent realism. Depicted elements are not shown in realistic ecological context
but used symbolically. In early landscape paintings, landscapes are background
scenes for telling narratives related to religion, mythology, legendary heroes or
historical events. It took some time before the landscape became the core subject
and landscape painting became a genre on its own. Clearly, the painted landscape is
not the landscape (Olwig 2004) (Fig. 11.1).

11.2.2 The Challenge of Perspective

Many landscape paintings use a fictive, elevated viewpoint allowing to create some
depth and overview. For a coherent depiction of the scenery, some system of
perspective is needed. All objects in the view are scaled and distorted relative to
the position and the distance to the viewer. The general experience that similar
sized objects are seen smaller when further away has always been the simplest way
to create the illusion of depth by diminishing the size of the objects with the

Fig. 11.1 René Magritte,

La Condition Humaine
(1933, Oil on canvas
100  81 cm, the National
Gallery of Art, Washington,
314 11 The Artist’s Landscape

The common experience that distant objects look fuzzy, pale and fade with a
blue-greyish colour also creates a sense of depth. This is called the atmospheric
perspective (sometimes confusingly called aerial perspective). Atmospheric atten-
uation increases with the distance and shift colours to shorter wavelengths (blue)
and decreases saturation and contrast (see also Chap. 6).
Projecting three-dimensional objects on a two-dimensional plane causes shape
distortions, in particular, foreshortening. Also, parallel lines not perpendicular to
the picture plane appear to converge in vanishing points. This is particularly
noticeable in scenes with buildings. However, most landscape scenes are not
geometric and simple linear perspectives are seldom adequate.
To represent objects as volumes in the scenery in a convincing and coherent
manner, several geometrical projection methods have been developed. Figure 11.2
shows a classification of planar graphical projections with the shaded areas indi-
cating the kind of projections and perspectives used in landscape representations.
The main subdivision is between parallel oblique projections, called pictorials, and
linear perspectives using vanishing point on the horizon.
In landscape painting, some special perspective techniques have been developed
to construct map-like pictures and panoramas. They use an imaginary, elevated
viewpoint (a bird’s eye perspective), a wide field of view and graphical projection
methods to create the illusion of depth in a formal, conventional and systematic
manner, but different from our visual experience of perspective. They are com-
monly called pictorials.

Fig. 11.2 Typology of planar graphical projections. The ones commonly used in landscape
painting are shaded grey
11.2 Landscape Painting – The Imagined Landscape 315

Other reasons can override these graphical conventions. A common example is

to represent more important elements, i.e. figures, taller regardless their distance to
the viewer. They allow constructing landscape views we can never experience in
reality. Nowadays, with cameras and computers, all kinds of visualisations can be
To read the landscapes represented on paintings demands some understanding of
perspectives and in particular the construction of pictorials. Many of the techniques
involved are older than the introduction of the linear perspective during the
Renaissance and were applied worldwide.
To represent rectilinear objects such as houses, some kind of oblique projection
was used. Two types are commonly found on old landscape paintings: the axono-
metric and oblique projection. Both use a parallel projection of the view lines
looking at the tilted and rotated object, resulting in a pictorial or plan oblique. These
produce convincing three-dimensional representations of the objects and some
types allow taking measurements as they have a constant scale in different dimen-
sions. However, the view is distorted and has no ‘natural’ perspective.
The axonometric projection is commonly used in architectural and technical
drawing and several variants exist. In the isometric projection, the direction of
viewing is such that the three axes of space appear at an angle of 120 between
them. The object is represented looking at the nearest corner. Consequently, the
scale and length of all sides are preserved and there is no foreshortening, which
causes a distortion of the shape. In the diametric projection, only two axes have the
same scale, while all three axes are different in the trimetric projection.
The group of oblique projections also consists of different types, which are
commonly used. The cabinet projection or perspective is a variant of the isometric
perspective, showing one face of the object parallel to the viewing plane. The third
axis is projected as going off in an angle with the viewing plane (typically 30 or
45 ) and a foreshortening of the receding planes of one-half is applied.
The cavalier projection (also called cavalier perspective or high viewpoint)
shows the receding lines at an angle of 45 with the horizontal base line at the
same scale. It was popular among French military engineers in the eighteenth
century to depict fortifications. It is, therefore, also known as military projection.
These projections are used for representing individual objects, not a three-
dimensional view of the landscape as a whole or its spatial patterns such as fields.
However, they are useful for creating wide panoramas and horizontal scroll paint-
ings, as the offer a constant perspective when panning the view along the scenery.
Linear perspectives depict parallel lines that are oblique to the viewing plane as
converging lines to one or more vanishing points. The number of vanishing points
chosen, their position and the distance of the object to the viewer control the
distortions. All kinds of perspectives can be constructed and not all correspond to
a familiar ‘natural’ perspective. Roman murals show the use of central linear
perspectives, i.e. using one vanishing point, in depicting architectural views, in
particular looking out of windows and porches. To correct the unnatural distortions,
multiple horizons were used, creating a composite ‘eyeball perspective’ (Schodde
316 11 The Artist’s Landscape

The theory of perspective developed in the fourteenth century Italian Renais-

sance and became popular in graphical arts ever since. Today, cameras, scanners
and computer visualisation make all kinds of curvilinear perspectives possible
as well.

11.2.3 Setting the Scene: A Brief Overview of Important

Historical Steps

This section focuses on the important historical steps in representing landscape as a

core subject in graphical arts (Fig. 11.3). The focus is on the changing ‘ways of
seeing’ in constructing landscape as an art genre.
The two main traditions for landscape art are the western and Chinese tradition
and both go back well over a 1000 years. Major differences between landscape
painting in the West and East are its appreciation and acceptance of the genre. The
classic Chinese mountain-water ink painting (shanshui style) was considered – as
was calligraphy – the most prestigious form of visual art. In Europe, it took a long
time before the landscape genre became accepted as serious art form. Nevertheless,
the Flemish and Dutch landscape painting became a very successful genre of
its own.
The meaning of the landscape in the painting is essentially different between the
West and the East. In East Asia, the focus was on spiritual qualities of nature and
landscape as a moral inspiration for humans. In Europe, the focus was on history
painting with mythological or religious subjects, demanding a spatial setting, which
gradually became a landscape background that also got additional symbolic mean-
ing. Landscape became only the core subject of the painting with the Flemish and
Dutch landscape paintings, in particular since the sixteenth–seventeenth century.
An important factor was the spreading of Protestantism. Dutch Calvinism forbade
religious painting, hence a variety of new genres appeared: portraits, landscapes,
townscapes, maritime paintings, scenes of peasantry, animals, etc. Landscape
paintings became a definite separate genre that was successfully commercialized. Ancient Origins

The earliest representations depict landscape elements such as a mountain silhou-

ette and sketches of plants and animals in the wild, and can hardly be seen as
landscape, but are rather ideographic elements setting a scene of action, often
hunting (Vos 2000; Jellicoe and Jellicoe 1975). Landscape as a whole is imaginary,
but contains realistic features painted in detail, such as accurate representation of
fauna and flora. Most often landscapes are compositions of separate more or less
realistic elements, setting the scene for a narrative, to illustrate an impression or
symbolizing spiritual ideas.
11.2 Landscape Painting – The Imagined Landscape 317

Fig. 11.3 Timeline landscape painting in the west and east

318 11 The Artist’s Landscape

The oldest painted landscape features preserved are decorative wall paintings in
Egyptian tombs and in palaces and villas. Typical are the Nilotic landscape scenes
(Pre-dynastic period from c.3500–3100 BCE). These scenes do not depict the
landscape as a whole in a coherent was, but have often a strong sense of place,
using the river as a connecting element or using an enclosed wall. The emphasis is
on individual plant forms and human and animal figures which are represented in
detail. Hunting scenes are also common.
This style spread to the Minoan Agaean (c.2750 BCE) and presumably further to
the ancient Greece, the Hellenic world and to Rome. The only tangible examples
are mural frescos in Minoan sites such as Akrotiri (Thera) and Cnossos (Crete).
Only in Greek texts there is evidence that wall and panel paintings were
common. Agatharchus of Athensas discussed techniques of perspective in the
context of creating illusion of depth, referred to as skenographia in theatrical
scenery. Plato’s Critias (107b–108b) gives some indication about the difficulties
in painting landscapes:
“... if we look at the portraiture of divine and of human bodies as executed by painters, in
respect of the ease or difficulty with which they succeed in imitating their subjects in the
opinion of onlookers, we shall notice in the first place that as regards the earth and
mountains and rivers and woods and the whole of heaven, with the things that exist and
move therein, we are content if a man is able to represent them with even a small degree of
likeness ...” (Plato, Critias, 107b–108b)

From literary evidence, mural paintings existed in the Hellenistic period, but no
examples survive (Honour and Fleming 1984). From the Roman period, examples
have been preserved from the first century BCE, in particular in murals at the
archaeological sites of Pompeii and Herculaneum. These frescos show that Roman
painting was innovative in representing landscapes, in particular in using tech-
niques of perspective, which are, however, different from the formal linear per-
spectives that were developed later in the Renaissance. The landscapes depicted
were fictive landscapes, as a setting of narratives from mythology, such as from the
Odyssey, as well as scenes with abundant vegetation, in particular gardens, and
architectural vistas with urban buildings. This made von Humboldt say that, in
classical antiquity, “landscape painting long served merely as a background to
historical composition”, and he made the reflection “that which the Greeks and
the Romans regarded as attractive in a landscape, seems to have been almost
exclusively the agreeably habitable, and not what we call the wild and romantic.”
(von Humboldt 1849).
Although there is evidence for an ancient Persian tradition of mural painting,
only miniatures were preserved and tell something about the subjects represented.
The miniature painting was influenced by western tradition and after the Mongol
conquest also by Chinese traditions. It became a significant Persian genre in the
thirteenth century with a climax in the fifteenth and sixteenth centuries. It devel-
oped proper conventions in the representation of different subjects. Placing more
distant figures higher in the picture plane created depth. Important figures were
represented larger. The background was important, whether landscape or buildings,
and depicted with great detail. The fictive viewpoint was always somewhat
11.2 Landscape Painting – The Imagined Landscape 319

elevated. Landscapes are often mountainous, represented by a high undulating

skyline and outcrops rock. The influence of the Chinese art is obvious. Late Antiquity – Dark Ages – Han to Tang Dynasty

In Europe, the Late Antiquity refers to the continuity of Roman elements in the
Dark Ages from the third to the sixth century. The Early Middle Age refers to
medieval characteristics from the sixth to tenth century. This period corresponds to
the late Han Dynasty to the Tang Dynasty in China.
In the West, the Roman Empire suffered from multiple, serious crises during
the third century leading to its collapse near the end of the fourth century. This
period is referred to as Late Antiquity and characterized by population decline
and massive migrations and marks the Dark Ages of the beginnings of the
medieval period. In was a period of cultural decline from which slowly a new
culture should emerge. Rewilding and land abandonment gradually effaced the
older landscapes.
In China, the imperial order was founded during the Han Dynasty and Confu-
cianism were established. The Tang Dynasty was a period a cosmopolitan golden
age with territorial expansion. Buddhism became influential. The oldest preserved
Chinese landscape art are mainly ink paintings on silk (later paper) mounted on
scrolls and using the same techniques as calligraphy. They date from the Tang
Dynasty (518–906 CE) and by the late Tang dynasty (608–901 CE), landscape
painting became an independent genre. With these shanshui paintings, the purpose
was to grasp an emotion or atmosphere, to catch the ‘rhythm’ of nature, rather than
to reproduce the landscape realistically (Moffat 2008).
In early European medieval art, the interest in landscape disappeared almost
entirely and this would last until the Renaissance. High Middle Ages and Tang to Song Dynasties

This period covers the eleventh to thirteenth centuries and corresponds to the High
Middle Ages in Europe and the Tang and Song Dynasties and the early Yuan
dynasty of the Mongols in China. It is the period of the ‘Great age of Chinese
landscape painting’. The shanshui-style of ink paintings, using similar techniques
as calligraphy, was regarded as the highest art form. In this style, landscapes were
not reproduced realistically, but the purpose was to grasp an atmosphere and to
catch the rhythm of nature. Significant is that the word for landscape in Chinese is
made up of the two words for mountain (shan) and water (shui). In Chinese
mythology, the mountains form the connection between heaven (yin) and earth
(yang) and are the place inhabited by the Chinese immortals (Wong 2009). That is
why these two elements feature in most landscape paintings. Spirituality is prom-
inent in art and relies on Confucianism, Taoism and Buddhism, which developed
during that time. Landscape symbolizes the system of nature, which became a
320 11 The Artist’s Landscape

metaphor for the well-regulated state. The Song Dynasty began with a northern
tradition of vertical scroll paintings with towing mountains and a misty atmosphere.
From the twelfth century emerges a new tradition in the south with long horizontal
scroll paintings with panoramic views depicting more realistic scenes. One of the
most prominent examples is the scroll Along the River During the Qingming
Festival by Zhang Zeduan (Fig. 11.7).
In Europe, the High Middle Ages were characterized by a rapid population
growth during the tenth to thirteenth century, requiring vast land clearings and
reclamation of the wilderness to expand the agricultural and grazing land. This is
when the basic structure of many traditional agrarian landscapes was laid out. The
impact on the landscape is only known indirectly from historical texts, land
registers, such as the Domesday Book in England, archaeological evidence and
pollen analysis. The Rise of the ‘Pure’ Landscape Paintings (Fourteenth

to Seventeenth Century)

The 14th to the sixteenth century cover most of the Renaissance in Europe and
correspond to the late Yuan Dynasty and Ming Dynasty in China. During the
fourteenth century, painters in Europe, such as Giotto, introduced landscape
features as background to set the scene for the narrative of the painting.
Landscape painting developed in two different styles and regions: one in the
north (Low Countries of Flanders, The Netherlands and in Germany) and one in
Italy. While painters from the north visited the Mediterranean countries looking
for inspiration, their Italian colleagues were less enthusiastic of the ‘pure’
landscape style that emerged in the north. An example of the mentality about
landscape painting is Michelangelo’s commentary about Flemish painting
recorded by Francisco de Hollanda in his treatise Da pinturea antigua
(de Hollanda 1548):
In Flanders they paint to deceive the external view, such as things that may cheer you and of
which you cannot speak ill, as for example saints and prophets. They paint drapery and
masonry, the green fields, the shadow of trees, and rivers and bridges, which they call
landscapes, with many (human) figures here and there. And all this, even though it may look
good to some eyes, the truth is that it is made without any reason or art, without symmetry
nor proportion, without careful choice or boldness/audacity. Ultimately without any sub-
stance or nerve; anyhow elsewhere paintings are worse than in Flanders. I am not speaking
ill of the Flemish painting, because not all of it is bad. But because it aims so high at making
everything well (each of them very demanding by itself) that (at the end) none is done well.
Only work done in Italy can truly be called painting. . . . (Francesco de Hollanda 1548,
Dialogue with Michelangelo, translated by I. Loupa Ramos).

The sense of realism and naturalism of the Flemish painters is renowned. As a

meticulous observer of nature, von Humboldt wrote “it is therefore a fact of no little
importance [. . .] that the celebrated brothers, Hubert and John van Eyck, belonged
essentially to a school of miniature painters, which, since the second half of the
11.2 Landscape Painting – The Imagined Landscape 321

fourteenth century, had reached a high degree of perfection in Flanders. It is in the

historical paintings of the brothers Van Eyck that we first meet with a careful
elaboration of the landscape portion of the picture.” (von Humboldt 1849). He is
referring to the Ghent Altarpiece, the Adoration of the Mystic Lamb, completed in
1432 and now exposed in the cathedral in Ghent. During the current restoration, a
detailed study was made of the flora depicted on the painting and 75 species could
be identified. Although the naturalism of the depicted elements is obvious, there is
no unity in space and time in the painting as a whole. Spring flowers are depicted
alongside summer plants, Mediterranean flowers are combined with endemic plants
from northern Europe. The choice and even representation of the plants is full of
symbolism (Van Crombrugghe and Van den Bremt 2016). The landscape is typi-
cally fictive and a composition of naturalistic elements.
Early in the fifteenth century, landscape painting became a genre of its own in
Europe and near the end of the century, pure landscape drawings, prints and
paintings became successful commercial products. Several styles developed.

The Re-invention of Linear Perspective

In 1305, the fresco painter Giotto di Bondone introduced linear perspective, but in
an imperfect and not very convincing way. Only in 1413, Filippo Brunelleschi
made a convincing experiment in Florence establishing linear perspective as a solid
technique. Soon many artists started experimenting with all kinds of perspective
constructions in their work. Leon Battista Alberti documented these in his book
Della Pittura that he published in 1435.

Miniatures and Window Views

In the fifteenth century, miniature landscape features appear in manuscript illu-

mination and as background scenery seen through a window. Some have a
remarkable detail of landscape elements, most often related to practices in
agriculture and forestry. Most famous is the French Gothic manuscript of the
Très Riches Heures du Duc de Berry, a book of hours painted by the three
‘Limbourg brothers’ between c.1412 and 1416 for their patron the Duke of
Berry (Manion 1995). Most miniatures glorify the possessions of the prince,
showing his presumptuous castles and their surroundings. As their denomination
suggests the painters were native from the Low Countries and the detailed
calendar miniatures give a comprehensive overview of the current land utilisation
practices throughout the year (see Fig. 11.4).
322 11 The Artist’s Landscape

Fig. 11.4 Les Très Riches Heures du Duc de Berry – Limbourg Brothers, detail from Calendar –
Folio 3, March, painting on vellum, 22.5 cm  13.6 cm, between 1412 and 1416 and ca. 1440,
Musée Condé. Land utilisation practices such as pruning vines, tilling the land, preparing for
sowing and herding

World Landscapes

One of the first to produce paintings with the landscape as the core subject was the
Walloon painter Joachim Patinir (c. 1480–1524). His friend Dürer called him “the
good landscape painter” (“der gute Landschaftsmaler”) and he was recognised as
the first real landscape painter in Western Europe. His works show imaginary
panoramic scenes from a bird’s eye perspective as setting for telling a myth or
legend. Figure 11.5 shows a typical example. The story reads commonly from right
to left and offers a visual walk through the different parts of the landscape. The
landscape is an imaginary composition of presumably real landscapes such as rock
formations (in Patinir’s work very similar to the ones near his native town Dinant),
villages, houses and vegetation typical for the Low Countries, and distant mountain
ranges as seen by travellers. The different scenes are theatrically ordered in
successive depth planes. People and human artefacts are miniaturised (Huvenne
This genre was successful and inspired artists from the Danube school, such as
Albrecht Altdorfer, and Pieter Breugel the Elder in Flanders, who added the realism
of daily life in the scene (van Hogendorp Prosperetti 2009). This painting style
11.2 Landscape Painting – The Imagined Landscape 323

Fig. 11.5 A world landscape by Joachim Patinir: Landscape with the Flight into Egypt
(ca. 1516–1517), The painting a fictive landscape, but is composed by a series of realistic scenes.
Presumably, the painter got the inspiration for the rocky outcrops from the ones near his native
town of Dinant. Atmospheric perspective is used to indicate successive depth layers. Oil painting
on panel, 18.3  22.4 cm. Antwerp, Koninklijk Museum voor Schone Kunsten)

gives “an idealized composite of the world taken in at a single Olympian glance”
(Schama 1995) and “were imaginary compilations of the most appealing and
spectacular aspects of European geography, assembled for the delight of the
wealthy armchair traveller” (Harris 2008). In Germany, this genre became known
as “Weltlandschaft”, i.e. “world landscape” (Huvenne 2004; Zinke 1977), and is
also referred to as “the northern realism”. Nevertheless, in this kind of painting, the
landscape as a whole is not realistic at all, but rather a background setting of the
graphical storytelling. However, the details in the composing scenes are realistic
(cf. the comments on The Hay Harvest of Pieter Breugel, Fig. 3.1 in Chap. 3).

Small Landscapes

This style refers to drawings and prints of pure landscape scenes without religious
of mythological narratives. Significant are the series printed and successfully
published by Hieronymus Cock in Antwerp. The first series of 14 sheets was
324 11 The Artist’s Landscape

published in 1559 with the Flemish title “Vele ende seer fraeye gheleghentheden
van diversche dorphuysinghen, hoeven, velden met alderhande beestkens verciert.
Al te samen ghecpmterfeyt naer dleven”, which can be translated shortly as: “Many
and very beautiful places of diverse village dwellings, farmsteads, fields decorated
with all sorts of animals. Altogether made from life”. The title page only stated that
the landscapes were drawn from nature in the vicinity of Antwerp (Silver 2012).
Cock published a second series of 30 sheets in 1561 with the title “Elegant pictures
of country manors, villas and cottages drawn from life in copper plates”. A third
edition was published by Philips Galle in 1601 in Antwerp entitled “Small Counties
and Villages primarily in the duchy of Brabant”. The Flemish artists were not
identified and the style became known as the ‘Master of the Small Landscapes’
(in Dutch: ‘Meester van de Kleine Landschappen’) (Schmidt et al. 2014).
The commercial success of the landscape prints was made possible by the
innovation in printing technology brought by Johannes Gutenberg in the early
fifteenth century. The Small Landscape prints, as well as the work of Pieter Bruegel
the Elder and the painters of world landscapes, established landscape art as an
independent genre and characteristics for the northern Renaissance art. This new
genre reflects also a shift in the clientele for art towards the urban middle-class and
illustrates a move towards a humanist worldview, using the countryside as a
metaphor of an Arcadian, peaceful life (Fig. 11.6).

Fig. 11.6 Master of the small landscapes, Village road with a flock of sheep, etching on paper, h
134 mm  w 203 mm, etching probably by Johannes or Lucas van Doetechum, published by
Hieronymus Cock in 1559–1561 in Antwerp, Rijksmuseum
11.2 Landscape Painting – The Imagined Landscape 325

Topographical Views and Panoramas

World landscapes are pictorials using only the scale and atmospheric perspective to
create an illusion of depth, but have no coherent geometric perspective, except for
the independent scenic details. This is necessary when specific landscapes have to
be represented so they can be recognised. Typical are oblique city views from a
fictive elevated viewpoint. Many give panoramic views of the landscape. The use of
a topographical view, offering a map-like representation of the landscape. These
painters-mapmakers are the predecessors of cartographers.
In China, panoramic paintings are an important subset of handscroll paintings.
The most famous one is Along the River During the Qingming Festival, painted by
Zhang Zeduan (1085–1145) in the Song dynasty. It shows in great detail the daily
life of people and the landscape of the capital Bianjing, today’s Kaifeng (Fig. 11.7).
A typical Chinese scroll can be several meters wide and is viewed by unrolling it
from right to left, thus revealing the story in time and every scene is seen sequen-
tially. Consequently, no explicit vanishing points and no explicit light source or cast
shadows should distract the view. Therefore, the lines along the vertical axis are
drawn as parallel lines, giving the effect of placing the horizon at an imaginary line,
infinitely high above the painting. This gives in an axonometric projection applied
on all buildings separately.
The Chinese scroll painting ‘Along the River’ is also a good example of such a
panoramic view.
Another example of a panoramic painting, but with one fictive viewpoint and
the use of oblique perspectives, is the Panoramic View of Ghent, realised by an
unknown painter and dated 1534 (Figs. 11.8 and 11.9). It shows the whole

Fig. 11.7 Detail of Cathay on the Chinese scroll painting Along the River During the Qingming
Festival (length 11 m). Projection lines indicate several vantage points and an oblique projection
326 11 The Artist’s Landscape

Fig. 11.8 The Panoramic View of Ghent (Ghent, STAM). North is on the right side of the
painting. Small white arrows indicate perspective lines of some buildings indicating the use of
oblique projections. Some important landmarks are connected by a polygon and compared to their
position on the topographical map (the shape indicated by the polygon ‘map’), giving an idea of
the oblique foreshortening. The rectangle indicates detail of the three towers (see Fig. 11.9)

walled city in detail with a distorted geometry to fit in the frame. However, the
important buildings are more or less at the right position relative to each other.
Also, there are indications that the painter used tower views to position the
landmarks. All buildings are represented in a cabinet perspective, but not in a
systematic manner.
This kind of topographical landscapes appeared in paintings, drawings and
prints and they intended to be a documentary rather than rather purely aesthetic
fantasies. These topographical townscapes are the precursors of map-making dur-
ing the Age of Discovery. Mapping the World

During the Ming Dynasty, the Chinese coastline and rivers were already accurately
mapped around 971 CE. Meanwhile, in medieval Europe, maps were mainly
symbolic and a figuration of classic geographical descriptions.
In the fifteenth century, overseas explorations and trade intensified and marks
the beginning of the Age of Discovery. In this period globalisation begins with the
11.2 Landscape Painting – The Imagined Landscape 327

Fig. 11.9 Detail from the 1534 panorama of Ghent with the three towers showing the perspective.
The view lines from the top of the cathedral to the parish church (a) and Belfry (b) are compared to
their position on the topographical map (a and b’), showing no significant difference, hence the
painter used views from the towers to position the main buildings. Different cabinet projections are
used: the cathedral (c) is depicted using an angle at 60 and the Belfry (d ) with an angle at 40 ,
which matches the different orientations of the buildings

expansion of the European culture and rise of colonialism and mercantilism. There
was a growing demand for accurate navigation maps and surveying instruments.
New printing techniques opened a market for the distribution of maps. The end of
the sixteenth century became the golden age of Dutch and Flemish map making
with inventive cartographers as Gemma Frisius, Gerardus Mercator and Jacob van
Deventer. In 1589, Mercator bundled maps in the first atlas. At the end of the
seventeenth century, the French astronomer Giovanni Cassini instigated the first
modern to