NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment

September 6-9, 2004

Lecce, Italy

Minimum Data Sets for Landscape Indicators using GIS

(A Case Study: Copa Mic Area, Romania)

Octavian-Liviu Muntean1, Lucian Drgu2, Nicolae Baciu2

1
Babe-Bolyai University, Faculty of Environmental Science, P-a tefan Cel Mare, No. 4,
400192, Cluj-Napoca, Romania, liviumuntean@yahoo.com , Phone: 0040-264-405316, Fax:
0040-264-599444
2
Babe-Bolyai University, Faculty of Geography, Clinicilor Street, No. 5-7, 400006, Cluj-
Napoca, Romania, lucian@geografie.ubbcluj.ro , nicubaciu@yahoo.com, Phone: 0040-264-
596116, Fax: 0040-264-597988

Key words: DTM assessment, land cover assessment, GIS, landscape indicators,
environmental decline

Abstract (Times 12pt bold, left align)

This work is part of the NATO / CCMC national pilot study named
The assessment of environmental decline in Copsa Mica area (Romania). The
national pilot study is foccusing on the roots and the consequences of the environmental
decline. The research was focused on the environmental impact assessment (EIA), integrated
territorial assessment (ITA) and a SWOT analysis of the area. The assessment of the
environmental decline was based on four categories of indicators: social, economic, health
and environmental. We have tried to establish a relational indicator, viewed as a tool for
community development and decision-makers. The methodology and results of assessment
(assessment matrix, environmental map and plan of environmental reconversion) will be
implemented into the local environmental planning.

Introduction (Times 12pt bold, left align)

Policy makers and the public at large need reliable and well-synthesised information
about the environment without getting lost in detail. This is why experts have recently
expressed increasing interest in a reduced number of environmental indicators selected from
existing larger data sets. The call for higher-order, more integrative indices is also becoming
louder. At the same time reservations continue to be voiced about the limitations of
aggregated indices, their perceived opacity, and potential for misinterpretation (OECD,
2002).
By combining the information contained in two or more indicators, aggregated indices
make it possible to convey simple messages about complex environmental issues. Among
their strengths is the potential to simplify the public communication process and to reach
audiences that currently receive little environmental information at all. However, reducing the
number of indicators by condensing information also runs the risk of misinterpretation
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy
because users are not always aware of the scope and limitations of the index methodology,
and because the message conveyed may be distorted by data gaps.
However, the indicators relevance varies by country and by context. They must be
reported and interpreted in the appropiate contex, taking into account places different
ecological, geographical, social, economic and institutional features (OECD, 2004; UNEP,
2004).
The article discusses the various aggregation methodologies used for each of the
indicators and identifies the strengths and weaknesses of the resulting indices for decision
making and public information. The methodology is applied to assess the environmental
decline in Copa Mic area and Trnava Mare Coridor (Romania).
The Copa Mic Area, one of the Europes most polluted sites, is situated within
Trnava Mare Corridor (part of Transylvania Tableland) in central part of Romania and its
environmental decline is due to the pollution type and its effects upon territorial systems,
human health, ecosystems, quality of life and more environmental components (air, soil,
water, vegetation, fauna, other human components etc).
From environmental point of view, Trnava Mare Corridor is a very interesting region
because the relations between environmental system and human activity are very unstable. As
an economic and human axe of the Trnavelor Tableland, the corridor is a distinctive region,
geographically and environmentally (junction environmental region).
We may consider Trnava Mare Corridor as a running and convergent
environmental model (Muntean, 2003), where Trnava Mare River and Copa Mic Area
were determinant factors. Practically, the human activities capitalized the entirely natural
components and influencing them at a large scale.

Tabel 1. Landscape metrics on 1:50000 land cover map basis

Index NC MPS ED SHDI

Settlement (ha) (m/ha)
Copa Mic 8 908.50 170.30 1.70
Trnava 9 1118.08 123.19 1.71
Micsasa 9 592.24 103.37 1.65
Axente Sever 8 747.67 104.71 1.64

Tabel 2. Landscape metrics on 1:50000 CORINE Land Cover dataset basis

Index NC MPS ED SHDI
Settlement (ha) (m/ha)
Copa Mic 6 1057.68 62.98 1.54
Trnava 9 1118.61 62.44 1.76
Micsasa 9 580.94 51.44 1.71
Axente Sever 7 845.26 58.83 1.68

Write introductory text here. (Times 12pt, justify)

Figures and tables should be incorporated in text and placed at the end of the paper after the
references. Place all captions on a separate sheet and remember to number tables and figures.
For equations use this layout:

2 xi x 2 (1)
x
i n

Use only grey colour images with high contrast. Send also a copy of each image in TIF
format at 300 dpi.
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy
References should be cited as (Quinlan, 1992) or (Breiman et al., 1984) or (Quinlan, 1992;
Dietterich, 1997).

The study was structured on the following steps:

- Setting up the indicators: environmental indicators; quality of life indicators; others
indicators (demographical, social, health, geographical);

- Configuration of database (type and quality of database);

- Using the data sets and indicators on GIS support.

The methodology of analysis and assessment of environmental decline (Muntean et al.,

2003) is based on:

- Integrated Territorial Analysis (ITA) of geographical components;

- Environmental Impact Assessment (EIA) of environmental decline (based on RIAM

matrix elaborated by Pastakia and Jensen, 1998);

- Aggregated Environmental Index (AEI, viewed as a tool for planners and decision
factors);

- EIA based on combined Leopolds Matrix and RIAM (Muntean, 2004);

- The assessment of human pressure (index of human pressure based on following

indicators: demographical pression; environmental resources, infrastructure; homan
impact on the environmental components);

- The build up of environmental data base and using GIS approach based on large part of
these indicators.

The selection of environmental indicators and building of minimum data sets.

All data and indicators of our study were selected from official data and scientific
reports (Environmental Assessments and Studies, Urban General Plan, Legal Normatives,
Environmental Protection Agency reports)

The environmental indicators is regarding to: terrain attributes, water, atmospheric

features, natural vegetation, fauna, soils, buildings, habitational space, green spaces,
landscape indicators.

The quality of life indicators (social, economic, health) is foccusing on: human health,
life expectancy, rate of unemployment, educational achievement, level of incomes

The Aggregated Environmental Index (AEI) is an expression of the RIAM matrix

assessment. AEI has some advantages: is easy to read and understand by decision makers
and assessors; facilitate an objective comparison between different areas and environmental
components; is useful to realize a map of environmental impact.
The selection and building of the minimum
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy
data sets are actions which depends on the data quality, data accesibility, importance and
relevance of environmental data within the Romanian context. Despite of all official reports
this context is characterized by uncertainty and partial lack of environmental data.

Using the environmental database and indicators on GIS support.

GIS techniques sustained the deline assessment and they were used for:
mapping the environmental components; cartographic presentation of the EIA results
(environmental scores); evaluating the human pressure on the territory
and mapping the land use/land cover categories of Tarnava Mare Corridor.
The main outcomes of these actions were represented by some thematic maps of
Copsa Mica area and Tarnava Mare Corridor:
map of environmental impacts; map of human pressure on territory; map of land use/land
cover (European CORINE base, 1992).

The selection of minimum data sets for landscape indicators. We consider that the
standardization amongst CCMS partners is a very important point in order to get comparable
results, based on the similar methodologies.
Even if methods are the same, their application on different data sets (in term of
quality, resolution, scale) will lead to different outputs. The standardization is a requirement
which is strongly dependent on data availability and accesibility at local and national level.
We realized an assessment of the most available data sets, whose results could be used
as a starting point towards the necessity of data sets standardization. In our case, these types
of datasets are:
Digital Terrain Models (ASTER DEM, SRTM) to derive terrain attributes and Land Cover
data sets (CORINE LC) to capture landscape structure indicators.

The DTM assesment was based on ASTER (Advanced Space borne Thermal
Emission and Reflection Radiometer) DEM and SRTM (Shuttle Radar Topography Mission).
These DTMs were chosen to be evaluated because of they are the most available DTMs.
They were compared against 10 m contour lines generated DTM and we observed some
errors in the stream flow direction. The visual comparison between different DTMs shows us
these errors (see Figure 2).

Land Cover Data Sets Assessment. Because of it is broadly available for European
space, we choose the
CORINE LC data set in our evaluation.
Some landscape metrics index, namely number of classes (NC), mean patch size (MPS), edge
density (ED), and Shannons diversity index (SHDI) were calculated (using Patch Analyst
extension of ArcView package) on the basis of both CORINE data set, and a 1:50000 land
cover map.
Landscape metrics on 1:50000 land cover map basis and CORINE LC dataset are
presented in the tables 1 and 2.
Although the two data sets presented are built at different scales (1:50000 vs.
1:100000), a great part of index are surprisingly close in term of values. Even if absolute
values are very different (see ED), they are however proportional, which means that
landscape structure is correct assessed eventually. Number of classes were reduced in two
situations, because of merging together small areas of less important land cover categories (e.
g. vineyard).
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy
Very close values are obtained for MPS and SHDI. However, for the commune
Unguras, which covers a more fragmented hilly area, the values of SHDI are quite different.
This happens because areas less than 15 ha are captured in the CORINE data set as
homogeneous agricultural areas.

Results and discussion

Some discussions on DTM assessment.
Although there are important differences in resolutions between SRTM (90 m) and contour
lines generated by DTM (10 m), elevations stand within a range of difference of under 50
m.
Differences in elevation are recorded mainly in the sharp contact areas. ASTER DEM is very
different from the contour lines generated DTM in term of elevation range ( 150 m).
Practically the two surfaces appear too far away from each other to be useful. Better
results might be obtained by using absolute models, but this would make the acquisition more
complicated. Whether the 90 m resolution of SRTM is considered satisfactory, and its
assessment in other areas produce results as good as in our area, this dataset could be used
within the CCMS national projects in order to assure comparable outputs.
Some discussions on Land Cover Data Sets Assessment. After this preliminary
assessment, CORINE data set seems to be confident enough for landscape metrics
applications.
This assessment and its results could be discussed and applied in the future development of
our project.

Conclusions
The next step in our study is to develop a distributed model of the landscape in order to use
resulting facets as basis for the environmental assessment and planning. This model is based
on Digital Terrain Model analysis and object-oriented image analysis (eCognition 3.0),
following a methodology created by Blaschke & Dragut, 2003.
Also,
the study will be developed together with Prof. Dr. R. Lenz from University of Nrtingen
(Germany).
The next challanges is the validation of the methodology, indicators, database and results in
practice through the calibrating of the spatial model of Copsa Mica area and Tarnava Mare
Corridor.
We shall try to realise some comparisons between our results and similar studies or
methodologies made in other countries.

Acknowledgements
This work was supported by National Scientific Research Council, Romania through the
Grant CNCSIS 50/364
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy
References

Blaschke, T., and Drgu, L. (2003),

Mac, I., Muntean, O.L., (2000) Geosystem: constraints and environmental decline, The 29th
International congress of geography, Abstracts book, pages 189-190, Seoul.

Muntean, O. L. (2004) Impactul antropic asupra mediului nconjurtor n Culoarul Trnavei

Mari (sectorul Vntori-Micsasa). Studiu de evaluare i planificare a mediului
nconjurtor. Ed. Casa Crii de tiin, Cluj-Napoca.

Muntean, O. L., (2003), A study of environmental reconversion in Copa Mic area

(Romania). Wrzburger Geographisches Manuskripte, Heft 63, Wrzburg am Main.

Muntean, O. L., Drgu, L., and Baciu, N. (2003) Environmental decline assessment in Copa
Mic area (Romania). EcoSys, Kiel

Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J. (1984) Classification and
regression trees. Wadsworth.
Quinlan, J. R. (1992) Learning with continuous classes. Proceedings of the Australian joint
conference on artificial intelligence, pages 343-348. World Scientific, Singapore.
Dietterich, T. G. (1997) Machine-learning research: Four current directions. AI Magazine,
18(4):97-136.

Copsa Mica Area The Environmental Context

COPSA MICA (5000 inhabitants) is an environmental area within TARNAVA Mare Corridor
(located in central part of Romania, Transylvanian Tableland)
After 1989, COPSA MICA area became a subject of environmental concern even at national
and international level and obtained the status of less-favored area of Romania in November
2000 (status extended for 10 years)
lThe landscape fragility and environmental decline have been generated by old and severe
pollution with heavy metals (lead, zinc, cadmium, copper) and black smoke
lThe rate of unemployment is very high (more than 60% in year 2004)
The lifes expectancy rate is lowest with 10 years (about 50 years) than medium national rate
lThe economic and social environment are depending on single major plant: SC Sometra SA
(Mytilineos Holdings, Greece), the main polluter of area
lThe advanced environmental decline is constraining the quality of life and the state of
territorial system

The Geographical Location of Copsa Mica Area in Romania and Tarnava Mare River
Corridor

1. Bardac, D. (1999), Copa Mic. Elemente de monografie medical i social, Editura Tribuna, Sibiu.
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy
2. Lenz, R. (2001), Environmental Indicators as Landscape Assessment Tools?, Development of European
Landscapes, IALE European Conference 2001 Proceedings, Tartu, Publicationes Instituti Geographici
Universitatis Tartuensis.
3. Mac, I., Muntean, O. L. (2000), Geosystem, Constraint and Environmental Decline, Abstract Book, The 29th
International Geographical Congress, Seoul.
4. Pastakia, C. M. R., Jensen, A. (1998), The Rapid Impact Assessment Matrix (RIAM) for EIA, Environmental
Impact Assessment Revue, 18, 461-482, Elsevier.
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy

Figure 1. The Geographical Location of Copa Mic Area in Romania and Tarnava Mare River Corridor
 NATO/CCMS Pilot Study on the Use of Landscape Sciences for Environmental Assessment
September 6-9, 2004
Lecce, Italy

The mapping process and extraction of the LU/LC categories were based on
CORINE LC Program (1992)

Figure 2. Map of Land Use/Land Cover in Copsa Mica Area and Tarnava Mare Corridor