GTZ im TZ-Verlag

Nill, D.; Schwertmann, U.; S bel-Koschella, U.;

Bernhard, M .; Breuer, J.
Soil Erosion by Water
in Africa
Principles, Prediction
and Protection
Deutsche Gesellschaft fur
Technische Zusiimmenarbeit (GI )H
Die Deutsche Hibliothek - CIP-Einheitsaufn;ll~r~~e
NU, D.; Schwertmann, U.; Sabel-Koschella, U.; Bernhard,M.;
Breuer, J.:
Soil Erosion by water in Africa / Nill, I).; Schwert~nann, [I.;
Sahel-Koschella, 17.; Bernhard,M.; Hreuer, J .
[Deutsche Gesellschaft fiir Technische Zusarnmenarlxit (GTZ)
G1nhH1.- Kossdorf: TZ-Verl.-Ges. 1996
ISBN 3-88085-5 14-5 (GTZ
NE: Nill, I).; Schwertrnann, LT:
I ~l l i s l i ecl 1 : l>eutsche Gescllschaft fiir
Technische Zusammen:ir-lxit (GTZ) G 1lll3H
1'ostk~ch 5 180
D-05726 Esch1x)rn
Contents
Contents
List of tables
List of figures
Preliminary note
1 Causes for soil erosion
2 Damages caused by erosion
2.1 Damages in agriculture (on-site)
2.2 Off-site damages
3 The erosion process
4 Soil loss determining factors
4.1 Rainfall
4.2 Soil properties
4.3 Topography
4.4
Cover, tillage and protection techniques
4.4.1 Reduced tillage and notillage
4.4.2 Contouring
4.4.3
Soil cover by organic mulch
4.4.4 Inorganic mulch
4.4.5 Surface forming practices (contour ridging,
heaping, tied ridging, bedding etc.)
4.4.6 Bufferstrips
4.4.7 Contour bunds (stone bunds, earthen bunds,
diguettes)
4.4.8 Protective ditches
4.4.9 Terraces
5 Indicators for soil erosion
6 Assessment of soil erosion
6.1 Rainfall simulator studies
6.1.1 Laboratory studies with simulated rainfall
6.1.2 Field studies with simulated rainpal
6.2 Runoff plots
6.3
Erosion measurement within existing fields
6.3.1 Erosion nails
6.3.2 Sediment traps
6.3.3 Diverse techniques
6.4 Sediment yield from river basins
7
Soil loss prediction with the Universal Soil Loss Equation
7.1 The erosivity of rain (R Factor)
Contents
7.2 Soil erodibility (K factor) 99
7.3 The topographic factor (LS factor) 1 1 I
7.4 The cover and management (C) factor 117
7.5
The effect of protective methods -
Support practice factor (P) 145
7.5.1 Contouring, contour-ridging, tied-ridging 145
7.5.2 Bufferstrips 153
7.5.3 Contour bunds and heaps 156
7.5.4 Ditches and terraces 158
7.6 Soil loss tolerance limrts 160
Annex 1 Rainfall and erosivity 163
Annex 1 . 1 Erosivity for single sites 1 64
Annex 1.2 Erosivity regressions 168
Annex 1.3 National iso-erodent maps 171
Annex 1.4 Regional iso-erodent maps 177
Annex 1.5 National rainfall distrib~~tion maps 179
Annex 1.6 Rain volun~e and distribution for \ ~ngl e I94
Annex 1.7 Estimation of the erocivity of the 10 year stonn 2 10
Annex 2 Slope length and gradient 2 12
Annex 2.1 Device for meawing \lope length and gradient 2 13
Annex 2.2 Conversion o f slope gradient in degrees to percent 2 14
Annex 3 Cover and management factor 215
Annex 3.1 Nurnber of day in the year- and co~responding datc 2 16
Annex 3.2
Field methods for the meacuremcnt of mulch
cover and canopy cover 217
Anncx 3.3 Growth curve\ for mono- and mixcd crop4 226
Annex 3.4 Detailed C factors 230
Annex 4 Protection and management 257
Anncx 4.1 Detailed support and management ( P) factor\ 258
Ani ~cx 4.2
Some useful $pecic$ for soil and water
conservation 263
Literature 269
List of tables (short titles)
Tahle 21-1:
Table 21-2:
Table 3-1 :
Table 42-1 :
'l'able 42-2:
Table 449-1:
Tahle 5- 1 :
Table 64-1:
Tahle 7 1-3:
Table 72- 1 :
'l'able 72-2:
Table 72-3:
Table 72-4:
Table 72-5:
Table 72-6:
Table 73- 1 :
Table 73-2:
'I'able 73-3:
Table 74-1 :
'l'ahle 74-2:
Tahle 74-3:
Table 74-4:
Table 74-5:
Table 74-6:
Table 74-7:
Sediment enrichment ratio for different grain
sizes on an Alfisol 17
Sediment enrichment ratio for organic carbon
and major nutrients under different cropping
systems 17
Mean runoff coefficients from natural rain 28
Soil loss on different barefallow soils 3 5
Soil propel-ties influencing soil erosion 36
Characteristics and applications for different
types of terraces 56
Soil erodibility of soils from different
parent materials 66
Annual suspended sediment load of African rivers 83
Erosivity and rain data for sites,
countries and regions 93
Soil loss and erosion depth under barekillow 100
Structul-c classes for use in the USLE 102
Permeability classes as used in the USLE 102
Determination of per~neability class by profile
information 103
Runoff, soil loss and soil properties for the
three erodibility groups 107
Convcrsion of Keqa to Ktrop for soils in
group 1 and 2 1 1 0
Slopc lcngth exponent ( m) for tiifferent gradients 1 12
Soil loss of slope segments
115
Examplc fur the consideralion of ;in irregular
slopc with changes in soil erodibility i n the USLE I I6
Crop stages as del'ined for the USLE 117
C hictor for a groundnut-maize system I I9
Annual erosivity distribution for
Douala/ Camcrow 13 1
Residual effects of savannah and forest fallows 127
Annual C I'uctor calculated with the erosivity
clistrihution of sitcs from diffi.1-ent climatic zones 128
Average C k ~ t o r for notill
130
Subfactor c2 for the cl'l'ect of mulch covcr 132
Contents
Table 74-9a: Average C factors for forest, bush and grass vegetation 133
Table 74-9b: Alternative determination of C factor 134
Table 74-10: Average C factors for banana 134
Table 74-11: Average C factors for pineapple 136
Table 74-12: Average C factors for cassava 137
Table 74-13: C factors for ~niscellaneous perennial crops 138
Table 74-14: Average C factors for groundnut 139
Table 74-15: Average C factors for maize 140
Table 74-16: C factors for millet and sorghum 141
Table 74-17: C factors for upland rice 143
Table 74-18: Average C factors for miscellaneous crops 144
Table 751-1: P factor for contouring 146
Table 751-2: Correction of P factors for ridges with side slopes 150
Table 751-3: C x P factor for temporary established ridges 153
Table 752-1: P factors for bufferstrips as calculated by the RUSLE 156
Table 753-1: Correction factor for bunds on different slopes 158
Table 754-1: Sediment delivery ratios for side-slopes 159
Table 76-1 : Rates of soil weathering 161
Table 76-2: Tolerance limits proposed for tropical soils 162
Table 11-lAnnex: Erosivity and rain volurne for single sites 165
Table 12-IAnnex: Regressions for the calculation of erosivity 168
Table 16-1Annex: Annual rain volume and monthly distribution 194
Table 34-1Annex: Detailed C factors for forest, bush and
grass vegetation 230
Table 34-2Annex: Detailed C factors for banana 234
Table 34-3Annex: Detailed C factors for pineapple 235
Table 34-4Annex: Detailed C factors for cassava 238
Table 34-5Annex: Detailed C factors for groundnut 243
Table 34-6Annex: Detailed C factors for maize 245
Table 34-7Annex: Detailed C factors for diverse crops 250
Table 34-8Annex: Detailed support and management fhctors
for notillage 255
Table 41-1Annex: Detailed P f'actors for contouring 258
Table 41-2Annex: Detailed P factors for ridges 259
Table 41-3Annex: Detailed P factors for mounds 260
Table 41-4Annex: Detailed P factors for bunds 260
Table 41-SAnnex: Detailed P factors for bufferstrips 26 1
Table 42-1Annex: Useful species for erosion control, etc. 263
Table 42-2Annex: Useful trees according to rainfall area 268
List of figures (short titles)
Figure 1-1: Causes at the origin of accelerated erosion 12
Figure 1-2: Importance of activities which destroy the soil resource 13
Figure 21-1: Impact of increasing bulk density on productivity 16
Figure 21-2: Influence of soil erosion on potential productivity 18
Figure 21-3: The vicious cycle of soil erosion 19
Figure 3-1: Transpot capacity increases with runoff velocity 23
Figure 3-2: Runoff velocity increases with gradient 24
Figure 3-3: Runoff velocity as related to gradient 24
Figure 3-4: A surface water layer decreases or increases soil loss 25
Figure 3-5: Runoff generation on slopes with sealing and
permeable soils 27
Figure 3-6: Infiltration as influenced by management and vegetation 29
Figure 3-7: Rain volume per unit area and surface storage decrease
with increasing gradient 3 0
Figure 445-1: Ridging, tied-ridging and bedding
44
Figure 445-2: Influence of slope on height of heaps
46
Figure 447-1: Contour bunds form small terraces after some years
49
Figure 447-2: Permeable, semi permeable or impermeable bunds
49
Figure 447-3: Dykes form deposits which are used for cultivation
50
Figure 448-1: Diagram of drainage ditches and Fanya Juu terraces
5 1
Figure 448-2: Hillside ditches f~cilitate access and transport
52
Figure 448-3: Watershed conservation plan with hillside ditches
52
Figure 449-1 : Key parameters for terrace planning
54
Figure 449-2: Different types of bench terraces
54
Figure 449-3: Intermittent terraces can be transformed to bench terraces 55
Figure 449-4: Orchard terraces in combination with individual basins 56
Figure 5-1: Concepts of a peneplain and a pediplain 59
Figure 5-2: Pediplains at different altitudes as formed in Cameroon 60
Figure 5-3: Development of gullies from initiation to maturity 6 1
Figure 5-4: Blockslide and translational slide 6 3
Figure 5-5: Critical storm duration and intensity for landslides 64
Figure 5-6:
Sapre growth of trees caused by soil creep
65
Figure 5-7:
Runoff and suspended sediment load of watersheds with
different vegetation 67
Figure 611-1: Diagramm of a flume for laboratory rainfall sirnulation
71
Figure 612-1: Set-up for rainfall simulation tests i n the field
73
Figure 612-2: Runoff/soil loss diagramm for a dry- and a wet-run
74
Contents
Figure 62-1: Divider tank systern and Coshocton wheel for the
measurement of runoff and \oil loss 76
Figure 631-1: Set-up ofermi on nail\ on a slope 78
Figure 631-2: Measurement of nail height with a \lide calliper 79
Figure 632-1: Sediment trap for measurement of runoff and coil 105s 80
Figure 633-1: Calculating the time since exposure of tree root\ 82
Figure 633-2: Meawring the height of loqt soil by expmed tree root$ 83
Figure 64-1: Relationship between watershed drainage area and
sediment delivery ratio X 5
Figure 71-1: Strip chart of' a 20 rnrn \ t o m 90
Figure 71-2: Correction of' ero\ivity for the effect of water mulch 93
Figure 72-1: Change of \oil erodibility with cumulat~ve erosivity 100
Figure 72-2: Compar~son of calculated and mea\iired \oil erodi bll~ty 103
Figure 72-3: Mea\ured and predicted erodibilitie\ after applicat~on
of di scri n~~nant functions 105
Figure 73-1: Diagram for the determination of LS fi~ctor\ 113
Figure 73-2: Examplec tor the determination of erosive \lope length 1 13
Figure 74-1: Mean annual ero\ivity d~\tribution for coa\t;il. tnland
and northern Catncroon 122
Figure 74-2: Influence of 111ulch on so11 lo\\ 125
Figure 74-3: Subfactor as influenced by canopy cover and crop hc~ght 126
Figure 74-4: C factor for d~ffcrent banana dencitics 135
Figure 75 1-1 : P factor+ for different r ~dge he~ght r 137
Figure 751-2: Isohyete\ for the I0 year \tor111 volume 138
Figure 752-1: P fitctors tor riparian butfer\trips of d~fferent width\ 155
Figure 11 - 1Annex: Eror- as cawed by I~near ext~apolut~on of erovvlty lh3
Figure 17-1Annex: Stornl erosivity a\ related to \tomi \olunle 2 1 1
Figure 21-1Annex: Water level for n~e;r\urement of \lope-length
and gradient 2 13
Figure 32-1Annex: Me\urernent of cover by the metcr$tich method 2 17
Figure 32-2Annex: Cord and knot method for cover mc;l\uremcnt 1 18
Figure 32-JAnnex: Marh~ng out ;rn area with cl~fferent de\ ice\ 2 19
Figure 32-4Annex: Sclecteci cover:ige\ for the ciilihl;ttiol~ of the cye 220
Figure 32-5Annex: Sighting l'rainc for me;i~urement of caiiopy irnd
mulch cover 22 1
Figure 32-6Anncx: Oh\rrvntiori error as ~ntluenced by ~l i t ~l t height 22 1
Figure 32-7Annex: Modified sighting fratnc for errorlec\
coverage ar ~d cover height ~nea\ure~llent\ 222
Figure 32-8Annex: S~g h t ~n g frame w~t h mirror for tall crops 223
Figure 33-1Annex: Canopy cover of Barnbara nut. cmavalia
and cowpea 227
Figure 33-2Annex: Canopy cover of cawiva, ca\sava/mai~e
mixcrop and pigeon pea 227
Figure 33-3Annex: Canopy cover of cotton, sunflower and tobacco 228
Figure 33-4Annex: Canopy cover of groundnut and soya 228
Figure 33-5Annex: Canopy cover of maize, rice and wrghum 229
Figure 33-6Annex: Canopy cover of tea 229
Preliminary note
Enhanced soil erosion research in Africa looks back on about 25 years of
experimentation and recording with respect to soil loss prediction. A lot of
information was gathered during this time which led to contradictory results
about the applicability of the Universal Soil Loss Equation (USLE). This
book tries to synthesize the latest knowledge and to evaluate it for practical
soil loss prediction. This meant the gathering of a lot of data in tabular and
graphic form which might be cumbersome for some readers. Nevertheless,
we hope that it will be helpful to have these data assembled in order to save
time for searching in many different journals which often can hardly be
obtained locally. The ultimate aim of the book is to help understand the
processes, to make the reader sensitive for recognizing them in the landscape
and to allow him to quantify of the influence of agronomic measures on soil
loss rather than to give detailed technical data and sketches.
People starting to get occupied with erosion problems will find basic
knowledge about processes and effects and the necessary literature for more
details. The book is also thought as a help for people concerned with the
planning and realization of soil conservation activities. It allows to detect
areas of high erosion risk which in turn facilitates the allocation of
conservation efforts. The absolute results of calculations can be subject to
substantial error whereas the relative differences between single measures are
comprehensible. However, if a process varies in magnitude by a factor of
1000 (soil losses can be as small as 0.1 t/ha and as large as several 100 t/ha),
an estimate which is wrong by a factor of 2 (= IOO%I error) is still a
reasonable estimate. At the same time. this means that projects and research
should continue to improve and enlarge the database in order to improve
estimates. For this reason, the authors would appreciate to receive further
data on measurements.
In Chapters 1 to 6 the reader finds descriptive information about
causes, damages, processes, recognition and measurement of soil erosion.
Chapter 7 is dedicated to soil loss prediction with the USLE. For the pure
technical procedure of soil loss prediction the reader can refer to Chapter 7.
Causes for soil erosion
Soil erosion is a process acting over tens and hundreds of years. Its effects are
normally only obvious, if they become disastrous. Until now, research
focused on the physical causes of erosion. However, frequent failures of soil
conservation projects showed that the causes were much more complex. A
FA0 study revealed that the lack of adoption of new conservation practices
was a major reason for project failure even though technically sound
practices were used (Hudson, 1991). Today, i t has largely been agreed upon
that soil conservation will not be successful in many countries even by using
the best available PI-actices if Inan and the social, economic and political
conlext are not considered. Fones-Sundell (1992) summarized the problern
very pointedly by saying: 'Neither engineering nor biological measures alone
can eradicate erosion in a socio-economic system which makes non-optimal
use of natural resources a necessary and often profitable form of behaviour
for the individual.'
The cause-effect diagram in Figure 1 - 1 illustrates the problem of
soil erosion. Soil erosion as caused alone by natural, physical factors
(clim3te. soil, topography) is known as 'geologic' or 'normal' erosion
(Bennett & Chapline, 1928). It is small enough to allow the sustainable
growth of a natural eco-system. Geologic erosion ranges from several
hundred kilogran~s per hectare for tropical bush and grass vegetation to
below 100 kglha for tropical forest (Nill, 1993; Roose, 1975). Soil formation
in the tropics is supposed to be in the same range. According to
measurements in Central Africa, 150 to 400 kglha of new soil is formed each
year (Owens, 1974). Dunne et al. (1978) found an annual formation rate of
150 to 300 kglha in Kenya whercas Kaye ( 1959) reported a rate of 15 t/ha on
limestone i n Puerto ~ i c o l .
Soil loss becomes critical if socioeconomic and political factors favour
erosion (man-induced erosion). The main factors are:
1 more detail\ a b o ~ ~ t soil torination ratcx arc given in C'hirpter 7.6.
Chapter 1
Figure I - 1: Physic-ml as n3c.11 as socio-ecot2ot~zic crnd political c-ctuses (ire clt
the 01-igitz of crc*c.elerc~tucf ut-osiorz
Poverty of the farmers:
Small farmers are obliged to cultivate their land as often as possible in
order to assure their subsistence. The lack of capital hampers the
application of intensive conservation measures and the use of inputs to
restore soil fertility. Decreasing soil fertility leads to the extension of the
cropping area, soil mining and finally migration of the farmers. I t is
estimated that deforestation proceeds thirty times faster than reforestation
(FAO, 199 1 ). Overgrazing, deforestation and agricultural use are major
fxt ors for soil destruction (Figure 1 -2).
soil erosion
. -------------- r(
I
4
socio-economic
physical causes
political conditions
man-induced
erosion
geologic eiosion
Y
human activity
A i A
v v
1 I
otl
'iculture (28.0%) -
over-grazing (35
llnbalanced population density:
O\c~-lx)pi~l:it~oll u\uiilly I \ dctri~ncnt:~l lor the \ o ~l ~-tt\oi~rcc.
O\erpopi~lation i \ not only the rc\ult o f a high popillation ~sowt l i . LC11c\t
cle\clopment\ \u,gge\t that segional ovespopulatio~i i \ oftell clue to
1111$1-:1tion cau\ed by stseam\ of' rel'ugee\ from bar \ os cn\ i ~-onme~l t ~i l
cl~\a\ter\. For example, tlul-ing thc Sahel tl~oi~glit 01' the early 70\. 1 ~ i l ~ l l ~ o n
Rl~l-hint.\e cclu~il to one {ixtli ol' the co~~nt r i c\ popillation. left the11 l i o~~i e \
( FA( ) . 1900).
lJ~lde~-lxq' t~lat~on al\o cau\e\ \el-lous clalliagc. Voscl ( I O X X ) clc\cr~bcd ~ h c
dctc.t.~orat~o~i ot' a traditional re1-1-accd ago-5y5tt.m i n Ycmcll al'tcr thc
~i l i g r ; ~~~o n of the rural popi~l;~tion to ~i e~gl i boi ~~- i ~l g coi ~~i t ri c\ . Soil
c.o~l\cr\:~tion ~ o r h s 111 11ld1;~ bere ol'len :tbanclonecl clue to the rccruil~llenl
ot thc [ii~rha. wh~cti wese the iilorc acllcc i n \oil con\ e~- i at ~on. 111to the
K1.1ti\11 Arilly (Blaihic. 1OS5 1.
'I'he in\titutional frame:
Go\ clnmcnl in\lili~tion\ ~n\ i ~t f ~c~el l t l y contsol the u\e of natul-,ll r.esoulce\
' I lie I:~ch ot tlet~necl con\cscation ~ ~ o l ~ c y clilcl l abs to i.cpi~l,~tc the Lr\e of
I,~ncl. to de l ~ne tllc ow~l e r \ I ~~p ~illd to I ~i c~l ~l al c llie co~ll~lle~-c~~~l~\~~tloll 0 1
yood\ p~-c\ cnt \ t;ismer\ to apply \or1 con\er\ 1112 product~on method\
Politicians often give priority to short term benefits from export crops on
the expense of soil conservation. Extension and conservation services are
mostly inadccluately equipped and trained and suffer from a lack of
cool-dination (Sheng, 1989).
The land tenure system:
Only farmers who own their land or have secure access to them land for a
long time arc interested In longterm rnalntenance of thi\ I-csource.
Restricted access to fert~le land for social groups or farnily members (e.g.
Y O L I I I ~ people, wornen) lead\ to the exploitation of stccp slope\ and
mal-ginal, fi-agile soils. The traditional heritage system \ometimes favours
an extensive fragmentation of the land which obstructs the adoption of
conservation practices.
Tradition, believes and illiteracy:
The degree of illiteracy intluences the adoption of new conservation
practices and other cultural techniques. Small farmers commonly perceive
erosion as a natural process and are not aware about its influence on
productivity. Lack of knowledge exists along with effective local
conservation methods (Tato & Hurni, 1992). The adoption of new
practices always needs a11 effort and includes some risks. Farmers, as inost
other social groups, need time to adopt new ideas.
It is wrong to conclude from these comments that measures should
be solely applied riccording t o the socio-economic and political conditions.
Soil loss by erosion is irrever-sible. Therefore, conservation activities can not
be delayed unt i l socio-economic and political conditions are favourable.
Conservation thinking and conservation activities must proceed sirnul-
taneously.
Damages caused by erosion
Surplus rain water leaving a fielti on the soil surfi~ce is called runoff. Runoff
first causes dutnage on the field (on-site damage) by entraining fertile topsoi 1
and by reducing the available iumount of water for plant growth. Once left thc
I'ield. the runoff is enlarged from adjacent fields ancl may entel- rills, ditches.
s111all rivers, passes lakes and stl-earns and finally reaches the ocean. On its
way, sediment is picked up arid deposited which causes further darnage
outside the Sields (off-site darnage). Both, on-site and off-site cliumagc need t o
be co~~siclered i n order to assess the overall economic and ecologic effect 01'
Soil conservationists intend to protect the diverse functions of soils.
Firnctions like ini'iltration and storage capacity of a soil to prevent tloocis and
its filter function to purify water are o f tnajor concern for the urban
environl~ient. I n the rural cnvir-onment, however, it is soil fertility.
2.1 Damages in agriculture (on-site)
Soil proctuctivity depends on a nutnber of physical, chemical ant1 biological
soil properties. The ~nos t important physical ones are texture, structul-e ~unci
depth of the profile. They detel-mine the aniount of water aticl air stored in the
soil, its capability to infiltrate ; ~nd conduct water, its possibility /'or I-oot
2n)wth and the fines which can bind ant1 deliver nutrients t o t11c plant roots.
111 well structureel. deep soils even heavy storms infiltrate. Structural damage
011 sornc tropical soils is more severe than on temperate soils as show11 by the
influence of' bulk density on relative productivity (Figure 2 1 - 1 ). Profile tlepth
and surface soil depth determine the water slor~tge and t11e volumc for water
and nutrient uptake of thc roots.
Thc chen~ical fertility depencij on the aniount ot available nutrrenr\
rn : I {oil which 1 5 governed by \oil pH, 01-ganic matter content and othes
characteri\tic\. Thc\e are greatly inl'luencad by parent r11atc1-ial and the
conefition{ undet which a \oil wa\ So~.rned a\ well us by its 11~2.
r
e
l
a
t
i
v
e

p
r
o
d
u
c
t
i
v
i
t
y

[
-
I

treatment
barefallow
plow
traditional
notill
The etlrichlne~lt oj' the I'ille \oil fraction in the \edili~ent account\ for
the hlghel- nutrrent content\ of the eroded wdimcnt a\ col-nparecl to the
original \oil. Alli\on ( 1973; i n Bouwtlian, 1989) reported ;I \edimcnt
cnrichmcl~t ratio betweell 1. 3 and 5 for 5011 organic carbon. Aina ct al. ( 197C))
ticmon\tratcd the enrichment of organic carbon and m;qor nutrient$ untiet.
dil'l'cl-ent cropping \y\tenl\ (Table 2 1-2).
mean
cropping system I S O M1 N I P I K I Ca / M g
grain size [mm]
4J.002
1 . 6
2.3
1.2
1.5
1.7
barefallow
cassava monocrop
0.002-0.0.5
1.6
2.6
1.6
I .O
1.7
maize-cassava mix-crop
soybean-soybean
The iriipact of soil crosiorl on production depends on the depth of
the arable layer and the cluality of' the underlying hori7ons. Soil erwion i j
more cletriniental the shallower the \oil, t hi \ being aggravated in area\ ol'
irregular rainfall.
1.5
1.3
pigeonpea-pigeonpea
mean
7 \ cdi mr n~ r111-ichrricnt ratio = (prl-cc~itagc of a grain \ i / c clabb in rhr \ edi ~i i cnt ) 1 (prrcrlitaye 01'
rlic \ i / c i ~l a\ \ i n the 01-iyinal boil)
0.05-0.125
0.
0.8
0.7
1.3
0.9
1.3
1.3
1.3
1.3
1.1
1.3
0.125-0.25
X 1 .O
0.7
0.0
0.0
0.9
1.3
I . I
1 . 1
0.9
1 .0
1.2
0.25-0.5
0 . 9
0.7
0.0
1 .O
0.9
0.')
1.4
1.2
1.4
1.2
0.5- 1
0. 6
0.5
0.9
0.9
0.7
0.7
0.6
1-2
0.8
0. 3
0.0
0. 8
0. 7
I . O
0.8
1.2
1.2
1.3
0.0
1 .O
1.3
1.1
1.3
Loss of fertile topsoil is most harmful on extremely leached Ultisols
and Oxisols of the hulnid tropics where the subsoil contains very low
amounts of nutrients and SOM compared to the less leached soils of the drier
areas. SOM counteracts P fixation which explains why P fixation increases
with increasing topsoil loss. Mbagwu et al. (1984) showed that the removal
of 5 cm of topsoil reduced maize yield by 95% on a leached Ultisol but only
by 52% on a less leached Alfisol. Maize died off at 30 cm height on an
Ultisol in Canicroon which had lost its topsoil duri ~i g 5 years of barefallow
(Nill, 1993).
Some crops tolerate ero\ion better than other\. La1 (1976a)
mea\ured 52% less ma i ~ e yield but only 38% leg\ cowpea yield i f 10 cm o f
surface \oil wcre stripped off. Yield decrease\ are generally i n the order
gra~nineae > grain legir~nes > tuber crops (El-Swaify, 1990).
Fi,y~tt.c~ 21-2: l t ~f l ~i t v i ~* t ~ o/ soil o r o ~i o ~z 0 1 1 p o t t ~~~t i u l p ~ - o t l ~ i ( ~ t i \ ~ i t ~ ~
t.eltrtot1 t o soil t\,pe (Pic.1-c,e r t trl., 198.3)
FAVORABLE SURFACE
AND SUBSOIL
HORI ZONS
FAVORABLE SURFACE
FAVORABLE SURFACE AND CONSOLI DATED
AND UNFAVORABLE
OR COARSE FRAGMENT
SUBSOIL HORIZONS
SUBSOIL HORI ZONS
ACCUMl Jl ATED EROSION ( TI ME) -
The extcnt to which soil erosion efl'ects productivity depends on the
depth-distribution of fertility pat-a~neters in the prol'ile (t;igure 2 1-2). A dccp
itncl honiogeneous soil acts with a slow productivity decline with increasing
crosion whereas yields drop sharply with increasing soil loss on soils with
unf'avourable subsoil properties.
The econorliic damage of soil erosion is alarming. In Zirnbabwe i t is
estimated that farmers loose three times Inore nitrogen and phosphorus by
erosion than they apply to their fields. 20 to 50 US $ on arable land and I0 to
80 US $ on grazing land would be necessary to substitute these nutrient
losses by fertilizer (FAO, 1990). I t must be st]-cssed that most erosion
damages can hardly be cured (e.g. compaction, structure) or are completely
irreversible (e.g. water holding capacity).
increased
soil loss
degraded soil properties less plant cover
reduced structural stabilit and roots
Damages to agricultural productivity are not only caused by
degrading soil properties but also by direct impact of runoff. Roots and seeds
are washed out of' the soil. Seeds and seedlirigs on the foot-slopes are buried
by the deposited sedil-nent. Kills and gullies hamper access to the fields.
impede farm operations and transport. Deep rills and gullies form drainage
syste~ns which drain the adjacent areas and lead to considesable loss of water.
reduced fertility reduced plant growth
I,oss of soil fertility by soil erosion is a self-enl~ancing process. Soil
erosion reduces structural stability and soil fertility. Reduced structul-al
stability decreases infiltration and may increase the amount of transportable
rnateri:il. Reduced fertility causes poor plant growth, canopy covet- anti I-oot
soil inter-actions. In till-n. runoff and soil loss are accele~-ated (Figure 2 1-3).
2.2 Off-site damages
Part of the surfilce runof'f' ;inti the suspended sedimunt Icavc the ficlcts and
g s a ~ i ~ l g lands and are concentratecl in the surficial drainage system.
Depending o n the transport capacity of the flow. sediment is picket1 up or
deposited. Rills are widcncd to gullies, channels arc deepenecl (clia~lnel
erosion) and strean) banks ar-c undercut (stream b:uik erosion). Ditches. ro~ids
and bridges arc damaged. The kist runoff leads to :i loss of' water I'ronl rhc
landscape and results in a I xge fluctuation of' tlie rivers. Some ~-i \ . e~-s start to
become olily seasonal. Tlie groundw:itcr table is lowel-cd w11ic11 al'l'ect\ the
\ cgctation anti causes water shortages in wells.
Downstream st.clitiientation silts up irrigated fields. ditches.
channels. clams and harbours. 111 areas with intensive agriculture. pesticicles
ancl nutrients dissolved in ri~nofl' ol- ilttaclied to the sediment becorne a serious
problem. Disasters at this extent arc dil'l'icult to quantify but national
economics si~l'l'cr important expcnditur-cs 1.01- their restoration.
The amount ot 4cdimcnt transported by \olne \trcam\ can he
enormoil\. For cx~ui ~pl c. the r i ~ e r per her^.;^ 111 Kenya rcccicc\ :in a\ el-age ot
105 1 1 x 1 - ye;11- 1'I-0111 eacli 1iecla1.c of' i t \ 13 10 hm' large uatt.l.shcc1 (13ilnnc.
1075 In: Walling, 1084). corre\ponding to an avcragc l o ~ c r i ~ i g of thc
~~at er \ I i cd by ahout 15 clll in I0 yeat-\. Ilam height\ on thl-ce 01' Moroccoq\
clam\ hat1 to be ~nct-ea\etl In order to maintain the \tot-age capacity. In 01-dcl- to
prc\cr\c lllc curl-ent watel- \torage cap;icity i n Mon)cco. one tirw ~ ; I I N ~ ~ i l l i
150 million rn{ nueci\ to he con\tructcd C;ICII yea1 (FAO. 1993).
The i'lcclucnt Ilood cl~\a\tet-\ In lt~clia arc anotl~cr \bell-hnomn
cxamplc. They ,Ire expl;llncd by tile deforc\t:itio~l of the IIr~i ~al : ~qa\
Scdi ~nent \ of tlie Hrahrnaput ~-~~ i111d (;;lnge\ rlver In India hate tolmccl ,I
50000 h ~ i i ~ lar-gc \hallow In the Gult 01' k n g n l . I'hc KO\ I Kt\ el 111
Klhal-Ilndi:~ II:I\ bur.ied
15000 hrii3 of icrtilc land wrth gr.a\cl :uid wnd
( Kollm:~nnspergcr. 1079)
Stitdie\ on the \ve\t-coa\t ol Sumatra \bowed that the \eci~mcnt lo~tcl
ot tlie ~ - ~ \ c r \ wh~cl i iricrru\ecl hy ciclorc\tntion. rllrnlng and channel
construction led to the destruction of the coral riffs off-shore (Hettler. 1994).
Riffs are rich fishing grounds and help to protect the coast line. These are
some examples, out of a large number which could be cited here, in order to
\how the importance of off-site damage.
Present investments into soil conservation efforts are small compared to the
i~nrnense investment in civil engineering aiming to repair the results of
erosion. It is supposed that investing in soil conservation would have n higher
cost-efficiency ratio and would protect both the soil resource and the down-
stream areas and facilities.
3 The erosion process
Soil erosion can be regarded as a result of four processes
(Foster & Meyer, 1972):
- detachment by raindrop impact
-
transport by raindrop impact (splash erosion)
-
detachment by the shearing forces of flowing water
-
transport in surface runoff (sheet or interrill erosion, rill and gully
erosion)
Rain falling on a soil causes increasing water saturation orland the
formation of a seal at the soil surface. As a result of both processes,
infiltration into the soil is decreasing. Water on the soil surface occurs as
soon as rain intensity exceeds the infiltration rate. Before any runoff can
occur, a small amount of water is needed to humidify the soil surface
(detention storage). Once the detention storage is filled up, ponding occurs in
the small depressions and irregularities of the soil surface (surface roughness)
which form the retention or depression storage. Overflow of some
depressions provides excess water to the ponds underneath. These, if filled
up, in turn spill their water further down-slope. Thus, Inore and more water is
1
moving down-slope which may concentrate, dig out rills of increasing size
and finally may cut deep gullies into the soil.
The amount of surface runoff (SRi) during a storm can be expressed as:
SRi = Pi - ( I + DS + RS) ( 1 )
with Pi rain volume of storm i [mm]
I infiltration [mml
DS detention storage
RS retention storage
Sheet and splash erosion occur in areas of shallow sheet or interrill
flow (few millimeters deep) whereas rill erosion is caused by concentrated
rill flow. In the rills, fine sediment is transported as suspended load whereas
coarser particles are dragged along as bedload.
3 \?el- definition rills can bc closed by normal farm ope[-ations.
22
3 The erosion process
The amount of transported soil and the size of particles depends on the
transport capacity of the tlow. For a flow of given width, the transport
capacity increases with increasing flow velocity (Figure 3-1 ) and tlow depth.
Both depend on slope.
If the overland flow on a smooth surface is regarded as a water sheet with
a certain depth and velocity, a unit volume of water can be picked out as an
element with a defined weight (w) (Figure 3-2). W equals a force with a
down-slope component fl parallel to the slope and a component f2
perpendicular to the slope. The down-slope force f l increases with gradient
and speeds up the velocity of the element. In the Manning formula which is
widely used to calculate flow velocities in channels, velocity augments as a
function of the 0.5 power of slope (Figure 3-3) whereas transport capacity
increases with the cube of flow velocity (Engelund & Hansen, 1967).
Figure 3-1: Tr(rnsport cnprrcity increases with runoff veloc-it,' as sholzw by
the clr?!ount oftmn.(;yorted sedirnent (Auer.sw~nlc1, 1993)
mean flow velocity [m/s]
Chapter 3
Figure 3-2: Ruizc?f veloc-ity increnses with irzc.rea.sing grclclirnt due to t h ~
down .slope .fi)r.c.r ("'I) which is n con~ponrnt of the weight (ul) of'
gradient
3 The erosion process
Flow velocity is not equally distributed over the depth of the flow. In a
laminar flow, velocity (v) increases to the square of depth (d) (Horton et al.,
1934) which means that the water layer at the water surface is much faster
than the layer close to the soil. This velocity distribution causes a force which
lifts up soil particles from the ground and transport them. Depending on their
size, they are either rolled and dragged along the soil surface as bed load or
lifted up into the flow and transported as suspended load. In the shallow sheet
flow, velocity is s~iiall. However, soil loss is enhanced by the energy supplied
by the pounding rain drops. The drop impact causes turbulence in the tlow.
Particles are heaved up, settle down and are heaved up again, thus, being
transported towards the rills. In experiments of Mutchler & McGregor
( 1983), maximum soil loss occurred in tlow depths of 2 mni (Figure 3-4).
Figlire -3-4: A s~r~fircr cz7atrr lcryrr drc-r-eases or- itzc-rrcrses soil 1o.s~ drlwrzdirzg
or1 its cleptll (Mutc.l?ler & MC Gregor, 1983)
surface water depth [mm]
Flow depth i n rills i \ generally deep enough to minimize the
influence of raindrop action. The amount of soil loss in the rills, therefore,
depends altnost solely on the shearing forces of the flow and the saturation of
its transport capacity. If the transport capacity of the rill tlow is saturated
with 5ediriient from the interrill areas, the rills do not deepen. If the sediriient
concentration i \ \mailer than the transport capacity, the flow picks up more
sediment from the rills.
Chapter 3
Runoff is distinguished into two basic flow patterns attributed to
differences in runoff generation. Horton flow occurs if runoff is caused by a
rapid sealing of the soil which limits infiltration right at the surface (Horton
et al., 1934). Dunne flow occurs if runoff is caused by saturation of the soil
profile due to excess of rain, dense layers or shallow soil depth (Dunne
1978). Horton flow is characteristic for structurally weak soils which have
enough fine earth to form a seal. Infiltration is rapidly decreasing after the
onset of a rain even though the subsoil may still be dry. The runoff
coefficient may reach 70 to 80% for single rains. Soil loss is limited by the
amount of available sediment rather than by transport capacity. Dunne flow is
characteristic for structurally stable soils rich in oxides, clay and organic
matter. High infiltration rates can be maintained until the soil becomes
saturated. Even if enough sedirnent of transportable size is available at the
soil surface, soil loss is limited by runoff. Transport by splash erosion
becornes more important.
Rain falling on a slope causes either runoff from almost the entire
slope (sealing soils with Horton flow) (Figure 3-5a) or only from part of the
slope (soils with high infiltration rate and Dunne flow) (Figure 3-Sb). Close
to the upper slope end, runoff, even if present on structurally weak soils, is
still too small and slow to transport soil. On structurally stable soils runoff
seldom occurs on the upper part of the slope. It infiltrates into the soil and
proceeds vertically or laterally in the soil. The lateral or interflow may add to
soil saturation of the area further down-slope where runoff starts. Thus,
runoff on both soil types but Illore so on soils with Dunne flow, leaves a
'zone of no sheet erosion'. Soil profiles on the watershed boundary
are,therefore, relatively uneroded and can sometimes be used as a reference
for the extent of erosion damage on the mid and down slopes.
3 The erosion pl.ocess
Figure 3-5: Runc?ff gufzercrtiotz on slol)c..c n'ith sec/lirlg ( ( I ) cuzcl l)errnecrblu ( h)
.c.oils (clftrr Chorluy, 19 78)
4 runoff area b
+ -
4
outlet drain
interflow
The runoff volurne produced by a rain depends on rain
properties as well as soil and vegetation properties. Roose & Piot (1984)
measured mean runoff coefficients (RC) of 20 to 40% and as high as 70% for
individual storms. In own experiments with natural rain, RCs varied between
as much as 30 5% on an Alfisol to as little as 1 C/c on an Oxisol. Small rains of'
2-3 mm could already generate runoff on sealing soils (Table 3-1) (Nill,
1993). Runoff' starts on some soils only some minutes after the beginning of
rain. Pontanier et al. (1984) f'ound 1 to 4 min of artificial rain sufficient to
generate runoff on hard setting soils ('sols hard&'), 2 to 20 min on Vertisols
and 5 to 20 liiin for Ultisols. However, on the Acrustox shown in Table 3- 1 ,
1.5 hours of rain with an intensity of 64 mmlh did not cause any runoff.
Trrhlc 3- I : Mew11 ~- ~i i ~( ?f co<f f i c~i c~t ~t ~s fi-0111 i ~ o t ~ i r ( ~ l f - ~ri i ~ o 1 1 .\P\WI soils
(Nill, 1993)
Soil covered by vegetation generally infiltrates tnore watcr than
uncovered soil. Sabel-Koschella (1988) measured a 7 times higher infiltration
volume undcr a natural savannah grass fallow ( 14 I0 ~nrnlh) compared t o a
sealed barefallow ( 2 10 mmlh) (Figure 3-6). If plowed and cultivated. thc
raime (;oil infiltrated 450 mmlh.
soil
Paleustalf
Andisol
Kandiudalf
'l'rophumult
Tropudult
Hapludult
Acrustox
With increasing arca. the total runoff volume becorner more. Ar
rhown for plot4 of 70 to 550 rn' (Mutchler & Crees, 1980) and watersheds
between 0.1 and 100 k~i i ' (Dunne, 1978), runoff pcr unit area becolne5
rmaller with increasing watershed size due to longer travelling timc of the
overland flow. The longer the flow stays within the watershed the more water
and \oil can be retained in deprersionr or infiltrate.
number of storms
[-I
8 1
35 1
320
239
357
135
135
runoff coefficient
[%I
3 0
I8
I8
15
1 1
1 1
1
smallest runoff
generating storm
[mml
- 3
3
3
3
5
3
7
3 The erosion pr oce\
infiltration time [min]
Experiments on the influence of gradient and slope-length on runoff
led to varying results. The trials showed more, less or unaffected runoff
volumes with increasing slope. In seven out of eight studies in the US, t'or
example. annual runoff volume increased logarithrnical with gradient,
whereas slope length had no effect on the amount of runoff per unit area
(Wischmeier, 1966). One reason for a positive relation between slope and
runoff is the decreasing surface retention with increasing slope comparable to
a cup of water which is more and more inclined (Figure 3-7). In contrast to
these results in the US. Poesen (1984) measured less runoff with increasing
slope on sealing soils. The compaction of the soil by impacting drops is less
because the impacting force does not act perpendicular to the surface and the
number of drops per unit area is smallel- (Figure 3-7). Thus, on steep slopes
surface sealing is weaker and runoff can be smallel- than on gentle one' s.
Additionally, the number and depth of rills were higher on the steep slopes.
The I-ills dissected the seals and enlarged the infiltrating surface area.
Chapter 3
Figul-c -1-7: Rrrirr ~ y o l u ~ r ~ ~ prr rrrrit area a d surfrrce \toroge r1ecrea.s~ 011 the
slr~pe of lerlgtl~ AB with increrrsirlg gradient. Sl?lo.slz is alnrrj, ~
trcrrz.\l7ortr~cl further don,t~slopo that1 upslope
rain
I surface /
4 Soil loss determining factors
4.1 Rainfall
One driving force for water erosion is rainfall. The raindrops which pound on
the soil surface either infiltrate into the soil or leave the field as surface
runoff. The rain volume which runs off on the soil surface not only depends
on the properties of the soil, vegetation and topography, which will be
discussed in the subsequent chapters, but also on the quantity, distribution
and type of rain. Tnvestigations showed that soil loss is largely determined by
rain volume, energy load, intensity and their distribution within single storms
(Flanagan et al., 1988) and during annual seasons (Lal, 1990).
An example for the last effect was given by Temple (1972) who
noted 8 times more runoff from a rain at the end o f the rainy season
compared to a similar rain at the beginning of the rainy season.
Kinetic energy (E) of a storin is calculated by (Morgan, 1986):
with m mass of falling rain [kg]
v terminal velocity of the falling drops [m/sJ
Terminal velocity of raindrops increases with diameter to a
maximum of 9 to 10 mlsec for the largest drops which have diameters of
about 6 rnnn (Gunn & Kinzer, 1949; Laws, 1941; Laws & Parsons, 1943).
Drop diameter increases with increasing storm intensity up to intensities
between 76 and 100 mmlh (Carter et al., 1974; Hudson, 1963). Pressures
between 2 and 6 MPa are exerted to the soil for very short times (50 ms)
when a rain drop hits the soil surface (Ghadiri & Payne, 1981). This
pounding action destroys aggregates, displaces particles (splash erosion) and
has a sorting effect which leaves a thin layer o f coarser particles at the soil
surface. Thin water layers of 14 to 30%- of the drop diameter in thickness
enhance splash erosion whereas thicker layers protect the soil (water mulch)
(Mutchler & Young, 1975).
Tropical rains are characterized by high and distinct intensity peaks.
Maxirna of up to 800 mmlh are reported from Jamaica (El-Swaify & Dangler,
1982). For northern Nigeria, Kowal & Kassatn ( 1977) measured common
peak intensities of 120 to 160 rnmlh and showed that mean drop diameters
where higher in tropical storms than in temperate areas. From western
Nigeria, intensity peaks of 190 mmlh are reported (Wilkinson, 1975). Peaks
occurred during the first five minutes in more than half of the storms. Hudson
(1961) measured peak intensities of up to 340 mmlh in southern Africa. The
erosivity of storms may additionally be enhanced by strong winds (La1 et al.,
1980). In convective storms high windspeeds commonly coincide with
intensity peaks (Raussen, 1990).
In order to predict soil erosion, Wischmeier & Smith (1958) found
out that the product of a storms total kinetic energy (E) times its maxirnum 30
minute intensity islinearly related to soil loss:
R =
( E :,: 13(,) [Nlh]
I = I
and
E = xn ( 1 1 .X9 + 8.73 logli) r; Pi lo-' [k~lrn'l
I = I
with R longterm mean annual erosivity [Nlh j
E kinetic energy (kJIm21
I30
maximum storm intensity during 30 n~i n [mmlh]
I,
intensity for storm interval i [mmlh]
fot- 0.05 < I < 76.2 mm; for I > 76.2 rnm I = 76.2 mm
P,
rainfall volume during interval i [mm]
11 number of storrn intervals with equal intensity [- 1
m numbel- of erosive \toi-111s pcr year 1-1
The R factor of Wischrneier & Stnith ( 1958) has proven appropriate
for temperate areas. For tropical Africa, however, several constraints are to
be faced. The calculation of reliable R factors depends on daily rainfall
records over 22 year periods (Wischmeier & Smith, 1978). The necessal-y
subdivision of individual storms into intervals of similar intensity and the
I-ecognition of the rnaxitnum 30 min intensity asks for self-recording
raingages with low paper feed rates. These data are nor~nally not available fot-
a sufficient number of years and meteorological stations. The R factor
overestimates large storms which causc only little runoff but underesti~nates
small storms with much runoff (Foster et al., 1982; Laflen et al., 1985). Both
occur- frequently on tropical soils. Therefore, other authors proposed a
number of different erosivity indices for tropical areas, which were either
easier to calculate or which can be better applied to the local conditions.
Fournier ( 1962) developed an index for river basins in West Africa:
where P,,,,, is the annual amount of rainfall and Pm,,,,, the rainfall amount
during the wettest month. A regression of a modified version of Fournier's
index with the R factor was used by FA0 for the design of an iso-erodent
map of Africa north of the equator and the Middle East (Arnoldus, 1978).
For southern Africa, Hudson ( 1 986) reported that only intensities above 1
inchlh (25.4 mmdh) caused significant splash. Therefore, his index KE>I
considers only the energy of rain falling at intensities > 25.411im. For the
calculation of kinetic energy he used:
with 1 storm intensity [mmlh]
The energy term as calculated by Kowal & Kassam ( 1 977)
with Pi storm volume [ mm 1
described soil loss better than EIjo (Salako et al., 199 1 ).
Delwaulle (1973) si~nplified the calculation of erosivity by
substituting rainfall energy by rainfall amount ( Pi ) and Lal (197hb)
additionally used shorter intervals for thc maxiinum intensity (I,,,;,,):
Chapter 4
For I,,,, he chose the maximum 7.5 min intensity. Sabel-Koschella ( 1988)
obtained similar results for m values between 5 and 25 min.
Roose ( 1977) evaluated mainly 13 stations in West Africa and found
a linear I-egression between the E130i and monsoon type rainfall (Pi) between
June and September of:
and a curvi-linear regression for high intensity storms during the rest of the
year. As an empirical approach for the estimation of erosivity in West Africa
he proposed:
R = (0.85 (+/-) 0.05) :;: Pann [N/h]
(10)
Roose (1977) verified his regression for 20 rainfall stations and
drew an iso-erodent map of West Africa. Further iw-erodent maps were
cornpiled for Zimbabwe, Kenya, Tanzania and Uganda based on KE > I
(Moore, 1979; Stocking & Elwell, 1976). An iso-erodent map for Zambia
wa5 supplied by Lenvain et al. (19x8) using:
with Pm mean monthly rainfall [cml
b mean number of days with rains > lmm [-1
c mean maximum daily rainfall per month [cm]
For the iso-erodent map of South Africa (Smithen & Schulze. 1982)
erosivity was estimated by 'effective rainedll', a modified Fournier's Index
and a 'burst factor'.
4.2 Soil PI-operties
4.2 Soil properties
The influence of soil properties on soil loss can be ideally studied on runoff
plots stripped from all vegetation for some years. Thus, it is assured that no
influences of the former vegetation bias the results. Table 42- 1 demonstrate?
the influence of soil properties on barefallow soils subject to 1200 mm/a.
Soil losses are as low as < It/ha on an Oxisol and as high as 280 t/ha on an
Andisol (Nill, 1993).
Tclhle 42-1: Soil lo.v.\ on diferetzr hczrqfnllow .soils c-orrt.c.tt.c/ to an L I I I ~ I U Q /
erosivity of' 800 N/lz ( ( q?pr~x. 1200 r?znz/n).
However, the soil properties causing these differences are not
evident as mostly a range of soil properties found in different soils and their
combination are responsible. The important soil properties decisive for the
extent of erosion are listed in Table 42-2.
soil type (US soil taxonmy)
Acrustox
Tropudult
Trophum~~lt
Hapludult
Kandiudalf
Paleustalf
Andisol
Mineral composition, especially the content of metal oxides, is
known to influence soil erodibility. Metal oxides act as binding agents
between soil particles, thus increasing structural stability. Soil loss on
subsoils decreases with increasing content of Al- and Fe-oxides (Roth et a].,
1974; Romkens et al., 1977). It is supposed that especially the amorphous
part of the Fe-oxides is reactive. In experiments of Chauvel et al. (1976)
kaolinitic clay mixed with > 5% iron oxides showed a self structuring
behaviour (formation of shrinkage cracks) when drying out whereas at iron
oxidc contents < 5% it formed a coherent matrix. Only 3% of the total Fe-
oxides were actively participating in the aggregation process. Rapidly sealing
mean annual sail loss[t/ha]
0.5
12
20
5 7
8 9
147
280
Soil lo\a was calculated Srorn soil el-odibility values which were adjusted to 800 N/h mean
annual erosivity (approx. 1200 mr nf a) .
Chapter 4
soils generally suffer higher soil losses than non-sealing soils. The type of
clay mineral also influences the formation of seals and the infiltration
capacity of the soils. Soils rich in smectitic clay (e.g. Vertisols) swell and
shrink with varying moisture content. Infiltration is, therefore, high in the dry
state while cracks are open. In the moist state these soils become extremely
sticky and plastic, cracks are closed and infiltration reaches very small
values. Soils rich in kaolinitic-oxidic clay, on the contrary, are well
aggregated in the dry and moist state. They are less susceptible to sealing
than soils with 2: 1 clays (Levy & van der Watt, 1988; Shainberg et al., 199 1 ).
The stable structure of the former enables high infiltration rates.
Table 42-2: Soil properties irzfluerzcirzg soil erosiorz.
soil properties
permanent
mineralogy
Fe-, Al-oxides
texture
soil organic matter
pH and exchangeable cations
aggregate stability and size
bulk density
electric conductivity of soil water
soil temperature
v antecedent rnoi\tur-e
variable
Type and quality of the parent rock act on the texture of the formed
soil. For example, sandy soils form from granite whereas clayey soils form
from basalt. Soils high in silt and low in clay and sand are highly erodible.
Erodibility decreases with ;i decrease in silt, regardless whether the
corresponding increase is in the sand or the clay fraction (Wischmeier &
Manncring, 1969). The high erodibility of silty soils is explained by their
weak structural stability. They rapidly form surfice seals upon raindrop
impact. Erosion is less on clayey soils due to their better aggregation and on
sandy soils due to their- non-sealing surface. Fine sand (0.05-0.1 Inn]
diameter), however, behaves like silt and is therefore attributed to the silt
fraction for soil erosion aspects (Wischmeier 8r Smith, 3978).
4.2 Soil propertie\
Soil organic matter (SOM) influences soil loss by improving soil
structure. root penetration, water capacity and infiltration. With increasing
SOM, el-edibility decreases (Wischmeier & Smith, 1978). SOM consi\t\ of
very heterogeniou\ particle\ ranging in size between several Inn1 down to
< 0.002 mln. Chelnically very reactive organic molecule\ compare with more
i nact i ~e one' \ and re\i\tant component\ with rapid decompo\ing one' \. The
role of SOM as a binding agent i \ more important on \oil\ deficient of other
structuring coriiponents. Thei-ef'ore, the importance of SOM decreases with
increasing clay content (Wischmeier & Mannering, 1969). Valentin 24 Janc~tu
( 1989) found that structural stability was only improved by organic matter if
the ratio of organic matter to clay was 2 0.07. In tropical cr-opping systems
SOM is high after the fallow ancl declines I-apidly during the cl-opping per-ioct.
Tlii~s. erodibility changes during ;I cropping cycle from low values during the
trtllow anti at (he t)eginlling of ci~ltiv;~tiori to higher- values towards the end of'
culti~,ation. In own trials the erodibility during the Sirst year of bar-cfallow
after bush and I'orcst filllows was only 40 CX ant1 80 (2 ;:. respectively. of the
final erodibility which was rcaclicd al'ter about 3 years of bar-efallow (Nill.
1903).
Agy-cgate \17e and \lability have a pc~-~iianent and a Larlable
coliij7onent, the latter of which retlect4. alllong other influences. vegctatioli
allcl ~nanagcliiont. Eroclibrl lty dccrca\cs it11 lncl-ca\ing aggregate stitb~lr ty as
\cal formation i \ clelayed and ~nl' iltrat~on incrca\ed. Ho ~c v c r . the effect ol
aggregate \ i l c i \ le\\ clear. Mo\tly \oil lo\\ wa\ found to becorne \ln;~lle~-
L+ ~ t h incre:~\ing aggl-egatc (Ehwue. 199 I ; Falayi KL LaI, 1979) for.
, I Y Y I - ~ ~ , I ~ ~ C L cliumctcrs hetwccrl 0.5 and 50 mm. 1,uh ( 1083) tcstcd aggrcgatc
cla\\c\ betueen 0. 5 and 30 mm and tounci higher splash and sheet cros~on
I'rorli larger than frolii smaller agg~-cgate\. Wi\chmeier Kc Smith ( 1978) al\o
attributed Il~pliet erodibilitte5 to larger aggregate\. Howevcl-, Aniba\\:i-Klhl
& La1 ( 1093,) only Sound a \oil lo\\ decl-ea\c up to 10 rnm aggl-egatc
diameter-. For :~ggrcgate\ between 10 and I00 rnm 110 effect wa5 rnea\u~-cd.
The el'l'ect ot the cxcliange:tble cat ~on\ i \ e\pecially Important on
Ic\\ \beathered \o11\ ol' t l ~c {emi-arid to arid tropics. Tllc$e \oil\ are wtxthlq
\trircli~~-ed cli~e to low SUM ;uid oxide co~ltents and have often \ancly 10 loamy
~c\tul.e. Na. a\ a monobalent c ~~t i on, ha\ a pronounced d~\ pcr\ i ng influence
o n \oil \t~.ucturc. 3 to 5(/( 01' Na o n the exchange complex arc enough to
cl~\per.\c the \oil (Shainberg. 1985) ant1 crodibility i ncr cam with irlc~.ea\iny
Na content (S~ngcl- et al, 1980). Mg \aturated \oil\ wcr-c Ihund to be mor-e
el-od~hle than C'a sati1i.atc.d soils cau\cci by the lal-gel. hydl-atton shell 01' the
Chapter 4
Mg ion which weakens bonds between soil particles. The stronger
aggregation in the presence of exchangeable aluniiniurn explains the higher
stability of acid soils. The electrolyte composition of the soil solution also
exerts an influence on soil loss through tlocculationldispersion effects. For
example, saline soils rapidly disperse after dilution of the soil solution at the
on-set of rain.
On previously rnoist soil runoff starts earlier and reaches higher
runoff volurnes than on initially dry soil. For this reason, rains occurring at
the on-set of the rainy season generally cause less runoff and soil loss than
rains at the end of the rainy season. Not much data are available about the
influence of soil and water temperature on soil loss. With increasing
temperature water viscosity decreases. Aggregate destruction caused by the
pressure of encapsulated air during I-apid wetting of the aggregates is
enhanced if the wetting velocity is higher. Water rnulch by less viscose water
(= ,,more liquid") will be less protective against raindrop impact. Auerswald
( 1992) explained a soil loss difference of 17 % between artificial rain applied
in the morning and in the afternoon with a temperature difference of 8 "C.
4.3 Topography
Topography intluences soil loss by the length, gradient and shape of a slope.
Soil loss increases very sensitively with gradient and commences already on
slopes < 1 9%. Mutchler & Greer (1980) measured losses up to 5 tlha from dry
<oil and up to 1 1 tlha from wet soil on a 0.2% slope when a simulated 60 min
storrn was applied. In Senegal, annual losses from groundnut fields on a 1 C/ r 1
slope reached 15 tlha (Fournier, 1967).
Uncertainties arise, however, where the influence of gradient has to
bc quantified. Most studies propose an equation for soil loss tiom interrill
areas of the form
with A soil loss
S gradient
a, b constants
4.3 Topography
Values for b between 1.35 to 2 were suggested (Hudson, 1986;
Hudson & Jackson, 1959; Musgrave, 1947; Zingg, 1940). A value of b = 0.67
was suggested for soil 105s from rills (van Liew & Saxton, 1983). In the
USLE the influence of gradient is described by
S = (65.41 " sin2 u + 4.56 * sin u + 0.065 1-1
with u slope gradient [degrees]
More recent analysis of slopelsoil loss data revealed a change of the
relationship at > 9 % slope (McCool et al., 1987). Soil loss for very low
slopes was found to be overestimated by the LS factor (Murphee & Mutchler,
198 1 ). Runoff on low slopes flows slowly and quickly forms a water layer
deep enough to act as surface mulch. It further became apparent that soil loss
depends on the ratio of rill to interrill erosion. Soil loss is higher on soils very
susceptible to rilling (McCool et al., 1989) and the potential for rilling is
greater on steep slopes (Mutchler & Greer, 1980). S factors for the Revised
Universal Soil Loss Equation (RUSLE) are, therefore, calculated by:
with u slope gradient [%]
Equation 14 and 15 are used for slopes > 4.6 m long and gradients
of < 9 5% and > 9%1, respectively. On slopes < 4.6 m long rill erosion is
negligible on most soils and equation 16 is to be used.
Increasing slope length enhances soil loss as more runoff can
accu~nulate on long slopes. For slope length, the following term is used
(Wischmeier & Smith, 1978):
with 1 slope length [m]
rn slope length exponent I-]
The product L S is called the topography or LS factor. The LS
factor wa5 derived from $011 lo\\ data of \lopes ranging from 3 to 18 '/r and 9
to 90m (30 to 300ft) long (W~whrneier & Smith, 1978). Beyond the\e range\
n o mea\urement\ were taken. However, the equation wa\ regarded applicable
by the author\ to \lope\ 300 m long and 50% steep (cf. Chapter 7.3). Foster
et al. (1982) est~rnated that the 1,s factor can be appl~ed in the trop~c\ to
slopes up to 25 C/c whereas Hurni ( 1980) u\ed the LS factor tor \lope\ > 50 C/c .
The LS Eactor wa\ verified In West Africa on \lopes between 3.5 and 23.3 (/c
(Rome & Sarrailh, 1989) wherea\ Shcng (1990) reported an overe\t~matlon
of \oil lo\\ by the LS factor on 30 54 slope\.
The effect of slope length on soil loss is interrelated with slope
steepness. This is expressed i n the slope length exponent 111 of the LS factor
which is 0.5 for slopes > 5 ' X and decreases to m = 0.15 for s l opes 2 0.5 54
(uf. Chapter 7.3). In the earlier development of the ecluation, an exponent of
1.6 was used (Zingg, 1940). Dangler & El-Swaify (1976) reportod an
underestimation of soil loss by the L ['actor as used by the USLE. However.
on soils from West Africa Lo,? was found to give better results (Rooso 24
Sarrailh. 1989). In the RUSLE, m varies between 0.02 and 0.83 depending o n
rhe soils susceptibility to silting (McCool et al.. 1992).
Soil lo\\ i \ al\o ~nflucnced by the (hape of a \lope. It dccrea\e\ In
the orcler convex > regular > concave \lope l'or~ii. On a convex \lope. whcrc
the gradlent increa\e\ in the order up-slope < mid-\lope < down-(lope. a
large runoff vol u~i ~e coincide\ with the maximum gradient (down-\lope). 0 1 1
the contrary. on a concaLe \lope the maxi m~~~i i gradient I \ up-\lope uhcrc
runol'f I \ \ t i l l \~rialler.
4.4 Cover, tillage and protection techniques
Cover. tillage and protection techniques depend on management. i n contrast
to rain erosivity, soil erodibility and slope. This r~litkcs them of fol-emost
irnportat~cc to soil conservation. Managcmcnt pl-actices can be dictinguishcd
according to the basic erosion processes that they inrlucnce:
I . Methods which reduce the runoff' volume or the sediment transport
capacity
All ~llethod\ which incrua\c water inl'iltration o r reduce runoff vclocit y al \ o
seduce soil lo\$. I f ri11101'1' i \ \lowed down. the wares \ray\ longer on the t'icld
4 4 ( ' o ~ c r . t l l l i ~ge and protcctlon t c c h n ~ q i ~ c \
and gc15 tnol-e time to infiltrate. Additionally, the water layer on the \oil
\urf'ace becomes deeper and protect\ the \oil from raindrop impact (cf.
Chapter 3).
These methods include:
D reduced tillage
D no-tillage
Ci tillage and planting acres\ the slope (contouring)
C> soil cover by inorganic or organic mulch
C> surface fhl-nning practice\ (ridging, tied-ridging, bedding)
11. Methods which reduce the slope length
Thereby, the ~-unoff procluc~ng up-\lope area and I-unoff volume are
I-educccl. So11 tnaq \till be tr-an\pol-(ed bill phy\ical ob\tacle\ divcrt r-unol'f
andlor cauw depo\ition. Thi\ group includes:
P hillside-ditches
[> filter-strips with grasses, hedges or tree rows
D earthen and stone bunds
I> terraces
The single methods mentioned above are described and discussed in the
following.
4.4.1 Reduced tillage and notillage
Tillage breaks down soil ugpt.cyiites. disturbs soil structurc, pore continuity.
and biological activity and produces transportable soil material. Reducing
tillage intensity and frequency increases the number of continuous pores.
maintains :iggrepation and reduces organic tnatter decomposition. Thereby.
binding agents between soil pal-ticles like fine roots, fungal hyph:~e. root and
bacterial exudates are conserved. Thc soil stays more consolidated as
colnparcd to tilletl soil. Thus, infiltratio~l and resistance against impacting
drops and shearing forces of rhe water are higher than on tilled soil.
Chapter 4
4.4.2 Contouring
Contouring necessitates that all tillage and planting operations are carried out
across the slope. These operations produce a low surface relief across the
slope. Runoff is slowed down and the surface storage is increased. With
increasing gradient, the surface storage capacity decreases (cf. Figure 3-7)
and the risk of spilling over with consequent rill formation increases.
Therefore, the efficiency of contouring reaches a maximum on slopes
between 3 to 8% (Wischmeier & Smith, 1978).
However, the effect of contouring is uncertain in handtillage
systems where the soil is tilled from the bottom of a field moving up-slope
and were a general down-slope movement of the soil from hoeing can be
observed. Thus, tillage in such systems is not on contour in the strict sense.
Only contour planting can be achieved.
4.4.3 Soil cover by organic mulch
Surface cover is one of the most efficient measures for soil loss reduction.
Organic material is easily available i n areas with sufficient rain. Residues of
the previous crop, weeds and additional mulch material from outside the l'icld
can be used (leaves; twigs fro111 bushes, hedges; straw; wood cuttings;
organic household waste like peelings, shells and husks). Left at the soil
surface, they protect the soil against the pounding drops and prevent seal
f'orniation. The stalks and leaves form barriers where the water ponds
(= water mulch). Runoff is slowed down due to twisted pathways.
Additionally, residues and mulch reduce variations i n soil ternperatul-e,
humidity and thereby biological activity and improve structure and
infiltration. Earthworms which move to the soil surface in order to pick up
food create large continuous pores. Mulch efficiency depends o n the surface
area covered by the material (cf. Figirre 74-2).
Incorporation of the residues diminishes the coverage. The deeper the
residues are incorporated into the soil, the less protection they offer against
erosion. The efficiency of surface mulch may also be reduced on soils with
extre~nely unstable structure ( e. g soils with high sodium saturation , or hard
setting soils) where aggregates already disperse on moistening . On such soils
superficial incorporation may provide higher infiltration rates because the
A t e t can be can-ied OLII hy \ubrnel-ging dry aggregate4 o r fragrncntx of 1-2cm diameter in
water. Unstable aggregates break down immediately
3.4 Cover. tillage and protection tcchniqi~cs
decomposition products of the residues stabilize soil aggregates and the
residues act as stabilizing framework.
4.4.4 Inorganic mulch
Several forms of plastic foils are used in intensive agriculture and gardening
to protect the soil, reduce evaporation and suppress weed growth. Howevcr,
these mulches play a marginal role in s~nal l scale agriculture. An important
natural mulch tnaterial are stones and gravel of various sizes. Stone
pavements and surficial gravel concentrations on soil surfaces frequently
indicate erosion processes. The gravel was enriched at the surface by the
selective removal of the soil. With increasing cover of the surface. the soil
i11lde1-neath becomes protected. Howevcr, the active use of gravel as rnulch
~natel-ial deserves much more attention than rccently given. In highly leacheti
soils of the hi1111id tropics, the use of gravel from basic rocks may
additionally deliver sorne nutrients.
4.4.5 Surface forming practices (contour ridging, heaping, tied
ridging, bedding etc.)
The\e method\ create physical ob\tacle\ (Figure 445- 1 ) which reduce
e\pecially the slope length. Runoff is \lowed down. \topped or deviated
\idew:ly$ on a reduced gradient. Alike contouring. the protective effect of
thew methods depends on the gradient of the \lope. the \ide \lope and he~ght
ol' the ob\tacle\ and their distance frorn onc another. Contour ridges are \mall
earthen dam\ of about I0 to 30~117 height placed across a \lope.
bed
The protection by ridges is low on very low slopes hecause soil loss
is genet-ally low. On stccp slopcs. i t is low because the amount of' water
which can be retained by a ridge decreases with increasing gradient. A 15 urn
high ridge does not store any water on slopes >25 C/ c (Foster et al. 1992). At
the same time the risk ol' spilling over- and ol' break throughs in the ridges is
enhanced. Thus, maximum efficiency is obtained on medium slopes (cl'.
Figure 75 1 - 1 ). The eff'iciency of' ridges also depends on ridge height and the
side slope. The higher the ridge, the more water can be stored. The lower the
sidc slope 01' the I-idgc, the slower the runoi'f. Meyer- & Harrii(>n (1985)
\howeci that on side slopes < 0.5 C/c the suspended sedinlent in the runoff is
cleposited and most of' the setiiment originates from the ridge-sides. Above
254 sidc slope deposition ceases and the sediment is moved out of the field.
With sick slopes of' 5-6 % rilling of'thc I'LI~I-ows commenced.
The ci'lkct of' l.iin-ows also depends on stol-m size. Large storms may
\uspass the carrying capacity of' the I'LII-rows and cause overflowing of the
I-idge top. Overflowing Inay cause very severe damages and should be
a\,oided i n any case. Thus. efficiency is less for areas with frequent lar-ge
storms. The 10 year storm which is the largest. I-egularly occurring sto1.m
within 10 year-s. can be i ~scd to calculate the efficiency of contour ridging f or
an area.
A\ the ridge-sides act as runoff' producing area for the furrows. the
ri~nol't' ~xodi ~c i ng area increase\ with the length of the furrow\ and so doe\ the
ri1t101'1 volutne. In order to avoid overflowing of' the ridge top\ 01- rilling.
f'i~rrow length {hould be limited.
Not muc11 i \ hnown about the effect ol' ridges placed along the
slope. Mea\urement\ on \lopes of 7 t o 13 Cj c indicate an ero5ion enhancing
i nfl ~~eri ce ( P f.ictor\" between 1 and 6). On \lope\ o f 13 to 20 fZ the ncgativc
effect was Ic\+ pronounced ( P = 0.31 to 3.4) but \till important (Reining.
1091 ).
Heaps of' v;~ryirrg sizes are frequently used especially for tubers but
also for other crops like groundnut or bambara nut. The loose. rich topsoil
usecl for the heaps is I'avourable hi- tuber fon?iation. Watcr logging is
~ w c ~ i t c d by the heaps and mincrali7,ation is enhanced. However. not moch
is known about the effect of heaping on soil erosion. Unfortunately. sizc and
LII-rangenlent of heaps on a slope arc inostly nut described in literature.
A P Factor < I indicates less erosion compared to a harcfallow I'ield wherca\ P > I 1nclic.aL~4
n o ~wotection. Morc int'ormation i 4 given in (' l l apt ~r 7.5
Heaps on a slope A -r C2 enlarge the average gradient a by the
gradient U-cx on7the sidewalls ( Fi g~~r e 445-2). Taking the ri l axi ~n~~rn angle of
about 40" (a+B) into account which forms if topsoil is poured on a heap. the
actual gradient of the slope is changed on the heaps into a gradient of about
80 C/r (= 40' ) on the sidewalls. This also implies that the heaps become lower
with increasing slope as demonstrated by heap height (h l+h2) for heap
ABC I on level ground compared to h2 for the heap ABC2 in Figure 445-1.
Runoff from the heaps enters into the furrows aniong the hcaps and mo\,es
downward in a concentrated flow. Runoff volume depends on the size and
arrangements of the heaps. Compared to small heaps. largc heaps have a
larger runoff producing sidewall area and less drainage paths between the
heaps. If the heaps form furrows along the slope, rapid water movement
results. If they are arranged in quintuples, the water has t o flow around the
heaps and a reduction of the flow velocity can be assumed. If the heaps are
arranged up and down-slope, runoff is directed straight down-slope and will
reach a higher speed.
' The m~i xi mum alopc anglc fol. I~capcJ LIP aoil (angle of I-cpocc depend\ on the particle \ i / c
di\tribution. Surface \oil < 3 0 mm = 38.7": aggr cgat e~ 1-2 Inrn = 40.2' : <urlacc \oil sietetl
to < ?mnr = 37.6" ( i 2i i cl - ~ alcl. 1903 ).
4. 3 Cover. t l l l af e and protectton (echn~yucs
The influence of heaps on soil loss further depends on surface soil
depth. On soils with a thin surface horizon, all soil is scraped together for the
heaps. The underlying soil which can have very different properties is
exposed. A pattern of very different soil erodibilities is created this way
which can include, for example, a less erodible surface soil on the very steep
side-walls of the heaps and an erodible subsoil between the heaps. Overall
soil loss should be notably increased in this case.
4.4.6 Bufferstrips
Bufferstrips (filterstrips) are < I ni to several m large strips of' planted
grasses, hedges or natural vegetation on contour. They slow runoff down and
maintain higher infiltration rates within the strips as co~npared to the adjacent
field. If runoff occurs, soil is transported within the cropped area and
deposited in the vegetated buf'ferstrip. The runoff either infiltrates completely
in a bufferstrip or crosses the strip. If all runoff infiltrates, all transportecl
sediment is deposited. If part of the runoff passes through the strip only a part
of the sedirnent is deposited. First, the coarser, heavier sand particles and
aggregates settle whereas the srnaII particles of clay and organic matter use
furtl~er transportecl and may leave the strip on its lower side. Thus.
quantitatively a large part of' the sediment can be retained by a bufferstrip
while an irilportant amount of' fertile soil is still lost. Cornpared to temperate
soils. this is more relevant on leached tropical soils. Their low cation
exchange capacity (CEC) is largely associatctl with the organic matter.
Another- inconvenience of strips with incoriiplete infiltration is that the water
which leaves on the lower side can speed up again and pick up new sedirncnl.
1 to 4 n~ large buffer\trip\ on 4-20 C?c slope\ can reduce soil lo\\ by
10 to 00%. Strip\ w~t h natural fallow vegctatioi~ can already be \pared out
when cultivating the field. Compared to i ~ ~ c h 4tr-ip4, planted $trip$ have a
lower efficiency in the I st yeas. Efficiency decline< after an opt~murn due to
increasing \ed~rnentation in the strip\. A 40 rn wide bufferctrip for example
dropped from 99 to 75 cji, el'f~ciency during 9 month\ (Barfield & Albrecht.
1982). In agreement with other authors. Sct i a~~der & Auer~willd ( N9 2 ) could
\how that a 30m wide grass stw on a 8 54 slope retained 64(L of the
sediment which entered the strip (ca. It/m strip width). The efficiency
increaced with incl-ea\ing \t~-ip width (cf. Figure 752- 1 ).
Bufferstrips are Inore acceptable to fanner\ if they give solne yield.
Introduction of fruit tree\ or woody specie$ may be of more interest than PLISC
gra\\ strips and can encourage farmer\ t o protect the strips :tgain\t t ~r c j .
Some conimon g1.21\\ and tree species u\ed for bufSer\tripc and biological
control are li\ted in Table 42- 1 Annex. Ari extensive databanh on sutted
'r
woody s pec~e\ is ~i\ailablc from ICKAFJ Nairobi .
4.4.7 Contour bunds (stone hunds, earthen hunds, diguettes)
Buncl\ ai-e a form of Iiigli ridges which arc mounted at a distance o f \evcral m
from one anotlicl- (Figure 447- 1 ). They can be constructed fi-om \oil or \tone\
or both co~nbi ned (earthen core wltli stone\ o n the outside) (Figure 447-2).
Bund5 are il\ed t o control ero\ion and to conserve water. Impermeable bund\
(I'rom earth or with ii1I earthen core) stop runoff completely and direct it
\idcways. They are ~i i orc rigid in tlicir action and are to be con\tructed more
\olicl than pcrriieable bund\. Waterlogging in front of' impermeable bunds can
be a problem for 4ensitlvc crop\.
Runoff hi t t ~ng a permeable bund (e.g. \tone bund) i\ \lowed down,
\lowly penetrate\ the bund and leave\ partially on its lower side. Reduced
celocity al\o f a ~our 4 infiltration on thc lower \ide. However, the runot't' may
regain velocity and pich up new sediment. Thus, the ~t rca below a bund may
be eroded wlierea\ thc area in front of thc butids is wdiment-enriched.
Therefore. \mall terrace\ fonii after a couple of years.
Bi~ricls ;ise recotiiii~ended o n slope\ of less than 12 % (Table 449- 1 but are
also u\ed on steeper slopes with \ucce\\. In order to diminish maintenance
mot-k. t he bundi ihould be planted to permanent vegetation (gra\\es, woody
\pecie\, tree\ ).
' h1ultipur~x";e Trcr K: Shrub Datahasc (ca. 110.- \IS$) IC'IIAF. P.O. Box 30077
Nail.ohi. Kenya
4.1 Co\ el-. rlllagc ;111tl protectloll ~ c c h r ~ ~ c l ~ ~ c \
permeable bund
semi permeable bund
~mpermeable bund
Dykes are a large version of bunds which is especially used in semi-
arid to arid areas to store water and to slow down torrential floods. They are
the transition to the even larger darns. Alike bunds, dykes are constructed as
permeable (digues filtrantes) or impermeable obstacles (digues dkversantes).
Between the two extremes there are a couple of intermediate solutions with
impermeable lower and permeable upper parts (Figure 447-2). In the first
case, the water is slowed down and ~nomentarily stored while it percolates
through the dyke. In the second case, the water is stopped and stored behind
the dyke. The water quantity exceeding the dykes storage capacity either
flows over the top of the dyke or is conveyed by a weir or spillway. Behind
the dyke, water infiltration is increased and sediment deposited (Figure 447-
3). The deposits are either used for irrigated cultivation in the border zone
while the water is retreating or for a crop which uses the water stored in the
soil. Clayey deposits serve for brick construction.
SHED
4.4.8 Protective ditches
Several types of ditches can be established on contour to slow down runoff
and collect eroded sediment. Drainage ditches are constructed by disposing
the excavated soil down-slope of thc ditch (Figure 448-1 ). The ditches are
laid out on contour or with a slight side-slope of 0.4 to 0.5% (see Hi~dson.
1975for planning principles). Drainage ditches reduce slope length into
several segmenls. The down-slope concentration of runoff is thereby avoided.
The sediment spilled into the ditches nceds regular excavation. Maintenance
efforts are therefore high.
The Fanya Juu terraces are a modit'ied version of dl-ainage ditches especially
used in East Africa (Figure 448-1 ). The excavated soil is disposed up-slope
thereby forming an earthen bund which traps further sediment.
Hillside ditches (Figure 448-2) are a form of reverse slope or level bench
terrace. The bench is generally not used for cultivation but as foot path or road.
tr rrac7e.r
drainage ditches
Riser protected
with grass
Hi l l si de ditch
( f oot path) Tal l grass
.I
Cultivation is carried out on the graded interterrace area along with further
protection measures. Hillside ditchc\ may be used for slope\ of up to 47%
(Sheng, 1989). They divide the slope into \liorter segments and divert runoff
at non-erosive velocity.
A version of the hillside ditch is the broad based terrace used in
mechanized agriculture. It can only be used on gentle slopes. The interterrace
space and the terrace interval on the graded terrace is used for cultivation.
The terrace is kept as low as possible in order to allow the passage with I'arm
equipment. Figure 448-3 shows how these measures are laid out in a
watershed or farm. The runoff collected by the ditches must be disposed
saf'ely by constructed waterways or by conveying i t into densely vegetated
areas.
4.4.9 Terraces
Terrace\ are described by a number of characteristic\. Important features are
the vertical height. the horizontal length, the ratio of the raiser b/a, the revert
\lope a, the side slope. the terrace interval and the interterrace interval
(Figill-e 449- 1 ). They are used for a range of purpo\ec a\:
D to divide a slope into shorter segments
D to reduce the slope angle on the terrace interval
I> to convey the surfilcc runoff to controlled water ways at a non-erosivc
velocity
L> to harvest water from intertcrrace intervals for water conservation
D t o store water for paddy cultivation
b to store sediment eroded l'rom the interterrace interval
A number of different terrace type\ was developed to cope with
these tasks. Most oc them can be described as riiodified bench terrace4
(Figure 440-2 ).
The level bench terrace has a wide-spread use f'or paddy cultivation
whereas the reverse slope ancl outward slope bench terrace are favourablc for
upland crops of the hurnid tropics. The conservation bench terrace is used i n
semi-arid to arid areas for water harvesting. B5nch terraces are used on slopes
between 12 and 50C/r and are built by hand, animal drawn equipment or
machines.
7- i- y b horizontal length j + '
~nterterrace :, terrace -&- Y
interval ~nterval
1 LEVEL BENCt i TERRACES 3 CONSERVATION BENCH TERRACES
Runof f Co n t r ~ b u t ~ n g Area
~ O r ~ q ~ n o l Land Sur f ace
O ~ , Q , ~ O I Land Sur f ace
'---qv~ .. , . ee$ D Y k e .-\, . q,+ ~ e v e l Bench
- .
.
.
.
2 OUTWARD SLOPING TERRACES
4 REVERSE SLOPING TERRACES
Original Land Surf ace
Toe Or o .
Toe Or o ~ n
4.4 Cover. t~llage and pl-otectlon techn~cliie~
The intermittent terrace is used if terracing is not completely carried
out for the entire slope (Figure 449-3). The design is carried out as for bench
terraces but only every 3rd terrace is constructed. This leaves the option to
later construct further terraces in between which gradually transforms the
intermittent terraces into bench terraces.
Orchard terraces are used for tree plantations 011 very steep slopes in
order to facilitate access and maintenance (Figure 449-4). The terrace interval
is not planted to trees but stabilized by grasses. The distance between the
orchard terraces is determined by the spacing of the trees which are planted in
the interterrace interval. In combination with orchard terraces individual
basins can be used to plant the trees in the interterrace interval (Figure449-4).
The individual basins prevent erosion and loss of fertilizer and herbicides.
They conserve water especially if mulched.
The area between the individual basins is vegetated.
Chapter 4
Some characteristics and applications s u~nma r i ~e d by Sheng ( 1989) are listed
in Table 449- 1 .
terrace type width of terrace interval reverse dope land slope
[mJ L%l 1941
bench terraces (hand made) 2.5-5.0 5 12- 47
intermittent terraces 2.5-5.0 5 13- 17
natural terraces
(caused by bunds) 8 -20 < 13
orchard terraces 1.8 I0 37 -58
individual basins 1 5 ~- ouncl 10 < 58
hilhide ditches 1.8 2 0 I0 < 17
4.4 Cover. tillage and protection tcchniclue:,
The length of the terraces is generally < 100 m and a side slope of
l CTc is proposed. The terrace interval depends on slope and soil depth. The
gentler the slope and the deeper the soil, the larger the terrace interval. The
slope of the raiser is built with a ratio of 0.75 : I .
Indicators for soil erosion
Erosion leaves finger prints which also give information on the type and
intensity of the processes. Some of these finger prints are dramatic and hardly
to be overlooked while others are less distinct and hidden. Such parameters
and finger prints can be studied and provide useful information i n a first
survey on the general erosion risk.
The age of a landscape indicates its erosion susceptibility. Old
landscapes are characterized by gentle slopes, plateaus and plains whereas
young landscapes show a rugged relief with steep slopes and deeply incised
valleys (Roose, 1975) resulting in higher erosion potential. Long periods of
'normal' or 'geological' erosion cause a lowering of the landscape, the final
form of which is a peneplain. A peneplain is characterized by a low and
gently undulating relief (Figure 5- 1 ). However, often the process is
J
interrupted by tectonic upheaval or tilting of a landscape. The base level is
lowered and a new erosion cycle starts. Tectonic movements and several
erosion cycles create a landscape of plains at different altitudes ( pediplains )
which are separated by distinct scarps (Figure 5- 1 ). The oldest surface
corresponds lo the highest surface. Remnants of the older surface were
separated from the faster eroding pediplains and fhrrn isolated steep hills
(inselbergs) or plateaus on the pediplain below which occur frequently i n the
savannah areas of West and East Africa (Thornbury, 1985). These remnants
were maintained because they were resistant to erosion. An examplc ~I-OITI
Camcroon shows how pediplanation has created four levels during > 60
million years (Figure 5- 2) (Segalen, 1967).
Gully erosion leave\ very striking features i n the landscape and
destroys agricultural land. In Lesotho, for example. 11 is estimated that 3% ot
the arable area is occupied by gullies (Wenner. 1989). Gullies vary greatly i n
slze. They are defined us deep enough in order that crossing is impo5\ible
with agricultural machines while rills can still be closed by ordin;~ry tillage
111ethods (Hudson, 1986).
lowcct point of thc landwape to which thc water c;un Ilo\v
58
5 Indicators lor soil erosion
Figlo-e 5-1: Corzc,e/'t~ of rr ~ P I I C I ) / L I ~ I I ~l'itll ,yor~tIe, ~4rld~llr1tirlg relief' ( u / >o ~ v )
r111rI ( 1 /?e(iipl~lirl ~' i t l l LI di ~f i 11~' t S C . L I ~ ) hct~lrerl t\tlo lc.~~el\
(c!ftet- T / I ~ ~ I Z ~ ~ L ~ Y - \ - , / 98.5)
Chapter 5
k'iglrro 5-2: Podiylnir~s crt citflkl-erzt n1titudtl.s I(.\ fhl-rllecl irz Cnnzc~r-ool~ h?' ~llor-tj
tlllrrl 60 r~zilliol~ !,etrl-,\ c!f'rr-o,siol~ (after- Scgnlc7t1, 1967)
Altitude
>1500m
900-1500
600 - 900
< 600
x
x
X i.
o Maroua 2
++
++
* x x X X , ' ~ x ~ ~ ) ( * x :+*
AFRICAN REPUBLIC
Large gullies reach several tens of meters deep and wide and several
kilometers long. Gully incision starts were large runoff volumes are
concentrated into linear llow. Possible sites are runoff convergence points of
several fields or spillways from roads (Moeyersons, 1989). The water from
thc inipcrmeablc road surface collects i n the road ditch and, instead of
entci-ing in ii~tervals into a reinforced evacuation ditch, is often led into the
aclsj:iccnt area where i t triggers gully forination. Lowering of the base level or
large storms which coincide with a sparsely vegetated soil can also initiate
gully formation (Oostwoud Wijdencs & Bryan, 1 99 1 ).
Gully l'ormation is facilitated on soils with a coarse textured surface
soil underlain by clay-rich subsoils (Lal, 1992). Concentrated lateral
subsur-face tlow creates subsurface pipes which in turn can start gully erosion
(Firth & Whitlow, 1991). Once a gully is initiated, it is enlarged by regressive
gully head cutting along with undercutting and collapse of the side-walls. The
gully head moves more and rnorc up-slope and secondary gully branches
forrn. Soil cracks form parallel to the gully side-walls. Surface water enters
the cracks which increases pore water pressure and decreases soil coherence
thus destabilizing the side-walls. It was demonstrated on sodic soils that gully
Chapter- 5
head advance wa5 determined by rainfiill, antecedent \oil moi\ture, headcut
height (plunge-pool effect) and runoff contributing area (Stocking. 1981 ).
Gully development depends on the depth of the weathered layer and the
water\hed area. Gully inci\ion stops if the wlid rock ~lnderneath is reached.
Vegetation which forms during less erosive years can also \top further gully
enlargement. However, thi5 may only be temporarily. Heavy \term\ or man-
made damage to the protective vegetation can reactivate the gully. I f the
runoff producing area becomes \mailer with progressive head cut regre\sion,
a mature stage o f the gully is reached (Figure 5- 3) (Hoebl~ch, 1992). Gully
reclamation is laboriou\ and costly. It is more eff~cient to avoid concentrated
flow than to protect the \oils against its damaging effect.
Landslides are another form of easy recognizable down-slope soil
transport which causes disasters. 2 1.000 people were killed in 1970 in Peru.
when an earthquake st;irted the mover-nent of 25 million m3 of earth which
destroyed two entire towns (Schustcr, 1978). L,andslides occur if the weight
of u sloping soil rnass cxceeds the shear strength of the soil. Cracks occur on
the upper side of the soil mass and the soil slumps down-slope along a
weakness plane. Such weakness planes within a soil or geologic substrate c;ui
be duc to different layers of the soil (e.g. permeable layers over less
pcnneable layers) or natul-a1 layering of the geologic substrate (e.g. schist).
I~nbalances are caused either by increasing the soil weight on the slope (c.,g.
construction. water saturation) or the instability of the weakness planes
(undercutting by roadcuts, water saturation). Landslides are classed according
to material (soil, stone). humidity (e.g. ~i ~i ~df l ows ) , the type and direction of
the movement and its velocity ( c. g. cr-cep, flow). An example of a
translational and a block slide is given in Figure 5-4.
Majot determ~nant\ tor land\licles are slope, climate, geologj,
layering and hydrologic properties, seismic activity, vegetation and
human activities (Ga\\er & Zobl\ch, 1988). Mocyer\on\ (1989) reported
that landslidec in Rwanda occur especially on elope\ > 58C/( and 011 5chl\t
whereas C ~ I I sand ctonc and quartzilic rocks no \ l ~de\ were ob\erved. Slldc\ on
\lope\ < 58% oonly occurred 11. road con\truct~on cauced \lope in\tabll~ty.
Landslide\ were more frequent on \lope\ > 62% in Uganda (Temple & Rapp.
1972) and on \lopes > 53 C/ c in New Zealand (O'Laughlin. 198 1 ).
5 I ndi ca~or s for boil ero\ion
Figlrrv 5-4: Lrft: bloc-kslicle; I-igllt: tr-crr~.\ I~rtioilcrl 5lidc (ir7: Grr.\,\c>t- & Ziihi\c.lz
( 1988); ~( f t c r Gri ggs & Gilc81zhir.c.t ( 1 977) NI ZC/ SL' / ICIL~PI- ( 197-5))
Landslide frequency is determined by a rainfill duration-intensity
threshold which varies due to geology and climate (Figure 5- 5) (Larsen &
Simon, 1993). Long duration, low intensity rains cause deeper landslides on
volcano-clastic material in Puerto Rico whereas short and intense storms
causc shallow slumps (Larsen 8( Simon, 1993). Vegetation decreases
landslide frecluc~icy. Plant roots increase the shear strength of the soil. A
perm"nial vegetation consu~i ~es an iniportant part o f the rain as interception
water and f'or transpir:ition. Thus, a vegetarcd soil is drier than an ~~nvegetated
soil. These influences of vegetation outrule the physical weight of the
vegetation and the weight increase of' the vegetated soil due to incl-easctl
infiltration.
An inclication for soil creep is the 'sabre growth' o f trecs. Fl'hc
slowly down-slope moving soil inclines the trces which in turn redirect thcir
growth upwiu-ds. The result is a trunk forln which resembles a sabre (Figure
5- 6) .
Construction of terraccs and roads is ot'ten at the origin of'
Ir~ndslides. The c ~ ~ t s weaker1 the slopc stability or locally increase watel-
inf'iltration ( e. g on the foot of reverse sloped terraces) which changes soil
coherence. A similar influence is exerted by overgr;lzing of land. The grass
cover is locally destroyed and livestock paths cut the slopc and clestabili7,c i t
(Wenncr, 1989).
Chapter 5
Figurcl 5-5: Critiool ~1lrlucs of' \torin d,rrlrtio~, rrnd i/lturritj- ,for i / ~c r e o~i / ~j i
lrrr~rlslide ~frer/rrc/zcy (Wilson et ol., 1992; Ln, ~r / l & Si/lloir. 1993;
Ccrirlr, 1980) (fi-om Larserl & Simorl, 1993)
storm duration [h]
Another informative source for erosion susceptibility is the geologic
map. In-situ soils are formed from the parent material underneath. Areas with
parent materials which form medium to light textured soils are more
endangered by erosion than those with materials that generate clayey or very
sandy soils. A tentative classification for different parent materials is given i n
Table 5- 1.
5 Indicator for \oil ero\ l on
T~rblc) 5 - 1: Soil or-oclibilitj, c!f'.soils cleri~!ecl,fron~ di f orr~l t l~nr-erlt rllcrtericrl,\.
Soil classification as well gives rough indications f'or soil
erodibility. With rcspect to USDA Soil Taxonomy erodibility increases in the
order Oxisols < Ul tisols < Alf'isols, Vertisols, Mollisols < Aridisols
(Chi-ornec et al.. 1989). Andisols were found highly variable (El-Swaify.
1990) and Vertisols proved more erodible than lnceptisols (El-Swaify 8L
Dangler, 1982).
1
erodibility
Erosion susceptibility decreases with increasing organic matter
content which in turn increases with soil moisture and decreasing
temperature. Therefore, a soil will have more organic matter in a cool
highland climate and will bc less erodible than a similar soil in the lowland.
low
basil t
gabbro
\hale
coar\e/gravelly \and depo\its
carbonate rocks
A very evident indicator for soil erodibility is soil colour. Structural
stability and erosion resistance of red hematitic soils is higher than of yellow
goethitic soils. In other words, with decreasing hematite content, as seen fsom
the redness rating (Torrent & Barrcin, 1993), structure becomes weaker
(Chauvel et al., 1976; Muller, 1977). Organic matter content is also roughly
estimated by soil dar-kness (Munscll values).
medium
gneis\
diorite
ande\i te
high
volcanic a\h
granite
rhyolite
granod~ori te
fine sand depo\it\
\ i l t stone
lee\\
5 Indicators for \oil el-osion
mean anual runoff [mm]
A first iinprecs~on of \oil erosion can be gathered fro111 the sediment
concentration in the river water. Rivers coming out of forested watershed$
carry very low amounts of sedi~iients colnpared to cropped and gra7ed
watersheds (Figure 5-7). An extreme example is seen from a plane
approaching Madaga\car. The high sediinent load of the rivers leaves a red
corona close to the estuaries and coast.
Finally, a very good indicator for erosion processes is population
density. Agriculture creates soil erosion and agricultural systems are often
not conservative. Thus, information about the population density combined
with knowledge about cropping systems, quality of the soils and cli~uate give
already an idea of the erosion potential.
Chapter 5
Further indicators like washed out root\ and top\oil depth are
sometimes already useful to quantify soil loss. They are therefore discu\\ed
in Chapter 6.3.3.
6 Assessment of soil erosion
Methods for soil erosion measurement depend on the scale applied and the
accuracy needed. Measurements are carried out on plots of less than 1 m2 to
watersheds of several hundred km7. The accuracy ranges from an estimate of
the erosion dimension to ineasurements precise to the kglha. Choosing a
measurement system, therefcjrc. needs a clear definition of the problern to be
investigated and the accuracy of answer needed.
6.1 Rainfall simulator studies
Rainfall simulators are used in the laboratory or in the field in order to apply
storms of controlled length, intensity and drop size distribution to erosion
plots. Today, a nurnber of different rainfall simulators exist which apply
perrnanent or intermittent rain from needles with drop-formers, small hoses
or nozzles to plots of varying sizes. Plot size is limited by the size of the
sirnu1;ltors and the availability of water in the field . Simulators can be as
( 1
large to fill a big truck or as small to be carried by hand (Crouch & Collison,
1989; Kamphorst, 1987). Largest and smallest plots of field simulators
actually used in Germany and Switzerland are 42 and 0.38 m2 (Auerswald et
al., 1992b). A comprehensive review on rainfall simulators is given by Meyer
( 1988) and USDA (1979). A detailed description and discussion of
sinlulators used in Germany and Switzerland was published by Auerswald et
a1. ( 1992a. b, c), Auerswalcl 81 Eicher ( 1992). Bechcr ( 1990) and Kainz et al.
( 1992).
Using rainfiill simulators, soils can be tested fairly quick, under
standardized conditions and independent of hazardous natural rainfall. The
air-dry soil (simulations are advantageously carried out during the dry
season) is exposed to several rains with varying duration and intensity.
Intensity can be atljusted to the local conditions or to the international
standard of 63.5 rnmlh. The latter facilitates comparison with other studies.
The standard treatment comprises a first storm of 1 hour. 24 hours later a
second 30min storm and after a 15 min break a third 30mi n storm. This
stor111 sequence represents rain o n dry, moist and wet soil as it occurs undcr
natural rain.
l 0 As a thumbrule about 10001 of watcr is nccded for a 60rnin storm of 63.5 m~ n / h on a 10 m2
plot if the intensity is rnensurccl bcfore and after- the storm.
Runoff and soil loss from moist and wet soil are generally larger
compared to dry soil. In order to calculate a mean soil erodibility for all soil
moisture conditions during the year, soil loss from dry, moist and wet soil is
weighted in a ratio of 1 : 0.31 : 0.23 (Wischmeier et al.; 1971). This ratio
proved valid for the climate in mainland USA. An attempt to adjust the ratio
to other climates was made in Hawaii (Dangler & El-Swaify, 1976) by taking
the number of dry and wet" months to weigh the storms on dry and wet soil.
A storm on very wet soil was not carried out. Correct results were obtained in
Ca~neroon by applying a ratio of 1 : 1 for dry and wet soil (Nill, 1993).
6.1.1 Laboratory studies with simulated rainfall
Laboratory tests are used especially for the study of single erosion processes
like surface sealing, rill and interrill erosion, splash erosion, intluence of
mulch layers and different rain intensities. Relative differences in the
erodibility of soils can also be evaluated. However, if quantitative
inSol-mation about soil loss is needed, the results frorn laboratory tests are
better calibrated with to data from larger erosion plots under natural rainfall.
I n laboratory tests only a part of the soil profile (generally thc surface soil) is
used and the soil is disturbed i n its natural structure. Therefore, results on
ri~noff and soil loss can only be compared t o in situ soils if the runoff volu~ue
is determined by the surface layer (= rapidly sealing soils). If runoff is
especially determined by less permeable subsurface horizons or the degree of
presaturation of the soil, laboratory test are of limited use. This is of'ten the
case for well structured soils rich in oxides, clay and organic matter.
The advantage of laboratory tests are the controlled conditions of
\lope angle, water temperature and quality (wme test\ are carried out with
disrilled water), rain intensity and antecedent soil moi\ture. The \mall plots
can easily be handled which facilitates repetition\. The comparatively \ma11
amount of soil needed allows the collection of very different soils distant
from one another.
I I A wet month was characterized by the median rainfall exceeding class A pan evaporation lot-
the month.
6.1 Rainfall siniulator studies
plan view
flume
side view
For the tests, a layer of soil is packed into a flume about 10 cm deep
where it is compacted to its natural bulk density for which a roller can be
used. A large nurnber of different flumes and rainfall simulators are used.
Therefore only some general features will be given here. A flume is a
wooden or metal box with an inner and outer area (Figure 61 1 - 1 ). The inner
area is connected to an outlet which delivel-s runoff and sediment into tl
container. The outer area is also exposed to the artificial rain. The idea of an
outer area is that the amount of splash which leaves the inner measurement
plot is replaced by the amount splash from the outer plot entering into the
measurement plot. The bottorn of the flurne is perforated in order- to allow
percolation of' the infiltrating rain. The flume can be adjusted to several slope
angles. Some flumes are also variable in their length.
6.1.2 Field studies with simulated rainfall
Field studies with rainfall simulators are more tiresome and expensive than
laboratory studies. The whole equipment and the necessary crew must be
transported to the site. If water is not available in the vicinity it needs to bc
carried from several kilometers away and stocked beside the experimental
site. Test conditions are not as controllable as in the laboratory. Antecedent
nioisture, slope, water quality and temperature can riot be standardized. On
the other hand, the soil stays rather undisturbed and the whole soil profile is
tested instead of a single soil layer. For determining soil erodibility, the plot
and a 50 crn wide strip around the plot are tilled to maize seedbed conditions
before the rest. The tilled strip around the plot serves for the same purpose as
thc outel- area of the laboratory flume. As the soil docs not need to be
transported, the plot size is generally larger in field studies cornpared to
laboratory tests. Generally the field si~nulators need to be calibrated on runoff
plots with known erodibility.
Field simulators also allow the testing of the effect of various factors
011 erosion such as vegetation cover during different growth stages and
seasonal variation in strt~ctural stability. Use of field simulators also gives
most realistic infiltration data as it reflects closely the intluence of natural
rain. In order to reflect the variation in soils and treatments two to four
repetitions are carried out in most studies.
6.1 Rainfall simulator studies
Figure 612-1: Mobile rainfall simulator unit in the field. 1: simulator;
2: manometer; 3: spraying nozzle; 4: supply hose; 5: outlet hose; 6: 300 1
tanks; 7: electric pump; 8: electronic control system; 9: 5000 1 tank;
10: 1 1 sampling bottles; 11: 5 1 beaker,- 12: outlet; 13: dissembled simulator;
14: 120 1 barrels for water transport
Chapter 6
nietrurl
40
dry-run
60-
40-
20-
ediment concentral
0 0
0 1000 2000 3000
time [s]
1 .o
I wet-run
The plots are bordered by metal sheets which are driven into thc
ground to 10-15 cm depth. At the bottom end a metal triangle is put into the
soil which collects the runoff into a tube and finally into a graduated bucket
(Figure 6 12- 1 ). Standard reported data are:
D antecedent soil moisture before the rain
D surface roughness
D startofrain(time=O)
D time for the first runoff
D time for every litre of runoff and runoff samples depending
on the volume (e.g. lst, 2nd, Sth, 7th, loth, 15th, 20th, 30th
litre, etc.) for sediment determination.
D end of rain and end of runoff
D initial and final rain intensity
With these data runoff/soil loss diagrams can be set-up which demonstrate
the erosion process (Figure 6 12-2).
6.2 Runoff plots
Measurements of soil erosion were originally conducted on runoff plots
under natural rainfall. A standard plot of 22.1 n~ length. 1.87 m wide on a
uniform slope of 9% was taken as 'unit' plot. It served as reference in
comparative studies. The 'unit' plot was tilled up- and down-slope to maize
seedbed conditions. Seals were regularly destroyed by further tillage
(Wischmeier & Smith, 1978). Thus, all conditions were set to attain
maximum soil loss. For soil erodibility measurements, a period of at least 2
years under barefallow was recommended to exclude all influences of the
former vegetation.
Today, experiments are carried out on plots of different dimensions
and on different slopes. However, these plots can be corrected to 'unit' plot
conditions with the Universal Soil Loss Equation (USLE). Nevertheless, a
minimum length of 9- 10 r-n is recommended for erosion plots. Calculation is
facilitated if the sill-thce area is equal to an even fraction of an hectare (e.g
SO. 100, 500 or I000 n12) (Shcng, 1990). The plots should be large enough to
contain a representative unit of a cropping system or treatment. Runoff plots
range between < 100 m' to about I ha. On larger plots different slopcs and
soils as well as deposition within the plot create a more and more complex
sitiration which is difficult to interpret. Additionally, i t becomes difficult to
contt-ol and measure the large amount oC water and sediment.
The runoff plots are bordered (metal sheets, bricks, earthen ridges
planted to grass) in order to prevent outside runoSf frorn entering the plot.
Runoff and sediment are collected by a drainage ditch at the bottom of the
plots which leads to an outlet. Here, the volume and sediment can be
measured by a divider tank system (Figure 62-1) or a Coshocton wheel. A
steady rneas~~rcinent of the runoff rate is possible by using a standard t l u~ne
and a watel-level recorder.
The tank system consists of a large tank which can collect all runofl'.
I f thc expected runoff volume is too large, a system of several interconnected
tanks or ban-els is needed. If the first tank is filled the overtlow is separated
into a large aliquot (e.g. 90%) which spills into an outtlow ditch and u small
aliquot (c.g. 10%) which enters the second tank and so on. Most of the
sediment, especially the coarser material. is deposited in the first tank
whereas suspended soil particles will be found in the second tank. All tanks
should be provided with an underground outlet to hci litate water evacuation.
Chapter 6
Fi gr~~- ( > 62-1: I l i l i der tmrk \jystc>ilr fi)r tilt) i~rr~rslrwirrrilt of r ui l of o~r d o i l
lo.~.c ( Snhrl - Ko, s c~l ~cl l ~~, 1988)
HS - F l u me ( \ - f oot )
Connect i on Tube
Concr et e Pi t
Coshocton wheels are installed in the waterspill underneath a flume.
The water falls onto the whccl thereby making it rotate. A slot in the whccl
which is connected to a barrel passes underneath the spilling water taking
each time a srnall aliquot of runoff. Coshocton wheels can be purchased for
about 1500 to 2000 US$. Their advantage is that they can be installed in
several places, while cemented tanks become worthless after the
measurement in one place.
6.3 Erosion measurement within existing fields
If less incasure~nent accuracy is needed, there are a number of simple devices
which can bc used to estimate soil loss.
6.3 Ero\ion nlcasul-emcri~ within cxi\ting I'ield\
6.3.1 Erosion nails
Erosion nails also called erosion pins can be hammered into the soil until a
defined length ( e . g 20 crn) stays out above the soil surface. If this length is
reduced or enlarged, sedimentation or erosion has occurred i.e. the soil
surface has increased or decreased. The nails are placed along the slope of a
field with an interspace of some meters and a lateral displacement of 10 to 20
cm (Figure 63 1- 1 ) in order to avoid any interference on I-unoff fro111 onc nail
to the nail bclow (Ziibisch, 1986). Length measurement of the nails rnust be
carried out with high prccision. An crror of 1 nlln in length means an error of'
10-1 5 tlha if bulk density is supposed to be between 1 .O and 1.5 gIcm3.
Therefore, a metal plate 5 ctn in diameter is slipped onto the nails to
compensate for random roughness of the soil surface. The length is measured
with a slide calliper precise to 0.01 rntii (Figure 631-2). The mean length of'
all nails is con~pared to the mean initial length. The soil loss can than be
calculated by the missing soil height if the soils bulk density is known.
A\ the measut-ement error can be appreciable, this tilethod is
e~pecially suitable for measurernentc over a period ol' several year\ or f or
\ite\ with a high erosion potential.
Chapter 6
I
Oel % lateral displacement
to avoid interaction of nails
I
I
I I I 2 m
I
I I
slope
2 m
erosion nails
6.3 Erosion r neas~~r ement \ within exi\ting field\
Slide caliper
erosion pin
\ \
iron
/
soil
- -il
surface
6.3.2 Sediment traps
Sediment traps are simple and cheap devices which permit the measurement
of runoff and soil loss. They were extensively used in ~iieasurements in
Kenya (Zobisch, 1986). The traps consist of a 50 x 50cm metal box closed
on threc sides by a 5 cm high I-im. A IOcm long extension of the bottom is
left at the open front part. This extension can be pushed into the soil in order
to allow runoff to freely enter the fnr end of the box which is installed at a
slight angle. The box is covered by a lid ;to avoid direct access of rain water.
At the far end a funnel is attached to the trap which is connected to a 301
reservoir where runoff and sediment are collected.
6.3 Erosion mcawremcnt within exi4ting I'ield4
The sedinient traps are installed in places with homogenous slope. A
conversion ditch 10m in front of the traps diverts runoff from further up-
slope. The catchment area of 5 m' for the sediment traps is given by the
width of the traps (50cm) and the slope length (10m). It is assumed that
runoff which enters the area laterally equals the runoff volu~ne which leaves
laterally. A sketch of a sediment trap system is given in Figure 632- 1 .
A system with this de5ign collect\ about 6 mm of runoff. I t needs to
be emptied after each rain. For larger storrns either the re\ervoir mu\t bc
bigger or the catchment area reduced.
6.3.3 Diverse techniques
Root growth of many trees is evidently inhibited on the exposed root parts as
thc bark and cambium are damaged. This was found for Pillus aristata (La
Marche, 1968), Pinus cdulis, Juniperus scopulorum and J. osteosperma
(Carrara & Carroll, 1979). Soil loss is calculated as the depth since exposuse
of the upper root surface which is observed on the growth rings
(Figure 633- 1 ).
Figlcl-c) 633-1: Ccrlculafi~zg fho time sir1c.e exposure of' tree roofs hj' t l ~c
growth rirzgs mot ~ e r t i o ~ ~ s
eroded part
of root
Dunne et al. ( 1 978) used a similar approach in Kenya by measuring
the height of the mounds underneath Acacia drepanolobium, A. tortilis.
Scricomopsis pallida, Olea africana and Acocanthera species (Figure 633-2).
These tree species were chosen because they do not develop any superficial
roots. However. caution must be given that the mounds were not formcd by
water and wind erosion, termites or the trees themselves. The age of the trees
was given by a regression between number of growth rings and diameter of
thc stem. Acacia drepanalobium is reported to develop a physiological mark
on the stem at the levcl of the original soil surface (a bulge, branching or
change of bark colour). An indicator for very erodiblc sodic soils is
Colophospermum mopane (Stocking, 1988).
0. 3 Erosion measurement within existing I'iclds
minimum level of
former soil surface
-------------
resent soil surface
Some neth hods can orlly be used very site-specifically. Rhoton ct al.
( 199 1 ) used the gravel concentration of the surface soil in order to calculate
the eroded depth of a soil with rather homogeneous gravel content.
Chapter 6
6.4 Sediment yield from river basins
Suspended sediment yield of rivers is calculated by measuring the cross
section of rivers, the waterlevel, the discharge and the sediment
concentration. Sediment yield, however, only gives a rudimentary estimate of'
the soil loss from fields within the basin. Several factors bias the result:
- Sediment yield measurements generally only cover the suspended
sediment load. The bedload which is carried close to the river boltorn is
neglected. The bedload of African rivers accounts frequently between 5
and 1 0 % of the suspended load (Walling, 1984).
- The soil lost within a watershed is not entirely transported into the river.
Sedimentation occurs on foot-slopes, depressions and well vegetated parts
within the watershed. The sediment delivery ratio (SDR) which gives the
suspended sediment load of the river relative to the total soil loss in the
watershed varies with watershed s i ~ e (Figure 64- 1 ). The SDR from large
watersheds is s~naller than fro111 s111a11 watersheds as the mean gradient
declines and sedimentation increases with increasing watershed size.
- The soil lost from fields in the watershed can be subject to several cycles
of deposition and remobilization until i t reaches the river outlet. Thus,
measured suspended sediment load may reflect soil erosion of' formel-
periods.
- Suspended sediment is not only derived fi-orn sheet erosion within the
watershed but may also stern from gully erosion, landslides, channel and
streambank erosion.
river
Watari
Bun\~u-u
Senegal
1 Faleme
, Ham~nanl
Kebir Oue5t
1 Meunu
country I basin area (suspended sedimen
Nigeria
Nigeria
Mali
Mali
Algel-ia
Algeria
Ethiopia
(krn2J
FA0 soil loss estimate
load
[t/(ha;J:a)j
6.3 Sedinicnt yield I'ro~n I-i \ rr ba\in\
The difference between suspended sediment loads of rivers and
estimated soil erosion rates fror-n the respective fields is demonstrated in
Table 64- 1 . Suspended sediment load is generally an order of magnitude
lower than the estimated total soil loss.
Figltt.rCi4-1: Relcltiotl.~l?ip brfit'retz \t,crtur.sl~ecl tlrctirzcrgo at-eu CI I I L I
setlit~~erlt cleli\vrj rcrtio as 1tsc.d hjr the U.S. Soil
Corzset-11trtio1l S~t-i' ice ,fi)t- the c'etltt*er/ nrz~l ee~stem USA
( f i ~ 1 1 7 2 WCIIII' II~, / 9 NH)
0 01 0.1 1 10 100 1000
Drainage Area (km2)
7 Soil loss prediction with the Universal Soil
Loss Equation
Erosion has already been noticed in ancient times. Plato already described the
disastrous effects of the denudation of the hills around ancient Athens more
than 2000 years ago (in: Herkendell & Koch, 1991 ). However, more attention
to the problem was only given by the 1920s when the menacing extent of soil
loss in the US became aware (Bennett & Chapline, 1928; Lyon & Buckman,
1922). As a consequence the US Soil Conservation Service was created in
19.35. Soon i t became insufficient to notice, describe and measure soil
erosion. For a deeper comprehension of erosion and its assessment under
varying conditions, i t was important to understand the basic processes.
The development of mathematical models started with the equation
o f Zingg ( 1940) which related soil loss to slope length and gradient. Smith
( 1941 ) included factors for the influence of crops and conservation practices
on soil loss. The addition of a rainfall fiictor resulted in the Musgrave
equation (Musgrave, 1947). Finally data collection and analysis of 10.000
plot years from 49 locations led to the 'Universal Soil Loss Equation
(USLE)' (Wischti~eier & Smith, 1978) which, today, is still the basic tool for
soil conservation in the US and other countries.
The USLE is an enipirical model with widespread use in land use
planning, extension and the design of cropping systems and conservation
practices. I t allows to estimate soil loss under varying climatic. topographic
and management conditions on different soils with a set of relatively sirnple
parameters. The basic idea was to measure maximum possible soil loss of a
specific soil on a control plot with standard size, gradient and treatment, - the
'unit' plot. The 'unit' plot was 22.1 m long on a 9 % slope. Soil loss as
caused by gradients, slope lengths and management conditions different from
the standard conditions was examined relative to ~naximum soil loss on the
control plot which was achieved by barefallow tilled up- and down-slope to
maize seedbed conditions. The equation is expressed as:
with A rilean, longterrn annual soil loss [t/ha*a]
R erosivity of rain (Nlh]
7 Soil leas pl-ediction
K erodibility of a soil, i.e. its susceptibility to erosion
It+h/N:. bhal
L slope length factor [ - I
S slope steepness fdctor [-I
C management factor [-I
P support practice factor I-]
Soil loss ( A) gives the mean annual soil loss in tlha on a longterm
basis. Soil loss of a specific year may differ considerably from year to year.
Rainfall erosivity (R) is calculated from rainfall charts for single erosive rains
during a period of 22 years and represents the mean annual erosivity for this
period. Soil erodibility ( K) indicates a soil's susceptibility to the erosive
forces and gives the amount of soil loss per unit erosivity. K was defined
constant for a specific soil. L, S, C and P are expressed as ratios of soil loss
on a given plot to soil loss on the unit plot. For example, an L factor of 2.1 for
a 100 m long slope of 9% means that this slope will suffer 2.1 times the soil
loss of the 22.1 tn long unit plot if all other conditions (climate, soil,
management etc.) are alike. A C Factor of 0.2 for a crop signifies that soil
loss under this crop is only one fifth of the barefallowed unit plot provided
that all other factors remain constant.
The model parameters were calculated from a defined set of natural
and management conditions in the US. Therefore, it was not surprising that
the application of the USLE has led to contradictory results under tropical
conditions (Lal, 1980; Mtakwa et al., 1987; Ngatunga et al., 1984; Roose,
1977; Vanelslande et al., 1984). Part of the differences were however caused
by treatments very different from the one's defined by Wischmeier 8( Smith
(1978). Recent data show, that the USLE can be directly applied to a wide
range of tropical soils and corrections can bc made for most other soils (Nill,
1993). The most urgent need exists now in obtaining reliable data on tropical
cropping systems.
Today, several deterministic models exist which try to con\ider the
numerous, complicated processes which determine erosion. Mostly they need
a large amount of infortnation on climate, soils and management. Often they
are not tested under differing conditions. Compared to these models, the
USLE convinces by its simplicity, the large data base which was used fur its
development and its widespread application. Although empirical in principle,
it still includes all irnportant factors which influence soil loss. Its parameters,
possibilities and limitations will be outlined in the following chapters.
Chapter 7
The USLE was designed to predict longterm annual soil loss from a
given slope under specified land use and management conditions
(Wischmeier, 1976). It can be used for watersheds, if these are subdivided
into smaller units where the USLE factors apply. Using mean gradients,
erodibilities and slope lengths for the whole watershed may cause important
errors in the estimate. Soil loss, as estimated by the USLE should rather be
regarded as best available estimate than as absolute data. Soil loss from a
specific event can not be calculated with the USLE. Even annual soil loss of
a specific year may vary largely from longterm mean annual soil loss. The
USLE does not account for deposition of sediment along field borders, ridges
or on foot slopes and can not predict gully erosion.
Beside the USLE, a second important prediction model is applied in
southern Africa. The 'Soil Loss Estimator for Southern Africa (SLEMSA)'
(Elwell, 1980a) predicts mean annual soil loss (Z) on a given slope by:
with K mean annual soil lo\s from a 4.5 O/c slope, 30 m long under
conventional tilled bare soil
X adjustment factor for different slope length\ and -gradients
C adjustment factor for the influence of crop cover derived
from the annual energy distribulion curve and growth
c ~~r ve s of crops
An appreciable database was collected for this model. For furthcr details rcfer
to Elwell ( 1980b), Elwell ( 1984) and Elwell & Stocking ( 1976).
7.1 The erosivity of rain
7.1 The erosivity of rain (R factor)
Wixhrneier & Smith (1958) found that soil loss increased linearly with a
storm's total kinetic energy (E) times its inaxiinuin 30 minute intensity (I3()):
with R longterm mean annual erosivity [ ~ l h ] "
E kinetic energy of a storm j [kJ/rn2]
I,,, maximum storm intensity of storm j during 30 lnin [mm/h]
for I,,, > 63.5 mrnlh: I,,, = 63.5 mm/hl'
In number of erosive storms j per year [-1
The energy of a storm is calculated by:
with I i intensity for storm interval i [rnmlh]
for 0.05 < I < 76.2 mrnlh; for I > 76.2 rnrnlh 1 = 76.2 mm/ hl
PI rainfall volume during interval i [mm]
n number of storm intervals i with equal intensity [-1
R is calculated from raingage charts. Each storm is divided in i
intervals of constant intensity ( I ) . For each interval intensity, volume and
energy are calculated. The total storm energy is the sum of energy of all
intervals. An example is given in Figure 7 1 - 1 :
'' R is oftcn given i n IJS units us [hundreds foot tons 'I' inlac ph] or as Ifoot tons ": inlac ph] .
Multiply by 1.735 or 0.01735, respectively. to receivc INthl. 1 [Nth] = 10 IMJ :!: mmtha :': h]
1 ,,, was limited to 63.5 rnrnlh because col-I-elation coefficients between crosivety and \oil loss
irnproved by introducing this threshold (Wischmeicr & Smith, 1978)
The rnaxirnunn intensity was limited to 76. 2 mmlh because drop diameters do not increase any
rnorc for \.cry high intensities (cf. C'hapter 4.1)
Figcue 71 - 1: Strip rlztrt-t of rr 20 m n ~ storrzl registered n~itll tr .\elf'
rc.c.or-cling rcri~ gcrge rtyith cr pcrpe).,fLotl t-trtc of'hO t11171/11
A 20mr-n \term was registered by a raingage with a papenpeed of
60 mmlh and a cylinder which emptied automatically after each l 0mm of
rain (= vertical drop of the line). The storm \tarted at 17.00 hour and lasted
until 19.48 hour. It was divided into 4 intervals of approximatcly equal
intensity (= slope of the ascending curve). Energy is cornputed as follows:
The maximum rain volume during 30 min was 7.8 mm in interval 2. Thus,
I,,, equals 15.6 mmlh and R = E :!: I,,, = 0.4 9 15.6 = 6.2 N/h.
Only 'erosive' storms are used in the calculation. For the US, they
were defined as storms with at least 12.5 mrn (112 inch) of rain or, if less, a
maximum 30min intensity of at least 12.5 mmlh (Wischmeier & Smith.
energy
[kJ/m2]
0. 13
0. 22
0. 04
00.008
0.4
intensity
[ m d l
6. 6
12. 8
4. 7
1.4
7.1
rain volume
[mml
7
10
2.4
0.6
20
interval
1
2
3
4
1-4
duration
[min]
64
47
3 1
26
168
7.1 The erosivity of rain
In Germany. the limit for erosive storms was set to 10 mm height or I 0 mm/h
as maxirnurn 30 min intensity (Schwertrnann et al., 1987). The 10 mrn
threshold also proved to be valid for stations in Cameroon. Nigeria and
Kenya (Nill. 1993; Ulsaker & Kilewe, 1984; Wilkinson, 1975) and was used
for all further computations. Storms separated by less than 6 hours are
considered as one storm (Wischmeier & Smith, 1978).
The tedious procedure for the energy calculation is easier done by
cornpoter and di gi t i ~i ng board . Providing erosivity data on a nationwide
b;i\i\ i \ an important task for the national meteorological kervices.
In practice, the calculation o f reliable R factors faces several
constraints. Ideally, the calculation is based on daily rainfall records over 22
years (Wischmeier & Smith, 1978). However, in most countries it is already
\,cry satisfying if 10 to 15 years of complete data are available. The
obligatory subdivision of individual storms into intervals of similar intensity
ancl the recognition of the maximum 30 min intensity demands self-recording
raingages with high resolution. Very often these requirements are not met and
several estimation procedures and indices have been developed in different
countries in order to rcplace the R factor (cf. Chapter 4.1 ).
Determination of the R factor:
Erosivity for many locations in Africa must be estimated from
available data of different origin:
-
In some countries erosivity is calculated for single sites.
-
In other countries regressions exist which may be extrapolated
to the surroundings.
-
For some countries national or regional erosivity maps
(iso-erodent maps) are available.
For countries where no erosivity data are available Eljo must be derived
fro111 rain data or rainfall distribution maps.
l 5 A \ol'~\zar.c psogr-anlln for di gi t i ~i ng and analy/.ing sainl.all chart\ i \ a ~a i l a bl e Iiorn:
Dr. M.'. Martin. Bahcri\che\ Gcologisclies Lancle\rimt. Hel3str. 118. 80797 Miinchun. Ger-rnun
The quality o f the obtained erosivity values will be no re reliable for
sites or areas where E13,, was directly calculated from rain data (provided that
the measurement period was sufficiently long). If regressions are used. the
reliability decreases with increasing difference in climate and increasing
distance from the stations of which the regressions were derived. National
erosivity maps generally will be more precise than regional maps. For
estimates of erosivity from rain data or maps, several regressions can be
applied (Table 12- 1 Annex).
For the Sahel countries Roose' s regression is recommended (Roose,
1977). The regression of Bresch ( 1 993) was developed from 18 stations in
Cameroon with 700 to 4000mrnla. Its use is proposed for the semi-humid to
humid parts of West and Centl-a1 Africa. The equations for Zambia
(Pauwelyn et al., 1988) and Zimbabwe (Stocking & Elwell, 1976) are based
on a large and well described data base and are recommended for areas of
southern Africa with comparable climate. For the highland areas of East
Africa, the regression for Rwanda (Durand, 1983) and Kenya (Moore, 1979)
can be used.
If regressions for near by neighbour countries are available, the user
may decide whether the climate can be co~npared to these countries and
whether these regressions may provide reasonable estimates.
In order to obtain El3u for a particular location, look for the country
in Table 7 1 - 1 and check if the site or a site nearby is listed. Table 7 1 - 1
indicates the reference tables in the Annex where you can find the EIjo
values. If the site is not listed, look up a national or regional erosivity or
rainfiill map as indicated in Table 7 1 - 1 I".
A given rain falling on low slopes (between 0.2 and 4%) is not as
erosive as the sarne rain o n steeper slopes due to the formation of a protective
water mulch. Correction of erosivity on low slopes demands the erosivity of
the 10 year storm (EI,JIo)". E1,,410 was found to be inore suited than inran
annual erosivity as runoff depth is especially determined by the intensity of'
Daily rain data for a11 \lalion in Benin. Ruskinn Faw. Cameroon. Central Afsic:ui Repuplic.
Cliaci. Congo. Ciabon, Ivory ('oast. M:~li, Nigcr, Scnegal and Togo are a~ai l abl e 1'1-0111 the
Cornit6 In(erafr-ican d' Etude\ Hydl -a~t l i quc (CIEH). B.P. 3hO. Ouag,.aJougou. Buskina 1:;lso 111
l uo \erich. Sesie I : Station\ cctablislieci until 1965. Sesie 11: I965 - 1080.
" ,411 c\timation method lor E17,JI0 is descl-ihed in Annex 1.7.
~ndividual storm\ (Renard et al., 1992). With EI,,,/I 0 a correction fiaclor can
hc scad fro111 Figi~re 7 1-2. Enter the chart vel-tically from the x-axis with the
EI:,,/IO of the site. Choo\e the gradient of the slope and read the corrected
ero\ivity value by moving hori~ontally to the y-axi\.
Figllt-c~ 71-2: Not~logt-trph fi)r. tllt~ c'ort-c~c.tiotl of orosi ~i t j * for f l ~ e eff2,c.f o/
nvlirer r?r~ilch or! ,rlopc.c het ~vooi ~ 0.2 nlld 4 C/c
( Fot t or, per-so~lrrl c7ot?lt?zllnic.otiot~)
100 200 300 400 500 600 700
El30 of 10 year storm [Nlh]
If hail i 4 a frequent event a \ i n ronie mountain area\. the annual
cro\l\ity \hoi~ld be con-ec~ed by estimating the percentage of annual
pl-ccipi~a~ion as hail. This percentage of the annual ero\ivity mu,t be
multiplied by 2.5 (Hurni, 1980) and added to the remaining annual erosivity.
Chapter 7
Example:
It is e\timated that 20 52 of the annual rain falls during hailstorm5 i n an
area with a mean annual erosivity of 800 Nlh. 20 5% of 800 Nlh corre\ponds to
160 Nlh. Multiplied by 2.5 = 400 Nlh. Thus, the annual erwivity corrected for
hail i5: 640 Nlh (= 80 %) + 400 Nlh = 1040 Nlh.
country site
I I
map map map
MU, lN/hl
I
sites regression national regional
Annex
Algeria
rain volume
[mml
1. 1 1.2 1.3 1.4
X
Ciour~u-i (1shcr hi ~\ i ~l )
Hcri/ (Isser hnsin)
Macljouclj (Isser basin)
Sidi Mohiirnccl
Cherif ( 1ssc.r hasin )
Angola
Benin
Botswana
Burkina Faso
Boho-Dioulu\\o
Dori
Fada-N' Gou~.rna
F~iraho-Ba
Garnpela ncar
Ouagacioi~gou
Gaoua
( ; on\ & []car
Ouagadougou
Mog~cdo
Nii~ngoloko
Ouagaclougou
Ouahigouya
Saria (Mct cn)
Burundi
Mashit\i (Gihcta)
national
X
X
X
X
X
X
X X
X X
X
X
X
X
X X
X
X
X
X
X
X
X
7.1 The eso\i\ ity of rain
country site
Cameroon
B. f ' .
'I 1'1
B~l menda
Rang;itigte
Batouri
Dilxinlha
Dollala
Dwhang
Garoua
ML I ~ O L I ~
Mciganga
Nachtig;ll
Ngaoi~ndGrC
Nhount i j a
Penha Michel
( Banwa)
Poli
Y ~1ound6
Y o k o
Central
African
Repuhlic
Chad
Congo
Eg\ pt
13quatorial
Guinea
Ethiopia
Gahon
Ganihia
Ghana
Guinea
(;uinea-Bissau
I\or! Coast
4hldl'in
Ar,lpu16
Bo~lithk
countrj site
Ivory Coast
Divo
KorI1ogo
Kenya
Eldoret
Kat i ~~nani
(Macliako\)
Kisurnu
Kitale
Lodu iur
Muli~idi
Mornhasn
Nuiruhi
( Kahctc)
Nakul-LI
N:uiyirhi
Naroh
Voi
Lesotho
Liberia
Ldihya
hladagascar
Hefandriiu~ia
Malawihlnlr
hlorocco
klauretania
Xlocamhic~ue
Niger
AllohoLo
Nigeria
Alore
Culabal
Enugu
I bad~ui
Ih orn
N\ i ~kha
7.1 The e t o v L ~ t y ol ralu
country site
1 Nigeria
-
Onitahu
Oue r r i
Port- Harco111-t
Umildike
Rwanda
Butare
Galiir(a
( i i wnyl
Ka ~n c mb e
Ki p 1 1
(~lir[>(3rt)
Kuhengeri
Sao 'l'ome and
Principe
Senegal
Barnhe!
s 6 1.21
Sierra Leone
Somalia
South Africa
Sudan
Tanzania
'I'ogo
Tunisia
Llganda
Zaire
Zamhia
Chipatii
Kabompo
Ka bu e
Ka l ' ~~a Polder
K:r\:~rrla
Mwir11Iung;l
Ndola
Sedi eke
Chapter 7
rain volume
Ln1mI
national
map
country site
Zimbabwe
Bcibridgc
Chipinga
Chisumbarijc
Delt
Eastel-n Ilistricr
Enkeldoorn
Forr Victoria
Gokwe
Hi gh~el d
In) ungn
t i ur oi
Low~cl t l
Lupane
Middleveld
Si~lishury
Tjolotjo
'Tuli
Regional:
Sahel
El.,,, [ Nh l
sites regression national regional
map map map
X X
X
X
X
X
X
X
X
X
X
X
X
X
X
X X
7.7 Soil erodibility
7.2 Soil erodibility (K factor)
The soil erodibility factor K of the USLE expresses a soil's susceptibility to
erosion. It is defined as '... a quantitative value experimentally determined.
For a particular site, i t is the rate of soil loss per erosivity unit as measured o n
a 'unit plot'. A 'unit' plot is 72.6ft long, with a uniform lengthwise slope of'
9'k, in continuous fallow, tilled up- and down-slope.' (Wischrneier & Smith,
1978). Crusts on the soil which form during rains have to be regularly
destroyed by furrher tillage. In order to exclude influences of the previous
vegetation, the unit plot is kept under barefallow for at least 2 years before
determining erodibility. It is assumed, that by then soil loss is primarily a
function of inherent soil properties and increases linearly with the rainPall
erosivity. Erodibility is considered to be a specific constant for a soil and is
calculated by:
K =
A
R L S C P
I t h/N ha]
On a unit plot L, S, C and P equal I and the equation can be written as:
A
K = R [t h/N ha]
This basic concept of erodibility can also be applied to tropical soils.
As shown in Figure 72-la, erodibility initially increased after clearing of the
vegetation. After 1000 to 2000 N/h (which corresponded to 2-3 years in the
example) a steady erodibility value was approximated. However, on some
soils erodibility may still increase or decrease after some years of barefallow.
Erodibility of the soil in Figure 72-lb, for example, started to decrease
slightly after 4600 N/h. This is the case if a surface horizon is partly or
cornpletely eroded and tillage inixes the underlying horizon with a lower
erodibility more and more into thc initial surface horizon. Tropical soils
rnostly have surface horizons of less than 15 cm depth. Partial or complete
truncation
Figrrrc~ 72- 1: Chotzge of soi l erodibility with cr4i~z~rl~rti~~e e r o. ~i , ~i t ~~ oil [rtr
Ultisolfi-oi)i Cl11~0rooi7 ( ( 1 ) cltzd ail Alfisol f;.oi~ Nigeriu ( 1 1 )
soil
soil loss
[cdal
5.8
Andisol over basalt
[Yha*a]
698
7. 2 Soil cl-odihility
ol' a barefi_lllow \oil i i poq5ible under tl-opical rain within a few year\ a\
ihown by annual cro\ion depth\ of ioriie Canieroonian soils (Table 72- 1 ).
A decrease in erodibility occurs if the surf.iice soil or subsoil
contains coarser particles like quartz or iron oxide gravels. With the selective
rcriioval of the l'i~ic-earth. the gravel is enriched on the soil surflice and
protects the soil. Soil loss estimates for gravel-covered soil need, therefore. to
be corl-ected 1)s the protective influence of' thc covcr. An increase in
erodibility takes placc i f an irnstablc sirbsoil ( e . g with high sodicity) is more
and more incorporated into the surtlcc soil.
Measuring erodibility is time consul~iing arid expensive. Wischrneier
& Smith (1978), therefore, cariie up with an equation to calculate erodibility
from sirnple soil properties which arc moasured routinely:
K = 2.77'!' 10-0 MI 1-l ( 12-OM)+O.O43 (SC-2)
+O.O33'!' (3-PC)
M here
M I-] = (\i+fl'S) ( 100-cl)
~ i t h c1 clay [ ( X 1
\ i \ i ~ t [ ( x 1 ' '
ffS very fine sand (0.05-0.1 ~i i m) 1% ]
OM organlc rnatter [C/c I
SC str-ucturc clas\ 1-1
PC perrneability cia\\ [ - I
The equation shows that soil erodibility incl-cases with increasing
silt plus very fine sand content of the soil. It decreases with increasing clay
and organic niatter content.
Structure class of a soil (Table 72-2) docs not refer to the actual
structure of the soil surface of a field but to structure after 2 years of
barefallow. Therefore. sorne experience is needed in order to assign a
str~icture class to a soil. Soils with an ~rnsrable structur-e develop coarse
f r agr ~i ent ~i f ' t e~- prolonged barefallow periods whereas stable soils maintain
an aggregated surface. The coarser the final struct~rre, the higher the structure
class and erodibility.
structure
class
The permeability of a soil describes its inf'iltration capacity and
ability to conduct water. Per-nieability classes (Table 73-3) must be
detel-mined for all hol-i~ons down lo XO cm depth. For each horizon a
permeability class is chosen. The permeability class of the soil is deterrnined
by averaging the permeability classes of all I~orizons.
structure
size [mm]
I
3
-
3
4
It the hor i ~on with the Iowe\t pcrrncabil~ty I \ within the upper 30 cnl, i t \
pesnieab~l~ty is counted twice before averaging. If the lea\t permeable
hor17on I \ found within the uppel- 20 cm, i t determine\ the permeability cia\\
of the \oil.
mean aggregate
I
For field use, the permeability of a soil can be cstilnated by using
information on biological activity or structul-e in the profile descr-iption. An
exarnple is given in Tublc 72-4. However, use of such data needs experience
and should only be considered carefully.
very fine CI-umb
tine crumb
medium to coarw crumb
hl ochy. platy or mas\ive
< 1
1-2
2 10
> 10
7. 2 Soil erodibility
tigrlr-r 72-2: C'oirq?rrri,\orl 01 c.rrl(.r[lo~rd (K,,,,,) r r ~~d i~zen,sur-ed (K, , , , , , , ) soil
er-ociibilit~, for 28 .soil.\ fro111 Ccz~llcroor~ crild Nigc>r-io
description
very few pore<
few pore4
common poi-e\
many pored pot-ou4
very porous
very high biological
activity. very porou\
Equation No. (24) wa\ applied to soil\ with < 65 '/c \and and < 35 %
clay (Wischmeier et 31.. 197 1 ). K factors for \oils beyond the5e limits need to
be deternl~ned from a nomograph. However, ;I recent investigation indicated
n o quality lo\\ if erodibility wa\ calculated by the above equation for \oil\
beyond the\e textural propertie4 ( Ni l l . 1993).
permeability class
I
2
3
3
5
6
Chapter 7
Erodibility measurements on 28 tropical coils t'rom Cameroon and
Nigeria showed that equation (24) can not be applied to all tropical soil\ but
needc correction factors for 3 different \oil group\ (Figure 72-2). For part of
the \oil3 (group I ) , erodibility as calculated by equation (24) underectimated
the measured erodibility whereas group 3 wa\ clearly overe\ti~nated. For
group 2 (about half of the soil$), calculated erodibility agreed well w~ t h
meaqured erodibility.
A d~scrirninant analysis can distinguish between unknown sods by
using two di\crim~nant function,". 89 % of all sollz were correctly cl;l\\ed by
uqing:
-
bulk density of 5-10 c111 depth [g/cm7]; meajured one day after the
soil had been tilled with a hand hoe to \eedbed condition5 of ma i ~ e
-
\ilt content of the \urface \oil I(/( ]
-
organic matter content of' surfllce soil [%]
- pH in water of the \u~-face \oil; measured in 1 : 2.5 soillwater
~ 5 p e n s i o n after 18 h
-
amount o f air-dry aggregate\ of the surface \oil ( 0- 5 cln) bi t h
0.6-0.2 rnrn diameter [%I rnea\ured by dry-sieving
Wrong classification only becomes dangerous if a soil' s erodibility
is underestimated. If it is overestimatcd, too much conservation efforts may
be the result which means an exaggerated input of labour and money but no
virtual danger.
All soils with a very high erodibility (group I ) were correctly
classed. 9% of low erodible soils (group 3) were classed into group 2 and
would receive more conservation than necessary. 17 % of the medium to high
erodible soils (group 2) were assigned to group 3 and would receive
insufficient conservation.
Based on the two discriminant functions, Fisher's Linear Discriminant
Functions were derived which facilitate classification. They were:
'('function 1 = 13.03:i:BD(5- 10) + 0.169:!:si + 0.265:!:OM - 1.62:::pH - 0.066::agg(06-02) - 5.62
eigenvalue = 1 . OX, Wilks' Lambda = 0.19 sign. = 0.000 1
eigenvalue = I .OX, Wilhs' 1,ambda = 0.48 sign. = 0.0038
with BD(5-10) bulk density in 0 - 5 cm Iglcrn']
si si l t [ %]
O M organic rnatter [ %I
agg(O6-02) dry sieved aggregates with 0.6 to 0.3 m dian~elcr ('A 1
The group centr-aid\ for function 1 and 2, respectively. were -3.6 anti 0.93 for ~ I - ~ L I P I . 0.938
and 0.77 for groi~p 2 and -0.76 and I . 1 X for group 3.
7.7 Soil erodibility
group 1 = 1.118 si + 90.5 . BD(5-10) + 0.416 OM
+ 13.4 pH + 0.724 agg(06-02) - 100.6 (26)
group 2 = 1.7 . si + 134.7 a BD(5-10) + 1.395 OM
+ 7.577 .+ pH + 0.478 r agg(06-02) - 114.4
( 27)
group 3 = 1.343 4 si + 114.2 *BD(5-10) + 1.617 *OM
+ 7.816 * pH + 0.378 .;; agg(06-02) - 90.5 (28
In order to assign a soil to one of the groups, the three functions
must be solved. The soil belongs into the group whose function yields the
highest value. Once the group is selected, the erodibility of the soils (K-trop
[t - h/N -ha]) can be calculated by the following regressions:
Chapter 7
group 1: Kt,,, = 2.3 K,,, + 0.12 (29)
group 2:
Kt ,,,, = 1. 1 , K,,,
Applying the three regressions to the set of 11-opical soils mentioned
above, explained 92% of the variation in measured soil erodibility (Figure
72-3). The third regression was not significant. However. soils in group 3
have very low erodibilities (maximum K,,,,;,, = 0.026) and prediction errors
lmay be tolerated.
How can the typical soils for groups I to 3 be characterized? Table
72-5 shows average properties for the groups. Group 1 contains soils with
more sand, less clay and a higher amount of aggregates in the size fraction of
0.6 to 0.2 mrn and have a slightly higher pH than soils in ~ I - O L I ~ 2 itnd 3. Bulk
density is lower than in group 2. The surface soil of group 1 readily seals.
Their low bulk density and high ariiount of transportable material enables
high soil loss rates. They are characterized by an early occurring runoff, high
runof'f' rate and coefficient. The agronornically very important volcanic ash
soils belong to this group. An alternative ecluation to calculate erodibility was
developed by El-Swaify & Dangler ( 1977) for a group of seven residual soils
and five volcanic ash soils from ~awai i "' :
20 dimen\ion lol- K: [[(on ;: ac1.t. .:: IIOLII./ hundrt.d\ ol'ac1.c .:: fool t on\ ;:: inches]: i n ol.tit?r t o at't'I\'e
at I r :i: IIIN ::: ha1 tnultiply wirh 1.3.
7. 2 So11 erodibility
Tcrhle 72-5: A\wrcrge nrtlojfl .coil lo.~.s t r ~ e / soil pi-opurtius ,for t l ~u tl~roe
rrodihilit~ groups (1~cr1zte.s $tirll different lettens crre .sigil~fi'c'crrlfl~.
c/(ffet.et~f crt the 0.0.5 /oI)o/);
wit11 LT250 In percentage of soil which passes a 0. 25 t11r11 sieve by wet-
parameter
\and [(k]
very fine \and [(/c I
\lit 1% 1
clay [ % I
organic matter [ %I
pH
bulh den\~ty ( 5 - I0 cm) {g/cm3]
bulh dens~ty (0-10 cm) lg/cmz]
aggregate4 (0 6-0.2 mm) ['A]
Ktliea5 [ t h / N ha]
$tart runoff 141
111ean runoff rate I llmin m']
~naxlnium runoff rate [llnlln ,111'I
runoff coel'l'ic~ent 1 % ot l-aln]
$011 lo\\ [tlhu]
mean $o~l lo$$ rate (g/l I m~ n ]
maumun~ $011 lo55 rate [g/l min]
slevlng
MH (\iuid > 0. I 111111) (silt + Lery fine \and)
H C ba\cx \ ; i l ~l ~- ; ~t i o~i i l l 1 11 NH40Ac at pH 7
\ i percent \ i l t
s ;1 \and > 0.1 111111
Soils in group 3, as the other extretiie, tcnd to be more clayey iuid richer in
organic 1-naltcr than soil$ in gr o~l p 1 and 2. Bulk density is slightly lower than
in ~ r o u p 2 ancl there are le\\ aggl-cgates of 0.6 to 0. 2 mnI ilinrnetel.. Thew
soil\ h:it-dly \eal and I-unoff start\ very late if rain fall\ on dry soil. Soil lo\\
group I
5 2"
2.8"
1 6"
32"
4.2"
5.3"
0.87"
0.88"
19"
0.2775"
X60a
3.1
5.2"
3 2,'
8.0
33,l
49%'
means for:
group 2
43,l
3. I "
1 8"
40G'
3.8"
5. l L 1
1 . 0 6
0.97 'I'
2sh
0 0969''
1339"
2.2"
4 3"
23"
-. 3 9''
- 3?s1 -
36'
group 3
31"
3.0"
1 7"
52h
6.3"
5 .O"
0.87'
0.79
-- 37'1
0.0077L
3X73b
0.3"
0.8
3h
0.1'
3"
5"
Chapter 7
under natural rain OCCLISS especially after sequences of several storms which
presaturate these soils. Their runoff-soil loss behaviour is not determined by
surface sealing but by the permeability of the profile. The poor I-elationship
between surface soil and profile properties explains the insignificant
I-egression in Figure 72-2 for this group. Drop impact and storm have little
influence on soil loss. As runoff in group 3 requires a presaturation of the
soil, occasional rains during the dry season are of no danger. Especially
clayey, iron oxide rich soils from basic parent rock are found i n this group.
Soils in group 2 tend to have more silt and very fine sand, along
with mediurn clay and sand contents (Table 72- 5) . The organic matter content
is lower than in the othel- groups and bulk density higher. Their sul-face seals
as in group 1 but sealing and runoff occurs later during a rain resulting in
lower rnean runoff and soil loss. This is indicated by smaller differences of
~naximum runoff and soil loss rates of group I and group 2 soils compared to
mean rates. It can be assulned that maximum runoff is reached at the end of a
storm for soils in group 2. Croup 2 had 7 1 % of the mean runoff rate of soils
in group I but reached 83% of the maximum runoff rate of group 1 . The
values for mean and maximum soil loss were 7 0 and 74%, respectively.
Thus, with increasing rain volume the difference in runoff and soil loss
between group I and 2 became smaller. However, mean and maximum soil
loss rate did not differ as rnuch as mean and rnaximuln runoff rate. This
suggests that runoff increased more than soil loss. Typic soils in group 2 are
formed from metamorphic basement rocks.
Soil taxonotny gives some, though not very safe, indications.
Oxisols frequently are to be found in group 3 although some occur in group 2
as well. Ultisols, Inceptisols and Alfisols are especially found in group 2.
However, some soils in group 2 also have low erodibilities (Figure 72- 2) and
rather stable structure.
7.2 Soil credibility
L)etenninatioiz oft he K factor:
1 . Calculate K,,,, according to cquation (24). Silt, clay, very fine sand and
organic matter content ar-e taken from soil analysis o f the surface soil.
Structure class is choscn from Table 72-2. If doubts exist which class to
choose, erodibility can be calculated for two different classes in or-der to
receive the range in soil loss.
Permeability is calculated as explained on page 96 and shown in thc
following cxamplc:
In soil I , the lowest permeability corresponds to the deepest horizon
within 80 cm depth and per-rncability class of the soil is calculated as mean
permeability of all horizons to 80cm depth. In soil 2, horizon Btl has the
lowest permeability and lies within 40cm depth. Therefore, i t is counted
twice (sum of all classeslnurnber of horizons = 1615 = 3.2).
2. I n order to decide into which erodibility group a soil belongs,
equations (26) to (28) must be solved. The group with the highest result is
assigned to the soil.
3. Erodibility (K,,,,,,) is calculated for group 1 and 2 soils from equation
(29) and (30), respectively, or can be read from Table 72-6. The regression
for group 3 soils (equation (31)) was not significant. It is recornmended to
use the maxinlurn erodibility K,,,,;,, = 0.026 found for group 3 soils. As 3 0 %
of the group 3 soils had erodibilities between 0.01 and 0.026 and 70 %, of the
soils erodibilities < 0.01, most of the group 3 soils are overestimated by this
procedure.
soil 1: soil 2:
horizon depth permea- permea-
tcml bility bility class
A 0- I0 very hi gh 6
Bt l 0 - 0 ~nedium 3
Bt2 50-150 med~um 3
tllean perrneablllty ot soil: 4
horizon depth permea- permea
[cm] bility bility class
A 0-10 very h ~ g h 5
A/ B 10-25 high 4
Bt l 25-60 low 2 X 2
BC 60-150 medium 3
mean pertmeability of 5011: 3. 2
Tlrhlp 72-6: Corn1e,:cior, of K,,,,, ro K,,,,,, ti),- soils in gt.orq, I (old 2 (l/et-i~,cd
fro~lr ecluntiorl,~ ( 29) rulcl (30))
Kcq<,
0.00 1
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.0 1
0 02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0. I0
0. I I
0.12
0. I3
0. 14
0.15
0.16
0.17
0. I8
0. I 0
0.20
0.2 1
0.22
0.23
0.24
0.25
0.26
0.27
0 28
0.29
0.30
group 1 group 2
0.122 0.00 1
0. 125 0.002
0. 127 0.003
0.129 0.004
0.132 0.006
0.134 0.007
0. 136 0.008
0.138 0.009
0. 14 1 0.0 10
0.14 0.0 1
0. 17 0.02
0. I9 0.03
0.2 1 0.04
0.24 0.06
0.26 0.07
0.28 0.OX
0.30 0.09
0.33 0. I0
0.35 0. I I
0.37 0. 12
0.40 0.13
0.42 0.14
0.34 0. 15
0.47 0.17
0.49 0. 1 8
0. 5 1 0.19
0.53 0.20
0.56 0.2 1
0.58 0.22
0.60 0.23
0.63 0.24
0.65 0.25
0.67 0.26
0.70 0.28
0.72 0.20
0.74 0.30
0.76 0.3 1
0.79 0.32
0.8 1 0.33
K,,,,,
0.3 1
0.32
0.33
0.34
0.35
0.36
0.37
0.38
0.39
0.40
0.4 1
0.42
0.43
0.44
0.45
0.46
0.47
0.48
0.49
0.50
0. 5 1
0.52
0.53
0.53
0.55
0.56
0.57
0.58
0.59
0.60
0.6 l
0.62
0.63
0.64
0.65
0.66
0.67
0.08
0.6'1
K,,,,,
group 1 group 2
0.83 0.34
0.86 0.35
0.88 0.36
0.90 0.37
0.93 0.39
0.95 0.40
0.97 0.4 1
0.99 0.42
I .OO 0.43
1 .OO 0.44
1.00 0.45
1 .OO 0.46
I .OO 0.37
1 .OO 0.38
1 .OO 0.50
1 00 0.5 1
1 .OO 0 52
1 .OO 0.53
l .00 0.54
I .oo 0.55
1 .OO 0.56
1 .00 0.57
I .OO 0. 58
1 .OO 0.59
1 .OO 0.6 1
I .OO 0.62
l .OO 0.63
1 .OO 0.64
1 .OO 0.65
1 .OO 0.66
1 .OO 0.67
1 00 0.68
1.00 0.69
1 .OO 0.70
1 .OO 0.72
I .OO 0 73
I 00 0 74
1 .OO 0 75
1 .OO 0.7t7
7.3 Thc topogl-aphic tnctol-
If the analytical data for the solution of the discriminant functions
are n o t available the experience that volcanic ash soils (Andisols) are often in
groi~p I , soils from acid basement rocks are frequently in group 2 whereas
soils from basalt and other basic parent rock are often in group 3 can be used
for a crude soil loss estimate.
7.3 The topographic factor (LS factor)
Soil enlsion is fiivoul-ed with increasing slope length and -gradient1 (cf.
Chapter 4.3). The slope length factor ( L) gives soil loss on a given slope
length relative to soil loss on the USLE unit plot. The hctor for gradient ( S )
gives the ratio of soil loss on any given slope t o that of a 9% slope. The
combined topographic factor (L'I'S) allows to adjust soil loss on a given slope
length, gradient and slope form to that of the control plot. It is calculated by
(Wischrneier & Smith, 1978):
with
or
1 \lope length (m]
rn slope length exponent 1-1
; 1 gradient [ " I
with \ gradient I c X]
Thc slope length exponent (111) depend\ on the gradient and i 5 \rnaller for
low {lopes than f or steep slopes (Table 73- 1 ).
' I GI-;ltlient can be ~~~c a s ur c d by illclinon~elel-a ( 3s \peci;~lly ccll~il,pccl compaisc.,. A icr.! ,irnl?le
tie\,~ce to Inca\lu.e \lope - length and - g~-;lrlicnt i i illust~-~~tc.tl i n Anrlcx 7. 1 .
On low slopes, m becomes smaller because low obstacles as rills
and clods (surfiice roughness) produced by tillage slow down runoff. Thus,
rnore water stays on the field for a longer time and water depth on the field
increases. Tinie for infiltration is longer and at least part of the soil surface is
protected against drop impact by a water layer. LS factors can directly be
read from Figure 73- 1 . In order to a?just for less splash erosion on low slopes
and thc PI-otective water layer, an additional correction of the annual erosivity
is proposed on low slopes in the successor model of the USLE, - the Revised
Universal Soil Loss Equation (RUSLE) (Renard et al., 1992). This correction
factor can be obtained from Figure 7 1-2.
gradient f % 1
<=0. 5
0.6- 1 .O
1.1-3 4
3.5-4.9
> = 5
An exponent In < 1 shows that soil loss increases to a smaller extent
than slope length. Nevertheless, i n contrast to erosivity, soil erodibility, and
slope-gradient, slope length can be influenced easily by Inan and is an
important parat~leter for soil loss reduction. Slope length in the USLE is
defined as the distance fro111 the point where runoff begins to the point where
deposition occurs or where runoff enters a well-defined channel (Wischmeier
& Smith, 1978). As demonstrated in Figure 73-2, the lower slope end may be
presented by a small ditch or ridge along a field bot-der, a road ditch or a
drainage channel. In case of small rivers, the slope end generally does not
correspond to the river border because deposition generally starts earlier-. The
upper slope end can be formed by the watershed boundary or by ridges.
channels or deposition zones which limit a slope above. In general, the
definition of an upper slope limit is met if no runoff from slope seg~nents
above enters the slope.
m
0. 15
0.20
0.30
0.40
0. 5
7. 3 ' fht: topographic t act or
slope - length [m]
Kunofi' volulne and velocity increase along the slope. This causes an
increase of' soil loss pel- unit area with increasing distance down-slope. I n
order to calculate soil loss on a segr-nent of the slope, the slope is divided into
a small number of segments i with equal length and approximately equal
gradieiit. The segment on top of thc slope corresponds to i = 1 . The ratio of
soil loss on each segmenl to soil loss of the total slope can be described by:
with Ai relative \oil lox\ of segment i I-]
I (egmcnt nurnber
N nurnber of segllients with cqi~al length
111 \lope length exponent
On a uliiform \lope of 6C/( ( m = 0. 5) , for example, which wa\
divided into 3 \egments of equal length, the upper segment would provide
19%. the middle and lower 5egment 35 and 46 C7r of the total coil los\ on the
\lope. Ke\ul t s of ecluation 35 for different \lope exponents and \egmcnt
numbers are glven in Table 73-2.
A\ \o11 lo\\ I \ not equally distributed alol~g a \lope, \lope torm a\
well determine\ \oil lojs. On a concave \lope, the up- and inld-\lope part\ arc
\teeper than the foot-\lope wherea\ 011 a convex \lope the foot-\lope ha\ a
hrghcr gradlent Thc large runoff volutne which arrive\ down-\lope meet\ a
low gradient on the cot~cave but a h~gh gradient o n the convex \lope.
Submitting the \amc average gr-adlent, \oil lor\ on convex \lope\ I\ .
therefore. Inore \evere than on concave \lope\.
7~1h10 7.5-2: Soil lo\\ of \lol~e \r~/lzollt, u,itll oylrcrl /e/lgtll or1 1~/11fot-111 \lope(
/'rltrtr\~> to \or/ 1 0 5 s of t o t ~ ~ l \Iol?o 10)- L / I ~ P I - C ~ I ~ ~ / ~ L ~ / ~ T / ? P I - of
\cJglllcilr 5 clrlc/ c/iffe/-o/lt slope C ~ ~ J O I ~ C I ~ ~ ~ ( 1 x 1 ~ ~ 0 O I Z o~ll/(ltro/l (35))
Determirzation of the LS factor:
Rcad the LS lactor for un~l or m slopes fTrom Figure 71-I?' . In order to
correct soil loss for the effect of slope form, an irregular slope is d~vi ded into
:t small number of equal length segments with approximately unilorm
gradient. LS values l'or each segment are cho\en from F~gur e 73-1 by sing
the slope length of the entire slope and the gradient of the segment. The \o
der ~ked LS value\ are wc~ght ed by niultiply~ng them with the value\ from
Tablo 73-2. Summation of the prod~1ct5 gives the LS f ~ c t o r for the whole
\lol'e
number of
segments
3
3
4
5
Exa~npl e:
A 60 rn long convex slope is divided into three 20 In long segment\
wrth unifol-m gradlent of 10, 15 and 20% for the up-, mid- and down-dope
\cgment (segn~ent \ 1.2 and 3 in Table 73-3). I-e~pectively. The LS factor for
each \egrnent i \ cho\en from Figure 73- 1 by uslng a \lope length of 60 ITI and
rl~c gradient of each \egnlent (column 3, Table 73-3):
27 ,I C' OI I \ C' ~~~I OI I table Stom degree$ t o ~)c'l-cerit 14 y l e n In Annex 7.1
segment
number
I
2
1
3
-
3
I
3
-
3
3
I
3
-
3
4
5
slope exponent
m=0.5 m=0.4 m=0.3 m=0.2 m=0.15
0.35 0.38 0.4 1 0.44 0.45
0.65 0.62 0.59 0.56 0.55
0.19 0.2 1 0.24 0.27 0.28
0.35 0.35 0.35 0.35 0.34
0.46 0.43 0.4 1 0.39 0.37
0.13 0.13 0.16 0.19 0.20
0.23 0.24 0.24 0.25 0.25
0.30 0.29 0.28 0.27 0.27
0.35 0.33 0.3 1 0.29 0.28
0.09 0.1 1 0.12 0.14 0.16
0.16 0.17 0.18 0.10 0.19
0.2 1 0.2 1 0.2 1 0.2 1 0.2 1
0.25 0.24 0.23 0.22 0.22
0.28 0.27 0.25 0.23 0.23
Thew LS factors are weighted by the values from Table 73-2 for a
\lope 2 5 74 (111 = 0.5) and 3 segments (column 4 in Tablc 73-3). The product\
of all \egriients (column 5 ) are \ul-nrned up and give the LS factor (= 3.1 1 ) fol-
the ilope. Thi\ means that on a soil on this slope, \oil loss would be 4.11
tirile\ the (oil loss of the same soil on a 22.1 m long \lope of 9 C/r.
1
segment
I
2
3
Soil erodibility change\ on a \lope can be con\idered by thc \ame
prc~cedure. Soil erodibility for each segment (colurnn 6) I \ multiplieci w~ t h the
weighted LS factor\ Sor the segments ( col u~nn 5 ) wh ~c h give\ a KLS t'actor
for the {lope of 0.7 (column 7).
Change\ of the crop and management factor are dealt alike a\ long a i n o
deposition i \ induced by thc changes.
2
The slope length f ~c t o r also allows the calculation ol' a ni axi r n~~m
length if the maximum tolerable soil loss (T) is known:
If LS i \ known, the maximnm length can be chosen from Fiiguse 73-1. A
tolerable LS value of 2 on a 10% slope, for example. yields a maxilnum
slope length of 65 m i n order to keep soil loss within the tolerable lirniti.
3
gradient
I0
15
20
4
LS
factor
1.92
3.53
5.33
5
weighting
factor
0 I9
0.35
0.46
{urn:
6
corrected
LS factor
0.37
1. 23
2.50
4.1 1
7
K factor
0.02
0.13
0. 2 1
corrected
KLS factor
0.007
0. I h
0.53
0. 70
7 4 The co\ er ancl rnan,~gerllcnt (C' ) I Ci c~t or
7.4 The cover and management
The cover and management C of the USLE gives the ratio of soil loss
on a cropped plot to soil loss on a barefallow control plot of identical size,
slope length, gradient and soil. In contrast to the barefallow control plot
where soil loss per unit erosivity (= erodibility) is supposed to be a constant
(see Chapter 7.2). soil loss on a cropped plot is subject to changes over the
year which depend on crop growth and rnanagement. After planting, the
growitig canopy increasingly protects the soil surface while litter fr-om
senescent parts l'alls 10 the ground and forms a tnulch layer. The weeds in the
crop stand develop additional canopy covet- and act as mulch after weeding, i f
left in the field. The protection of the soil surfi~ce depends o n the amount and
cli~ality of coverage. Both are crop and management specific.
Howcvel.. ;in uncovered soil surface is only endangered if erosi\,e
storms occur. Therel'ore, in order to calculate the influence uf crop cover on
soil loss. the distribution of erosivity during the year must also be considereti.
As thc annual erosivity distribution is site specific, the salne cropping system
will cause different soil loss at different locations because of different
distribution of erosive rains. The incan annual erosivily distribution is the11
assigned to thc different crop stages (Table 74- 1 ).
- -- -
crop stage
crop (scedbccl to germination)
I0 (/( canopy cover until SO(;/( cover (ectablishrnent)
50 ' X to 75 (k canopy cover (development)
75 % canopy cover to harvest
h;u-ve\t 1 0 next plowing or seeding
description
F S B
SB - 10
For each crop stirgc ( i ) a coil loss ratio (SILK) i $ cnlculatud ac \oil
Ios\ of thc cn,pprcl plot (Acrop,) relative lo soil lo\\ of tlre control plot
(Ahare, ) during the same pel-iod:
rough fallow ( F) al'ter primary tillage (coarse tilth) to
seedbed (SB) preparation (= secondary tillage; fine tilth)
;~fter. seedbed preparation ~lntil 10 74 canopy cover of' the
Chapter 7
Acrop,
SLR, = -- I - ]
Abare,
The soil loss ratios indicate the degree of soil protection by a
specific crop stage. They are independent or site specific climate.
In order to avoid short term soil loss variations, the longterm mean
soil loss of the barefdllow is used instead of the actually measured soil loss. I t
is calculated by:
Abare , = R, : K (tlha)
(38)
with K soil erodibility [t"'h/N'l'ha]
R1
mean erosivity during crop stage i [Nlh]
The term (R, 'I: K) give\ the mean soil loss of the barefallow control
plot during crop \(age i . In order to reflect the site \pecific ermivity
distribution, the erosivity during crop stage i relat~ve to the annual erwivity i \
calculated:
Rrel, =
RI
R
I - I
with Rrel, proportion of annual R (relative erosivity) during
crop stage i [N/h]
R mean annual erosivity [Nlh]
The \oil loss ratio\ for each crop stage of a rotation are multiplied by
thc corresponding Kreli'\. Summation o f the products and \ub\equent
division by the duration of the rotation rcsults in an avcragc annual C factor:
with n number of crop stages i per year j
t duration of the rotation [a]
7.4 The cover and management (C) fitctor
An example for the calculation of a groundnut - maize rotation is
given in Table 74-2. Mean planting date of groundnut and maize was 15th
March and 5th August, respectively.
The crop stage duration is taken from the growth curves of the various
crops (Annex 3. 3). The Rrelils (column 4) are obtained from the mean annual
distribution of erosivity (Table 74-3). Alternatively, the erosivity , - distribution
can be estimated by calculating the relative rainfall distribution- .
The C factor is calculated by summation of the product of the relative
erosivity (colurnn 3) times the soil loss ratio (colurnn 4) for each crop stage
(column 5 ) .
" Estinlation of the el-osivity distribu~ion from I-ainfall clistribu~ion I'or 18 \tations in Cameroon
I - ~ ~ L I I ~ c ~ i n a mean ancl maxiln~lm crrol. of 1.3 ancl 13.6%. respectively (Bresch. 1993).
3 4 5 6
I . relative soil loss column 4 * 5
erosivity erosivity ratio (Ci)
[relative to (Rrel,) (SLR)
mean annual] I-] [-I
0.06 0.0 1 1 51 0 03
0.12 0 07 0 63 0 04
0 13 0. 02 0 02 0 00
0.24 0 10 0 07 0 .OO
0 54 0.29 0.03 0.00
0.49 0.06
0.68 0 14 0 56 0 O X
0 70 0 I I 0 51 0 06
0 84 0.06 0 32 0 02
0 99 0 15 0 05 0.0 1
1 05 0.05 0 00 0 00
0 51 0.17
1 0 23
column 1 1 2
gronnd-tint
make
crop stage duration
[dl
S13 - 10 16
10-50 18
50-75 7
75 - H 3 7
H-SH 5 5
SR - SR 143
SR - I0 3 I
10-50 I9
50 75 8
75 H 44
H - SB 120
SB S B 222
Total 365
Chapter 7
High contribution to the C factor results from crop stages where
little surface cover coincides with high erosivity: This was the case for crop
stage SB - 10 of maize which received 14 5% o f the annual erosivity (Rrel, =
0.14) in Table 74-2 in a state of little cover. 35 % of the annual soil loss
(column 6 in Table 74-2: 0.0810.23) occurred during this period. The soil loss
ratios are generally high during the initial crop stages when cover is poor. An
SLR > 1 for crop stage SB - I O of groundnut (SLR = 1.52) signifies that soil
loss on the cropped plot exceeded the mean soil loss of the barefallow plot
during this crop stage. This was due to compaction of the cultivated plot
during the planting operation and sealing by early rains. On the barefallow
plot, seals after a rain are raked (per definition) and no planting takes place.
The duration of' the crop stages shows the fister growth of
groundnut which needed 88 days from seedbed (SB) to harvest ( H) compared
to 102 days for maize. Groundnut in the example received 49 Cr/c of' the annual
erosivity (sum RRi of groundnut (SB to SB) = 0.49) compared to 5 1'/( for
maize (sum RRi (SB to SB) = 0.51). The contribution of groundnut to the C
fiictor was 0.06 (SB to SB) which corresponds to 26 C?c coriipared to 74 54
(0.17) for make. Thus, in the example, groundnut was more protective for
the soil than maize. Protection mea\ures (e.g. mulch) would thu\ be more
effectively applied during ma i ~e cultivation.
7.4 The coccr and inanagerncnt ( C ) fiictoi-
~ I F , I I ~ , . C P ~ ~ S - ~ I F - ~ I ~ ~ C P X ~ S - ~ I ~ ,
I t , , kc c a c d c d a a G ~ . - r - P 1
r-1 rl rl P I CI r-I ri ri ri PI C I CI r~ rr 'cT; r-i PI cr rr rl PI CI cr rl rr TI PI rr rr rr
The difference of erosivity distributions is shown by the three sites
from Camel-oon i n Figure 74- 1 . On the coasi (Douala), the very humid ocean
cli~nate has rather uniformly distributed erosivity during 9 months. The dry
season lasts about 3 months. The inland of southern Carneroon (Yaounde)
has two distinct rainy seasons separated by a dry spcll whereas in the north
(Maroua) nearly all erosivity is concentrated in a few months.
To establi\h soil lo\s ratios for different crop\ and management
systems needs field rneasurcments which are co\tly and time consuming. Sol1
lo\\ ratios for the rn~ijor c ~ ) ~ s - ' in the USA have been experimentally
determined for a range of management options2'.
'' maize. aoybcans, coltona a~nall grain. iorghum, wheat. rycgraas, potatoes. paitul-t.. range uncl
icllc land ancl forest
'' plow. notill. chi\el plow. contour tillugc, \(sipcrop. ridging. with and \vjtho~~t I I I L I ~ C ~ or
rcsiduei
7.4 The cover ant1 management ( C) factor
In order to calculate soil loss for further crops and systems.
Wischmeier (1975) proposed to divide the influence of the cropping system
into subf'actors. He defined a s ubf ~ct or for:
1. the influence of the canopy cover (c 1 )
2. the influence of mulch or of' vcgctation close to thc soil \urliice (c2)
3. tillage and residual effects of the former vegetation ( c3)
The C factor is calculated as the product of all 3 subfactors:
For trop~cal countrie\, the \ubfactor niethod i \ especially valuable
becau\e for Inany cropr no experimentally determined data are available. A
further complication i \ the large variety of small holder sy\tetns which are
difficult to compare to American standard\ ( e . g hand till~tge, mixed
cropping. heaping and bedding etc.).
Data for the rubfi~ctor calculation also are often not available but can
rather ea\ily be collccted. The procedure for \ubf;ictor deter~ilination i \
\~~b\ecluently explained.
Subfactor
The intluence of canopy is calculated by (Foster, 1 982):
with CC, effective canopy coverage I-]
He cf'fective canopy height [rnl
The canopy height effects thc velocity of drops falling of'f' the leaves ant1
thereby the energy of the drop irnpact on the soil. As drops may be formed by
lower and higher leaves on a plant and drops from higher Ier~vcs may he
intel-ceptecl by the lcaves below, the effective canopy height is used which
represents an average valuc. For practical considerations, He is estitnated as:
with
H,,,,,,
mean height of the uppcrrno\t horizontal leaf of
the plants in a crop \tand [rn]
Thc second variable in equation No. (42) - canopy cover - enter\
also as effective canopy covei-. Dr o p which fall froi i ~ the canopy may not
directly hit the soil s~i r f ~i ce but may fall on mulch material underneath
without cau\ing \oil loss. Ac a cover fl-om ~nulcli is 111ol-c protective than
from canopy, the effect of mulch is conjidered to 100 C/( wherea\ only the
canopy cover with no mulch underneath, i.e. the effective canopy cover
(CCe), is taken into account. It i \ calculated by:
with CC canopy cover 1 - 1
MC 111uIch cover [-I
I f canopy cover is 80'/(, for example, wit11 a mulch cover of' 20' A. the
effective canopy cover is 0.8 :I' ( 1-0.2) = 0.64.
Subfactor Q
The intlucnce of lnulch cover (c2) can be calculated by (Yodel- et al.. 1992):
bquation No. (45), which rcllect\ the curve u\ed by Wi\chmeier &
Snlith (1978) glve4 a consei-vatrve e\tlmate of the mulch etfect.
Mea\urements by numel-ous other author\ (Duma\, 1965; Kaln7, 1989: NiII,
1993) revealed a hlgher efficiency (cf. F~gur e 74-2). Nevci-thela\, ~t I \
Indicated to continue u\ing eyuation No. (45) in order to arrive at a cautlou\
estimate of so11 los\ reduct~on by mulch. Some stmple iiicthod\ for \oil cover
measurements are illu\trated and explained in Annex 3.2.
7 4 Thc co\ cl. and managcmcnt (C' ) 1';11,1o1-
Fi,qli/-o 74-2: / ~ ~ / / / i o ~ i c ~ o o f 111itlc.11 011 . $oi l 1 0 , s ~ (/,\ o \ ~ / l / i ( ~ t e ( / 173, ~ l i ~ o ~ - ~ i i t
cl/itllor..\. Sl l l ?f i l c' t o~- c.2 gi\1c>s thc /*[rtio of' \oil 1os. s 0 1 1 ( I
~ ~ o \ ~ ~ ~ / - ~ ~ ~ ~ ~ l l o t to (111 /i11~*0\1Ol"c~d / l I Ol .
mulch cover [%I
Subfactor c3
Not much data are available to determine subfactor c3 for tropical agl-o-
\y\tem\ which account, for the re5idual effect of the prcviou\ vegetation.
Ou. n mcasuremcnt\ re\ultcd in an average c3 of 0.8 for the 1st year after
11)rc\t I':illow and 0.4 after pl-ajs fallow (Table 74-4). A mean c3 01. 0.67 for
the first 2 ye:lrs after gra,\ fallow can he c\timated from data of Kilewc &
Mbuvi (1987) by the ratio of e md~hi l ~t y during the l i nt 2 year\ and
clr)dihility of thc 3rd to 5th year. For practical purpo\c\. the c3 value\ in
T;ihlc 7 1 % are proposed. The influence of. the gl-ac\ fallow re\iduc\ come\
very clo\e to the residual effect\ dew-ibed by Wischmeier & Smith ( 1978)
for turned \od. For the first year after plowed grassland they propo\ed 0.4.
0.45. 0.5 and 0.6 for crop stage\ SB - 50. 50-75. 75 - H and H - SB.
re\pectively. The same cnlp \tages during thc second year were weighted by
0.8. 0.85, 0.9 and 0.95.
Figl,,.e 74-3: Sl,hfictor c . 1 a\ i17/lu~,rr>r~l 17~3 efi~c-ti~~~ crrr~ol)\, collri- liiril c . f i ) / ~
height ([1/i et. Fo~trr-, 1982 rrrlll Wi.$chtrlrier, 10751
effective height
effective canopy cover [%I
7 4 The cover and managetnent ( C) factor
At the mornent, not enough data are available to calculate SLRs for
the multitude of' tropical cropping systel-rls. Nevertheless, soil loss c~u1 be
esti~nuted by the available data. In most experiments pi~blished in literature,
soil loss was measured o n a cropped plot and cornpared to soil loss on an
act.jacent control plot. Such data supply soil loss values for single cropping
seasons or years without considering different crop stages. C f'actors which
have a high variability due to a low number of repetitions can be calculated
fro111 such data. However, some crops have been lested in several
experiments and by comparing and averaging the results some reasonable
ti-ends can be observed.
fallow type
fore\t
- -
gsa\$
- -
- -
Such annual C factors from ciif't'erent locations include an unknown,
\ite specific variation caused by the el-ojivity distribution which can not be
accounted for. By using them in different sites, the same soil loss will be
predicted irrespective of the site specific erosivity distribution.
&st K factor c3
[t*h/N*ha]
[-I
0.0 105 0.0 135 0.78
0.0886 0. I I00 0.8 1
0.0 1 15 0.0236 0.49
0.0660 0.2000 0.33
0.1620 0.3450 0.47
To estirnate the error caused by ignoring the erosivity di\tribution, C
factors were calculated for a mixed cropping sy\tern measured in Cameroon
by using erosivity distribulion curves from cites with an annual erwivity
between 750 and 3231 N/h and mono- and bimodal rain distribution. The
maximum difference was \r-nall (16'k ) (Table 74-5) (Petri. 1902).
Furthermore, the liriiited ecological range of' 111o\t crops will also contribute
to keep the difference within certain limits becauw very large differences in
climate are generally also accompanied by a change in crop\.
The systern rain - canopy cover - soil loss can be regarded as sclf-
stabilizing within certain limits. More rain after seeding or germination will
enable faster and more growth provided that water is a limiting growth f'actor
as i t is in many regions at the onset of the rain. Sucli an auto-regulation also
favours similar annual C factors despite site specific differences in ternporal
rain distribution. However, some crops can be found in very contrasting
climatic zones. Groundnut and maize, e.g., are as well planted in the
teainforest as in much drier environments. In this case it is safer to choose a C
factor which was measured in a climate comparable to the site for which
calculations shall be carried out.
site mean mean annual C factor
annual rain erosivity
Bamonda 2470 1395 0. 26 1 04
Bafia 1370 XI8 0.29 116
ecological zone
hulllid rainli~rest
humid highland
hurnid savannah
savannah/ forest
transition
- -
Determination of the Cfactor
As previously clescribed, C factors can be derived from available
cxperimental data or be calculated by using subfilctors.
I. Ilerivation of C factors from experimental data
Choosc a table from Tables 74-8 to 74-1 8 according to the main
crop and look for a similar management system as your own in the
descriptions:
table no.
Ttrhle 74-8
title
C /tlc.torcfila forest, hlr $11 errlel gnis 5 \ ~ g -
c2tcltlorl (fclllon>~, p~r\t~rre) crrlcl \rthf(rc.tor, for
rr.5 iclralrl ef ec- t ~
Etart1171c. of trltrrrlcrti\~e ~~o t h o t l fir deter-
171inlrtioll of C f c~~t or for. the I st \ ucrr- for grtl,\ \,
c - ~ \ ~ c . t - c 1-017s crt~d hri\/~ / ~ I I I o c L ' \
C f i r ( tors for. hcrtlcrtlcr
C' fcic-for\ f or /?lile~1/?/710
C ftrc tors foi c tr ,(r \ l r r
C fcic tot.\ foi. tur sc~rllcrrloorr,, ~) o- c j / ~~~i t i l c.r-ol~c
C fcrc.tot.\ fol 81-orrrltlr~rrt
C ftrc tot-, fol- tlltrr;e
C' fcrc tot-c. fot t?rrllc't NI I CI \oi.g/lro?l
C ftrc.tot-5 /oi ~l l ?l r i ~~~l r-lc.c
C' fcrc.to)-s for / l r r s c c~llrrrloo~ic ( r ol ) ~
Table\ 74-8 to 74-18 contain average values der ~ved trom t l ~c
detailed data in Annex 3.4 (= \oLlrce refer\ to the line\ In the Annex table\).
The detailecl C tactor\ given in Annex 3.4 are not advised for unexpericncecl
il\er\. They were included Tor people who seek Inore inf'ormation uncl in orclet-
to allow control and improvement of the data-ba\e and the derived value\ in
the u\er wction ' . If you doubt about what to choose. take a11 average v;ilue.
With wme routine. col-I-ection\ for tiiffel-ences between de\cribed
and own \y\tem can be applied. If, for example, your crop is e\pecially well
developed, a \lnaller C factor \hould be chosen within the range given a\
' extremes' .
If a notill option is not i ncl ~~ded in one of the management systems,
the C factol- for the clean tilled variant can be taken and multiplied by one of'
the values in Table 74-6 which were derived frorn data in Table 34-8Annex:
?(' 11' >oil hayc li~e~-a,ur-c a\.ail;tble oil the sub,jec~ u hich is not inclucled in lhe tablcs 01'
, ~l l l l ca 4.4. the auIl ~or\ would be g~- a~cl ' i ~l for intlicutions or- a cop).
Chapter- 7
If a certain mulch cover is maintained in your systcm, you can
choose the C factor for the system without mulch and correct i t by
multiplying with a mulch factor (c2) from Table 74-7.
no.
1
2
If two crops are planted during the year, two C factors must be
chosen from the tables. In order to arrive at an annual C factor, the two C
factors and the periods between the two cropping seasons rnust be weighted
according to the erosivity which they receive.
Example:
notill system
without residues
with re\ i due~
A rotation consists of groundnut which is planted during the first
rainy season and is followed by plowed maize. In the dry spell between the
two cropping seasons, the field is left to the natural weeds. The C factors for
each crop and period arc lnultiplied by the relative amount of erosivity which
falls during the respective period i.e. 30 ' %I of the annual erosivity falls during
groundnut cultivation, the dry season receives 10 %, maize 50% and the 2nd
dry season another 10% of the annual erosivity. The sum of all products
gives the annual C factor of 0.35:
C factor
man extremes
0.65 0.45 to 0.8 1
0.22 0. I to 0.4 1
literature (lines in
Table 34-8Annex)
mean of no. 1 to 7
mean of no. 8 to 12
product
0.1 17
0.0 19
0.195
0.0 19
0.350
period
gr ound~~i ~t
dl-y \enson
rnai7e
dry seawn
total
C factor for
single periods
0.39
0.19
0.39
0.19
relative
erosivity
0.3
0.1
0.5
0. I
1 .O
7.4 The cover and management (C' ) factol
I n order to judge a systcrn, not only the cultivation period i \
regarded but thc whole rotation which includes the fallow period. I f in tlie
above example the groundnut-mai~e year i \ followed by two years of' grass
fallow, the annual C Pactor is (0.35 + 0.19 (Table 74-93, line 2) + 0.004
(Table 74-9a, line 3))/3 = 0.1 8.
11. Derivatiorl of Cjactnrs by subfactors
The C factor can be calculated from subf'actors by:
with subfactor: c l influence of canopy cover
c 2 influence of nlulch cover
c 3 residual intluence of former vegetation
I n order to derive c I to c3, the following information i s needed:
1 . the canopy cover curve and the canopy height to calculate \ubfactor
cl (equation No. (42))
1. the ~nulch cover curve for s~lbfactor c2 (cquation No. (45))
3. the residual influence of the fernier vegetation
3. the relative distribution of the annual ero\ivity
l'hc influence of notill can additionally be considered by lnultiplyi~lg with
the notill subfactors in Table 74-6.
D I . The canopy curve is either deterini~~ed by imeasunng canopy coverage
for the \y\te111 (methods in Annex 3.2) or by u\ing the typical growth
curves given in Annex 3.3. However, it \hould be kept in mind that the
variability included in the mean growth curves due to growing condition9
ai d cultivars may be appreciable. Calculations and ~neasurenients can be
crrrried out for crop stage periods (Table 74-1) or with a finer resolution
i.e. 10 day or weekly intervals. The effective canopy coverage is
calculated by equation No. 44. The canopy height can be rneawred or
estimated from experience and is used to calculate the effective height by
equation No. (43). With the effective height and the effective covcr
si~bt'actor c 1 can be read f'rorn Figure 74-3.
D 2. Mulch coverage is determined frorn the mulch cover curve which
shows mulch cover in the cropping system during the year. Subfactor c2
can be directly read from Table 74-7.
C> 3. Use c3 I'rorn Table 74%.
C : : > 3. Calculate the mean relative erosivity distribution fro111 as tnany years as
available (as in Table 74-3). 11' no erosivity data arc available. Llse the
relative rainfall distribution. Gcnerally. mcan curves are calculatecl by
~i~jcraging weekly or I 0 day intervals for as many years as possible.
mulch coverage
[%I
0
subfactor c2
[-I
1 .OO
mulch coverage
[%I
50
subfactor c2
[-I
0.28
7 4 The co\ er :und management ( C) factor
7irhlr 74-8: A\,~r(rgr C t o . o o r hrr~h orld gf-lrr \ )e~ot ~l f i ot z
(frrllo~rs. p ~ r ~ mr r ) lrrrd .\uhfirc-ior.s,/i,r tlrrir residrrrrl qtfr.ct.\
& - Z k b
c3 0 G
.c, rl 2
ZZr E
n:&
* g i a J
c - y 5
* = c
- 2 "Go
E G&-LI+
gc! yt-
.-
>, s- m -
- y = , S S
.s ' ; : >
ZSC gs
",' 5 1 1 '3
- ' - -
c /
" E*? ";!
c U c ' =
c3 -a.=
o '2 A A ,-
, - r S
a ? , &, zS
Y, P La-=, G
g c 3 u . c
- c 3 ; : g
" 2 p OJ u
S U 3 L E
La LC f 5
,z 2 '" ?J
'3 c
,z O E. - 4'
>, aJ - . -
% Z + . L
2 - 0 2 5
O C C 0
E5 " - .
7;rI)Io 73-9: E. I - ( I I ?I ~/ ( ~ o/ c117 c ~ / t ( ~ ~ - ~ ~ c r i i ~ ~ ~ I ? I P ~ / I O ( / f i ) t a t11e ~Ie~t~~1-11li11c/tio11 of' cr
C' filc' tor- for- t11c 1 \t I ' O ~ Z I - of'gru.\.\, c-ollel- c'l-op,\ crrlcl hlr\h flrllo\c
coverlperiod duration erosivity SLR weighted SL
during period (SLR*relative
erosivity)
[dl LN/ ~ I 1-1 1-1 C-I
I SLI 111 365 1450 / 1 0.24 I
Thc C fiictol- l or the 1 st year i \ the \ u ~ n o f thc weighted SILK'\ (0.24).
description
I Iea\cx plncccl around rrunhx :unci on
conlour; \pacing 5 x 3 m o n contour
, .\Itcrnatively I'or :I young plantation
I xt ycar)
2 :I\ aho1.c hut \pacing 7 x 3 m
3 ax above hut \pacing 3 x 3 m
1
4 a\ a how hut spacing 4 x 3 111
5 LL it11 con~plctc ITIUICII cover
I C factor 1 literature
mean
I
extremes (lines in Table
34-2Annex)
I
For- other- \pacings/ den\ities bctweerl 5 x 3 m (= 660 plantdha) and 2 x 3 111
( = 1650 plantslha) C' can bc taken from Figure 74-4.
7.4 'The coves and riiun:~gemcnt ( C) I.;~ctol.
Fi,qlrt,r 74-4: C' ftrc.fot- for- diffi)r.clrlt Ocrtlcrrltr t/tvl.\itie.c.
Chapter 7
*

1

l
e
v
e
l

g
r
o
u
n
d

=

n
o

m
o
u
n
d
s

o
r

r
i
d
g
e
s

w
e
r
e

f
o
r
m
e
d

3

C

f
a
c
t
o
r

m
e
a
n

e
x
t
r
e
m
e
s

0
.
3
6

0
.
1
2

t
o

0
.
5
6

0
.
1
8

0
.
0
2

t
o

0
.
5
9

0
.
6
3

0
.
5
7

t
o

0
.
6
9

d
e
s
c
r
i
p
t
i
o
n

p
l
a
n
t
e
d

o
n

l
e
v
e
l

o
n

l
e
v
e
l

g
r
o
u
n
d
:

l
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

m
a
i
z
e
.

r
i
c
e
.

c
o
w
p
e
a
.

s
o
y
a
,

p
h
a
s
e
o
l
u
s

o
r

c
o
c
o
y
a
m

f
o
r

i
n
t
e
r
c
r
o
p
p
i
n
g

w
i
t
h

m
a
i
z
e

w
.

l
i
t
e
r
a
t
u
r
e

(
l
i
n
e

i
n

T
a
b
l
e

3
4
-
4
A
n
n
e
x
)

m
e
a
n

o
f

n
o
.

1
-
5

m
e
a
n

o
f

n
o
.

1
6

t
o

I
8

a
n
d

2
0

t
o

2
6

n
o
.

2
7

c
a
s
s
a
v
a

1

2

3

m
o
n
o
c
r
o
p

i
n
t
e
r
c
r
o
p

s
u
b
f
a
c
t
o
r

Chapter 7
Chapter 7
7.4 The cover and Inanagetncnl ( C) t l ~ct o~-
Tlrl?lr 73- 18: A ~,r.r-c~grd C,firc-tors for rlli~c-ellnneous c.r-op.\
no.
I 1 bean and jack bean 1 1
C factor
man extremes
0.43
-
0.27 0.47 to 0.16
crop ' description
planted as monocrop
o n contour 011 4 out of 8
Balnbara nut
bean\
3
literature
lines in Table
34-7Annex
plowed: 3.5% {lope; jpacing
30 x 30 cm
data from mung bean, red
cabbage
no. 1
no. 2 to 6
4 /chili I 1 0.33 - 1 no. 8
1 different Indone5ian soil types 1 0.6 - 110. 7
5
6
7
8
no. 18 to 20
planred along \lope
notill. plu\ re\tdues of f.01-rner
I rnai7e. planted along slope
1
cotton
- - -
cowpea
- " -
0.005 0.0004 t o 0.02
planted along slope
211d cycle
plowed: without residues
plus residues ol' former maire;
10
I I
11
13
I
I residues humed: soya 1-esiducs I
I
0.29 -
0.5
0.24 0.2 1 to 0.27
0.06 0.002 to 0.28
In\h potatoe
lemon ~ r a \ \
1 4
15
16
17
no. 9
no. 1 I
no. 12 and 13
no. 14 to 1 7
papaya
\oya
surficially incorporated
rotation a\ ilbove but with notil I 0.04
-
-
- " -
sweet potatoes
tobacco
wheat-soya
incorpol-atecl
as above but all residue\
no. 33
wlthout cover crop
0.05
-
no. 34
0.22 -
0.434 -
notill without residue\
-
2nd cyclc
rotation on 12% slope. wheat
20 / wheat -m; ~i ~e / a\ :ibove, conventionally tilled.1 0.1 - 1 no. 35
no. 2 1
no. 22
2.1
-
0.26 0. l to 0. 4
no. 23
no. 24 to 27
0.103 -
0.23 -
0.5 -
0.1 13 -
no. 36
no. 28
no. 29
n o . 30 and 3 1
no. 32
residues incorporated
as above but notill 0. 0 14 -
yam
- -
(residues maintained)
on heaps
on heaps; intercropped; with
residue ~iiulch
0.23 0.16 to 0.8
0.07 0.04 to 0.09
no. 38
no. 39
7.5 Thc cflicr of protective methods - Support practicc factor ( P )
7.5 The effect of protective methods -
Support practice factor (P)
Protection measures must be adjusted to the possibilities and resources of
each farmer. For nearly each individual situation a set of suitable physical
and biological methods can assure sufficient soil protection.
7.5.1 Contouring, contour-ridging, tied-ridging
Generally speaking contouring means that all tillage operations and planting
are curried out across the slope. Contour tillage and planting with mechanical
tools leaves a roughness of the soil surface that is oriented across the slope.
This may be considered as micro-ridges on contour. Such a formed surface
redirects and retards the surface runoff. The efficiency depends on the degree
of roughness (ridge height), the side slope of the tillage marks and the
gradient of the overall slope. There is no clear limit between contouring and
contour-ridges or bunds. The latter could be regarded as extreme roughness.
Contouring in its original sense occurs ~ ~ n d e r mechanized tillage with crops
planted in rows. In handtilled systems only planting can strictly be achieved
on contour. Tillage with the handhoe is generally moving up-slope. The blade
of the hoe is placed on contour but no continuous roughness is created. There
is no information whether the roughness left by the handhoe marks can be
compared to contour tillage.
Contouring reaches its maximum protectiveness on slopes between
3 and 8% (Tablc 75 1 - 1 ). It is less efficient on slopes below 3% where runoff'
velocity is slow and a protective water mulch forms. On slopes above 8%. the
protectiveness declines as the water storage capacity of the ridges becomes
smaller with increasing gradient. For slopes > 2574, no protection is reached.
P factors which were calculated from recent soil loss s t ~~di es (Table
4 I - 1 Annex) support the values in Table 75 1 - 1 :
: ' I The 111akimum hlopo Ienyrh lnay be incsoascd by 7Sfk i t ' se\ i di ~c co\ cs ;~l' tt' ~- planting
reyul ~~sl y exceed\ SOc/ (
slope [%I
1 - 2
3 - 8
0- 12
13- 16
17-20
21 -25
>25
A P f'actor of I for slopes > 25% was based on the assumption that a
typical 15 crii high ridge in mechanized systerns retains no rnorc water on a
slope of 25% (Foster ct al.. 1992). If the storage capacity of the ridges is large
enough to prevent overf ow, maximum slopc lengths need not to be applied.
As the effectiveness of contour ridges depends on their- storage capacity, it
must also depend o n storm size. I n locations with frequent I~irgc stor~ns.
contour-ing is less effective than in locations with smaller storms. Theref'ore.
thc 10 year storm volume is chosen for ridge design purposes (Foster ct ul..
1992). If thc furrows can only carry the rnaximurn 2 year storrn, length limits
arc applicable ( Wischmeier. & Smith, 1978).
The procedure to estima~c the intlucnce of ridges applied by the
USLE gives a rough estimate and does not allow to distinguish between
dil'fcrent ridge heights. Ridges. however, play a11 important role i n tropical
~igro-systeriis. A more refined estimation is possible by using the P factors i n
Figill-c 751 - 1 i ~sed in the Kcviscd Universal Soil Loss Equation (RUSLE)
(Renard ct a].. 10'32). The curves were calculated on the basis o f a 10 year
storrri of 86 to 190 111111. hydrol ~gi c soil groirp C - I and clean tillage for row
P factor for
contouring
0.6
0.5
0.6
0.7
0.8
0.9
I .O
'' hydl . ol ori ~~ soil g1-oi1p C includes xoil\ with low inl'ilts~~tion sate5 wllcn wet. ~no\ t l \ ~ i t l l
i ~npendi np luyors or ~noclcratly fine roxturc (IISIIA. 1073: SCS Nalional Enginerr-ing
I 1;untlhook )
maximum slope
length Em]**
122
9 l
(3 1
24
1 X
15
13
7. 5 The el'i'cct of ~,l.orccri\e ~nerliod\ Support p~.aclicc I'itcror ( 1 ' )
F~g~~t - cl 751-1: P jilc.for-s f o r . difor.rrlt ritlgo hc>igl~t\ 101- trt-clri\ r~,rfll 10 \.t>tit.
\to/.llls bct\c.cerl 86 L I I I ~ I 190 111111 (111ii / ?\ ' ~/ I - O~O, ~I ( . \oil qr-0111) C
( I . o rtrr- e f ( I / . , 19913)
CI-opi u it11 n o c ~ L ~ ~ I - tlnd mlnlmum roughneis (cover-rnanagemcnt condit~on
6) . I n Weit Af r ~ca iuch I0 year itorms are found approximately In the belt
h c t ~ e e n I h northern latit~ide (north Senegal, north Burkina 1-aio) and the
co~ri t line (Figure 75 1-7). For other areas no infonilation on the 10 year itorm
M ; ~i found.
I2nr areas with a lower 10 year storm volume (e.g. north of I6
Intitode) the I-idpc cfficicncy will he underestimated by Figulc 75 1 - 1. whereas
f-ol- ai-c;is with higher 10 year stol-m volume ;in overestimation is possible.
Regarding the soils. hydrologic soil gnliip C may be applied t o the Aridisols.
Alfisols. Inccptisols and VCI-tisols oi the semi-humid to semi-arid/ arid area.
For the LJl~isols ancl Oxisols of the humid to s c ~ l ~ i - l ~ u ~ ~ ~ i d areas thc cfficicl~cy
i s underestimated by hydrologic soil groilp C.
Chapter 7
Figrrro 75 1-2: Isoh?.ctc~ f or - tlrr 10 ?vor {t onn ~lolirt?~e (CIEH. 198.5)
Contouring and contour ridges are mostly not exactly on contour. In
practice, they have a side slope either accidentally or in order to evacuate
excess water. For side slopes < 0.5%, all soil is deposited in the furrows (cf.
Chapter 4.3). For steeper side slopes the efficiency of contour ridging is
reduced. P factors corrected for side slope effects (Table 75 1-2) W~ I - c
calculated by (Foster et al., 1992):
P, = P,, + ( 1 - P,,)
( \ , 1 5 , ) ".i
m it11 P, P I'actor I'or ol'l'-gn~de contouring
P,,
P l'actor for on- ~r at l e contouring
,
grade along the furrow\ (sine 01' slope angle)
\,
\teepness of thc I~und (sine of \lope angle)
Mea\urcd \slue\ for ridges are given in Table 4 1 -'Annex, line I and
2. The table al\o indicate\ the di\astrou\ effect of up- and down-\lope ridges
( P = 0 0 to 4.3). practical purpo\e\ a P factor of 2 can be ~ ~ \ e d for thi\
practice.
slope
[%I
4
8
12
16
uncorrected
Pfactor
I
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0. 1
I
0.9
0.8
0.7
0.6
0.5
0 4
0.3
0. 2
0. I
1
0.9
0.8
0. 7
0.6
0. 5
0.4
0.3
0.2
0.1
1
0.9
0.8
0.7
0.6
0.5
corrected P factor
side slope of furrows [%I
0.5 1 .O 1.5 2.0 3.0 5.0
1.00 1.00 1 .OO 1 .OO 1 .OO 1 .OO
0.94 0.95 0.96 0.97 0.99 I .OO
0.87 0.90 0.93 0.95 0.98 I .OO
0.8 1 0.86 0.89 0.92 0.97 1 .OO
0.75 0.8 1 0.85 0.89 0.96 I .OO
0.68 0.76 0.82 0.87 0.95 I .OO
0.62 0.7 1 0.78 0.84 0.94 1 .OO
0.56 0.66 0.74 0.8 I 0.93 I .OO
0.49 0.62 0.7 1 0.79 0.92 1 .00
0.43 0.57 0.67 0.76 0.9 1 1.00
1 .OO 1.00 1.00 1 .00 1 .OO I .OO
0.93 0.94 0.94 0.95 0.96 0.98
0.85 0.87 0.89 0.90 0.92 0.96
0.78 0.8 1 0.83 0.85 0.88 0.94
0.70 0.74 0.77 0.80 0.85 0.92
0.63 0.68 0.72 0.75 0.8 1 0.90
0.55 0.6 1 0.66 0.70 0.77 0.87
0.48 0 55 0.60 0.65 0.73 0.85
0.40 0.48 0.55 0.60 0.69 0.83
0.33 0 42 0.49 0.55 0.65 0.8 1
1 .OO 1.00 1 .OO 1 .OO I .OO I 00
0.92 0.93 0.94 0.94 0.95 0.96
0.84 0 86 0.87 0.88 0.90 0.93
0.76 0.79 0.8 1 0.82 0.85 0.89
0.68 0.72 0.74 0.76 0.80 0.86
0.60 0.64 0.68 0.7 1 0.75 0.82
0.52 0.57 0.6 1 0.65 0.70 0.79
0.44 0.50 0.55 0.59 0.65 0.75
0.36 0.43 0.48 0.53 0.60 0.72
0.28 0.36 0.42 0.47 0.55 0.68
1 .OO 1 .OO 1.00 1 .OO 1 .OO 1 .OO
0.92 0.93 0.93 0.94 0.94 0.96
0.84 0.85 0.86 0.87 0.89 0.9 1
0.75 0.78 0.79 0.8 1 0.83 0.87
0.67 0.70 0.72 0.74 0.77 0.83
0.59 0.63 0.65 0.68 0.72 0.78
7.5 'l'he cffect of protective methods - Support practi ce factor (P)
slope
[%]
16
20
24
28
uncorrected
P factor
0.4
0.3
0.2
0. I
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
I
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0. 1
corrected P factor
side slope of furrows [%]
0.5 1.0 1.5 2.0 3.0 5.0
0.5 1 0.55 0.58 0.6 1 0.66 0.74
0.42 0.48 0.52 0.55 0.6 I 0.69
0.34 0.40 0.45 0.49 0.55 0.65
0.26 0.33 0.38 0.42 0.49 0.61
1 .OO I .oo 1 .oo I .oo 1 .OO 1 .oo
0.92 0.92 0.93 0.93 0.94 0.95
0.83 0.85 0.86 0.86 0.88 0.90
0.75 0.77 0.78 0.80 0.82 0.85
0.66 0.69 0.7 1 0.73 0.76 0.80
0.58 0.6 1 0.64 0.66 0.70 0.75
0.50 0.54 0.57 0.59 0.63 0.70
0.4 1 0.46 0.49 0.52 0.57 0.65
0.33 0.38 0.42 0.46 0.5 1 0.60
0.24 0.30 0.35 0.39 0.45 0.55
1 .OO 1 .OO 1 .OO 1 .OO I .OO l .OO
0.9 I 0.92 0.93 0.93 0.94 0.95
0.83 0.84 0.85 0.86 0.87 0.89
0.74 0.76 0.78 0.79 0.8 I 0.84
0.66 0.68 0.70 0.72 0.74 0.79
0.57 0.60 0.63 0.65 0.68 0.73
0.49 0.52 0.55 0.58 0.62 0.68
0.40 0.44 0.48 0.5 1 0.55 0.62
0.32 0.37 0.40 0.43 0.49 0.57
0.23 0.29 0.33 0.36 0.42 0.52
1 .OO 1.00 1.00 1 .00 1 .OO 1 .O0
0.9 I 0.92 0.92 0.93 0.93 0.94
0.83 0.84 0.85 0.85 0.87 0.89
0.74 0.76 0.77 0.78 0.80 0.83
0.65 0.68 0.69 0.7 1 0.73 0.77
0.57 0.60 0.62 0.64 0.67 0.72
0.48 0.52 0.54 0.56 0.60 0.66
0.40 0.43 0.47 0.49 0.53 0.60
0.3 1 0.35 0.39 0.42 0.47 0.54
0.22 0.27 0.3 1 0.35 0.40 0.49
Determination of the P factor for contouring and contour ridging
Contouring and contour-ridging
a. For simple tillage and planting of row crops on the contour, use P factor\
in Table 75 1- 1 according to the slope. If contouring and ridges were
established with side slopes, enter Table 75 1-2 for correction.
Example:
For a contoured \lope of 14 % with a side \lope of 3 C/c a P factor of 0.7
was chosen from Table 751-1. The side rlope effect is con\idered by
entering Table 75 1-2 for a 12 C/c and a 16 (;/c dope ( P corrected = 0.85 and
0.83, re\pectively) and interpolating the two values to a 14% slope ( P
corrected = 0.84).
b. For ridges with a height of more than 1 0 cm, choose a P filctor according
to slope and minimum ridge height l'rom Figure 75 1 - 1 . Correct it for thc
effects of an eventual side slopc as cxpluined in a..
c. If ridges do not persist during the entire year but are rnounted, for
examplc, during the growing pesiod of a crop and levelled during harvest.
the P factor can not be fully credited. In this case only the soil loss ratios
of those crop stages are rnultipliecl with the P factor for ridges for which
the ridges are intact. For the crop stage periods without ridges P equals 1 .
Example:
Ma i ~ e is planted on level ground on a 10% slope. When canopy covcr
reaches I O C/(, 15 cm high ridges are mounted with a 1 (k \idc \lope. The
term C x P f'actor is calculated like in Table 75 1-3.
7.5 71'lie effect ol'pro[ective method - Support practicc factor ( P )
The erosivity ratios were taken from Figure 74-1 from the 'north'
curve a4suming that crop stage SB - 10 4arted on the 130 day. The duration
of the crop stage periods and the soil loss ratios were taken from Table 74-2
(column 2 and 5 ) . A P factor- of 0.36 corresponds to 15 cm ridges on a 10 C/c
\lope (Figure 75 1 - 1 ). The corrected P factor i \ interpolated fi-om Table 75 1-2
((0.59 + 0.54)/2 = 0.57). The I-esulting C x P factor is 0.075 which compares
to C x P = 0.063 if the ridge4 would be credited for during the entir-e cropping
cycle.
Tied contour ridges
6
C x P
column 2 x 3 x 5
0.0 1 1
0. 020
0.01 1
0.0 15
0. 0 1 8
0.075
Values in literature for soil loss with tied contour ridges range between 0.21
and 0.035 on slopes between 4.5 and 7 % (Table 41-2Annex). Ties between
the ridges have no effect i f the ridges are perfectly on contour. I f the ridges
have a side-slope, ties will stop or reduce the sideways evacuation of runoff'.
A reduction of the ridge efficiency due to side slope will therefore be much
less. For practical purposes it is proposed to choose a P fhctor as for contour
ridging from Figure 75 1 - 1 and to dismiss the correction for the side slope.
7.5.2 Bufferstrips
4 3 1
Bufferstrips are < I to several m large strips within fields mostly composed
of quick growing species or natural vegetation. They are laid out on contour-
i n orcler to decrease runoff velocity thereby causing deposition of suspended
sediment. The efficiency of bufferstrips depends on the quality of the strip
(stsip widths. vegetation density). ils age and its position on the slope. The
runoff which asrives at the upper bufferstrip end has a certain transport
5 2
crop stage
SR - 10
10-50
50-75
75 - H
H - SR
SB - SR
erosivity
ratio
0.02
0.07
U.06
0.51
0.36
1.00
corrected
P factor
1 .OO
0.57
0.57
0.57
1 .OO
Total :
soil loss
ratio
0.56
0.5 1
0. 32
0.05
0.05
P factor
1 .OO
0.16
0.36
0.36
1 .OO
capacity and sediment load. Runoff velocity and tran\port capacity are
reduced in the bufferstrip by the higher hydraulic roughness and friction
exerted by the vegetation. Additionally, part of the runoff will infiltrate
within the strip which has generally a higher infiltration rate than the adjacent
cropped soil. If the transport capacity becomes less than the sediment load,
soil i 5 deposited in the bufferstrip. However, if runoff leaves the strip on the
lower end, it rnay regain \peed and pick up new sediment from the cultivated
\trip underneath. Thus, the mo\t t'avourable case i \ . i f no runoff leave\ the
strip.
Planted bufferstrips generally do not reach their full protection
efficiency during the first rainy season 01- the first year while the plants' root
and canopy system is still establishing. P factors for the second year are.
therefore, often lower than for the first year. There is also some evidence, that
the efficiency of bufferstrips may decrease with increasing sedimentation in
the strip (Barfield & Albrecht, 1982). This will depend on the growth habit of
the strip vegetation (e.g. canopy or twig density close to the ground) and how
tist the vegetation can grow up and cope with a heavy sediment load. If large
amounts of sediment arrive at the bufferstrip, a small terrace will form within
a couple of years.
A special case of bufferstrip are the riparian bufferstrips along rivers
which prevent sediment entry. However, they do not prevent soil loss from
the slope above. An indication for the effectiveness of riparian bufferstrip5
with increasing strip width is given in Figure 752- 1 .
7.5 The el'1'ec.l of protective method\ - Supporr pl-acrice I.21cro1- ( P)
Determination of P factors for bufferstrips
F~glrr-(1 7-52- I : P flrc*tor\ for ripnr-irrrl l ?l i f o~~t r- i p\ o/ I ~ ~ / / C - ' I . C ~ I ~ ~ 11,illt11\
~I ~i . i \ ' ~' ~l f l . o~l ( I 8(jc lo/)^ C L ' / ~ / I 1/11 11~7/ 1~11/ 1 \ ~~l i l i l ol l t l o ~ l ~ l of
cScl. I t/~tl Oliffor lt>11gtl1 ( SCI I I I I I CI C~I . c4; A ~ i o r \ ~ t ~ ~ / l ~ l , 1992)
a. Bufferstrips
a. Systc~natic trials for the effect of hufferstrips are still deficient. Somc P
factors as calculated by the RUSLE are givcn in Table 752- 1 .
0.95-
b. Further P factors can be taken from the experililerltally determined P
factors i n Table 4 I-SAnnex for comparable situations.
buffer width [m]
0.75-
z
5 0.70
m
2 0.65
0.60 0.55
0.50
0.45
0.25
y=369*? -304 1
----.
\% i - 1 ! -
i i \
r _ _ _ - _ . . - - -
\
1 I I 1
j I l l , l I l l I
I
1 1 1 1
0
1 1 1 1 -
5 10 15 20 25
Ttrhlcl 752- I : P Jilc.to~-.s ,fol- hziffi.~-.str-ip.~ crs ( Y I I ( . L I I L [ ~ P C I 17). f / 7 ~ RIJSLE
(Fo.vtc11- c)tcrl., 1992)
I percent of slope I position of strip*' I P factor
' ' 1 kw example 0.3-0.5 means f l i ~ ~ t the strip start\ after 4054 of the \lope Icngth doun-
\lope and ends after half' 01' the slope lcngth
covered by strip
2054 ~n 2 \ti-lp\
10% in 2 \trip\
h. Riparian bufferstrips
P factors for riparian bufferstrips can be chosen from Figure 752- 1 .
0.4-0.5 and 0.9- 1 .O
0.35-0.40 and 0.65-0.70
7.5.3 Contour bunds and heaps
0.67
0.7 1
Results from trials indicate the different efficiency of stone-bunds and
earthen bunds. Runoff occurring on the uppermost side of LL field picks up
e lo city and sedi~nent. Arriving at the first bund it is co~npletely stopped by
:in earthen bund or slowed down by a stone bund. The sediment is deposited
i l l front of thc bund. Usi ng eat-then bunds, t he process is repeated between
first and second bund. second and third bund and so on. However, stonc-
bunds, which are permeable, allow a part of the writer to pass the bunii. This
water regains velocity and transport capacity on the lower side of the bund
and entrain new sediment in addition to the runoff' produced o11 the lower side
itself.
The ci'ficiency of earthen bunds is thus much higher in the first year
compared to stone-bunds (Table 4 1 -4Annex). However, the data indicate that
in the second year the effect of the two types becorncs similar. The stone-
bunds become less permeable due to sediinent which progressively fills and
clogs the inner space of the bunds. Thc earthen bunds apparently became less
ef'f'icient duc to holes which occur in the bund or to the lowering of the bund
by raindrop innpact or overtopping. The decreasing efficiency of the earthen
bunds also make higher rnainte~iance necessary compared to stone-bunds.
7. 5 Tlir rf'l.rc,t ol'psotc.ctivt~ metliod~ - Support ~?s;rctic~r t;rctol. ( 1' )
Ilctermination of P factors for contour bunds and heaps
a. Stone-hunds and earthen bunds
Thc P kictors in Table 41-4Annex can be used for first and subsequent
years in comparable situations. As a bund can be compared to a ridge a
sirnilat- slope influence on the efficiency of bunds is assumed. Therei'ore, i r is
~~r oposed to multiply the available P factors with the ratio of the P tictor for a
15 cm ridgc (Figure 751-1) o f a given slope to the P factor on a 3 C/c slope
(Table 753- 1 ). If' maintenance is regularly carried out on earthen bunds each
year. the P fzlctor for the first year can also be used for subsequent years.
11. Heaps
Not tnany data are available for the specific effect of heaps on soil loss.
7'hc \.cry variable influence is shown by the data in Table 41-3Annex. The
influence of' heaps depends on their arrangement on the slope (up and down-
slope. on contour. in cluinti~plcs), their sire and height which depend on slope
and top soil dcpth (cf. Chapter 4. 4). The data in Table 41-3Annex should
only bc used if tlie intluence of mounds is not yet included in the C f'actos
(e.g. for yam in Table 74-1 8 it is not necessary to use additionally a P factol-
for heaps)).
7.5.4 Ditches and terraces
slope
[%I
0
I
7
-
3
4
5
6
7
8
9
10
1 1
12
13
14
15
Hillside and drainage ditches (cl'. Figure 44%- I ) clecrcase soil loss by
reducing tlie erosion effective slopc length ( L fiictor). Thus, the clown-slope
acceleration of runoff and its concentration is controlled. Slope length i n the
LISLE is defined as that part of a slopc where no major deposition is
I n the case of drainage ditches, the sediment charged water \pill\
11-ecly illto the ditch. The slope length is the distance between the lowcr ude
of a ditch t o the ilppcr {ide of the next ditch. For a Fanya Juu type terrace (cf.
F~gure 448-1 ). deposition begins in front of the excavated ridge and \lope
Icngth is calculated from the lowei- end of the ditch to thc area where
dcpo\ition begins in front of the next terrace.
Terraces not only reduce slope length but al \ o gradient which i \
con\idered in the LS factor. For \loping bench terraces (cf. Fig~tre 449-2). the
ratio
C-I
2 12
1.67
1.21
1 .OO
0.79
0.70
0.6 1
0.58
0.55
0.55
0.56
0.57
0.58
0.59
0.6 1
0.65
slope
[%I
16
17
I 8
19
20
2 1
- 3 - 3
23
24
2 5
2 6
2 7
28
29
30
ratio
[-I
0.70
0.76
0.82
0.9 I
1 .oo
1 . 1 1
1.2 1
1.33
1.45
I .64
1 .X2
I .9X
2.15
2.27
-.- 3 55
width of the bench is considered as slope length for soil loss prediction. The
soil eroded frorn the bench reaches the toe drain where it is either deposited
or washed off into the waterway anti out of' the field.
The soil deposited either in front of the Fanya JLILI terraces, in the
ditches or in the toe drains is not yet lost ttom the field. It can be regarded as
di\trihutcd within the field. Excavation of the ditches will partly put it bach
on to the field. The amount of soil which i\ actually transported out of the
f'icld relative to the amount of soil erodcci is callcd the sediment delivery
ratio. It varies with the 51de-slope ot the ditches (Foster & Highfill. 1983)
from 0.1 for level clitchcs to I for ditche\ with a side-\lope of 1 L/c (Table 754- 1 ).
.,. I including tcrr;lces with i~ndergrounti outlet$
":1 froin Wi\ch~lleier Kr Smith (1978): all other values from
STIR = 0.1 ';:c7.h4g: g = \ide \lope 1'4 I
' ' ' 3 ner ero\ion may occur in the channel\ depending on flow hydraulich and crodihilily ol' lht.
ch:unl~rl\; if c1i:uinel crohion occur\ SDR > I
terrace grade [%]
clo\ccl out l et '
0
0. 1
0.2
0.4
0.6
0.8
0.9
> 0.9- 3
Determination of P factors for terraces
sediment delivery
ratio
0.05 2
0.10
0. 13
0.17
0.29
0.49
0.83
I .OO
-
Calculate soil loss A = R'''K*'L'''S"C'I'P I'or each terrace by using slope
lengths as explained above. The gradient is either the slope-gradient i n the
case of hillside ditches or [he bcnch gradient in the case of bench terraces.
P may be compmed ot contouring factor if tillage and planting are
carried out on contour (cf. Chapter 7.5.1 ) which needs to be multiplied by the
\eciiment delivery ratio fi-om Table 754- 1 .
Example:
A slope is divided into I0 reverse-sloped ten-aces 10 m wide and 100 m
long with a bench gradient of 5% and a side-slope of 0.4%. The benches are
cropped to cassava with nlaize arranged on contour. Further data ( R = 500
Nlh; K = 0.15; LS = 0.3 1 (from Figure 73- 1 ); C = 0.2 1 (from Table 74- 12)).
P SOI- contouring is 0.5 (from Table 75 1 - 1 ) and the sediment delivery ratio for
a 0.4% side-slope is 0.29 (Table 754- 1 ) . Thus soil loss for this situation is A
= 500 ;:: 0.15 *: 0.3 1 'I: 0.2 1 ' I: 0.5 $: 0.29 = 0.7 1 tlha. Each terrace has 0. 1 ha.
All terraces together would thus loose 0.7 1 t.
7.6 Soil loss tolerance limits
Soil loss tolerance limits define the soil loss rates which are tolerable in
order to maintain the soil's diverse functions during a specified time. The
ef'f'ect of' soil loss depends strongly (311 the type of soil. Soil loss always
implies a loss of nutrients and structural components (clay. organic matter)
which are enriched in the sediment. The soil profile is shortened. rooting
depth and water storage capacity decreased. On very deep, homogeneous
soils. the damage will be less than on soils with unfiivourable layers or solid
rock close to the surface. Cornpared to less weathered soils of the temperate
and semi-arid zones, loss of surface soil is more severe on highly weathered
soils whose nutrient storage and availability depends largely on the organic
tnatter of the s ur f ~ce soil while the subsoil fertility is low.
The yield decline associated with erosion depends also on the crop.
Mbagwu et 31. (1984) showed that removal of 5 c111 of soil reduced ~nuize
yields by 95% on an Ultisol whereas on Alfisols mean yield decline was only
52%. Cowpea yields were only reduced by 63 and 22% on Ultisols and
Alfisols. respectively. In own measurements, maize yields on an Ultisol were
rero after four years of erosion had stripped off the surfice soil. On an
Alfisol, however. which had been exposed for 8 years, a poor yield was still
possible.
7.6 Soil lo\\ tolcrancr l i l ~ ~ i t \
Ideally, the soil loss rate should not exceed the soil formation rate of
the parent material. Most reported annual weathering rates for tropical
climates are below 500 kglha (Table 76-1) which is far below agricultural
soil loss rates. The lost productivity is irrecoverable by external inputs. This
is even rnore true ['or small scale farmers in developing countries which do
not have the necessary inputs to mitigate soil damage.
Tolerance value\ alco depend 011 the intended purpose of (oil
con\ervation. In general. the purpose for erosion control will be agricultural
production. However, in flood prone areas water retention can be the more
important goal whereas for the municipal authorities wdiment damage\ on
road ditche\. waterway5 or in the public sewerage 5ystem may be decicive.
In order to formulate tolerance limits, a decision about a seasonable
conservation time ~iiust be taken. This is rnore a political and social clucstion
than a scientific orit.. Can we tolel-ate a 50 C/c yiclcl decline in 50 years. 100
year-s, - or arc we still responsible h r the well-being of 0111- ancestors in 500
or I000 years? An answer to this clucstion must be founci in order to calcitlate
a mean. annual tolerable soil loss.
country
Cenlral
AI' r~ca
Puerto
KICO
Kcnq a
I
, Kenya
1 OSA
climate
L
sub-humicl
humid
humid
1 arid
parent annual literature
rate [kgka]
Toleraiice values. - first developed in the USA. were based on
e\timates ol' a sul-face +oil formation of 2.54 cm in 300 to 1000 year\
(Bennett, 1928). This e\timatc wa\ later changed to 2.54 cm in 30 year\
u hich i \ about 1 1 t/ha"'a (Pimcntel et al., 1976). Thi\ wa\ thc b:i\i{ fol- 5ettiny
150-300
< 150
300
Dunne et al. ( 1978
Dunnc el al. t 1078)
KII-hhy i 1980)
maximum annual soil loss rates in the USA to 1 1 tlha. Thus, setting of this
and all subsequent values was more based on expert judgemelit and PI-actical
considerations than on scientific data.
Tolerance values for tropical soils have not yet been formulated on
a n international level. However, some countries use tolerance values and
propositions were made by some ~luthors. A summary of existing values is
given in Table 76-2.
literature
Chin 61L Tan ( 1974)
Central African Federation
after Hudson ( 1986)
,
- -
Nyagumbo ( 1992)
Hudson ( 1986)
La1 ( 1980)
Lal ( 1983)
Hurni ( 1980)
applied for
tropical \oil\
clayey \oils
5a1idy \ ( MI \
undy \0115/Zimbabwe
\hallow. erodible \oil\
\hallow h~ghland mils
trop~cal soil\
Et h~op~a
tolerance limit
[t/ha*a]
15-25
I I
9
5
2-5
2.5
0.2-2
2
Annex 1 Rainfall and erosivity
Annex 1.1 Erosivity for single sites
Check if your site is included in Table I 1 - 1 Annex. Erosivity was
directly calculated for these sites. Also verify if erosivity data are available
fl-orn the meteon)lopical services or research stations-';. If your locatiotl is
near to one of the sites in Table 1 I-1Annex and has the same annual rain
volume. you may as well use the erosivity given in the table. If the rain
volurne of your site varies within 10% of a nearby station, you can
extrapolate the erosivity value linearly. The error for stations between 400
and 4000 mm in Table I I - 1 Annex is supposed to be less than 6 '/r (Figure 1 1 -
1 Annex).
E.t-c1111/710 ,fi)ta (~.~tl-eri~oll/tiorz: 7/10 ,site ill Tc//)/o 11-1 Ar7r1o.1- t - c ~~1i 1~) . <
1330 rllrll/cr (?/' r.clirl ,\,it11 trri c~t-o,si\~it>' of' 1249 N/lr. Yorrt- o ~ ~ i l .site rlcjtrt-1)). 1ltr.c
1 .(Of1 I I I I I I ~ I . E!.,,,,/~)I- ?,or rt- .site cwrl he1 c~trlc~~1krrpcl1~~ (1249/1440)'!'/ 300 = 1 I28 N/I1.
+
Bresch (1 993)
--..----.
error by 10% less
-
error by 10% more
annual rain volume [mm]
Erosivity for single site3
Tcrhle 1 I - 1 Atztzc1.r: Erosi\?itj?, ruirl ~lolunze, mea.c.urerner~t prriotl
Ji)r ~irzgle sites
country site erosivity measure- mean
literature
ment annual
period rain
[Nlh] la1 [mm 1
Algeria Go~~r ar i 139 3, 555 Marour ( 1997)
- - tleriz
53 2 338 - .. -
- - Madjoi14 5 0 1
330 - " -
- - Sidi Moharncrl Cherif 53 7
338 - " -
Burkina Faso Bobo-D~oi~litsso 09% 58 1 150 Galabert & Mlllogo
i 1973):"2
- -
Dori 468 4 7 530 - " -
- - Fudu- N'Gourma 772
48 X O O - " -
- - Faritho- Ba
84 1 6 10x3 - " -
- - Gaollit
1076 53 I240 - " -
- - Gunab 599
5 709 Roose ( 1975)
- - Mogteclo 656 6 754 Galabcrt & Millogo
( 1973 ):,:2
- - Niangoloko 1161 1 3 1330 - " -
- - Oi~agadi)i~gou 763 2 1 8x0 - " -
- - Ouahigouq a 607 49 700 - " -
s. ' . . ,rrl,c (Mctco) 72') 30 840 - " -
Rurundi Mushitsi (C;iheta) 499 2 1 157 Stocking & ~1wcI1
i 1976)
(:ameroon Rnf'ia 818 3 4 2 Bresch ( 1993)
- - Bamrnda 1395 6
3315 " -
Biuiganptc 509 I 3 Nill ( 1993)
- - Bat o~r i 750 I I 1472 Rrcsch ( l993)
- - Dihum hit
1627 I 2220 Nill ( 1993)
- - Dnuala 323 1 I I 3566 Rresch ( 1093)
- - Dschiuip
I OX4 4 1070 Seguj ( 197 1 ):'.2
- - Giu-oua 400 X 974 Rre5ch ( 1993
- - Marouu
540
12 751 - " -
Meiganga 858 10 1477 - ' I -
- - Nachtigal l Oh3 3 1330 Nil1 ( 1903
- - Ngaounci6rC 746 14 1485 Brcsch ( 1093)
- - Nlioi~nclja 1015
X 1901 - " -
Perlh~r Mlchcl iBanso~1) 777 4 1560 Nill i 1903)
- Pol i 1326 4 l3XX Bresch ( 1993)
- Y aoundP 933 13 1593 - " -
Cameroon Yoko 667 8 1543 - " -
Chad Dell 054 3, 1 1 100 Audr!, i 1973) 1
Annex 1 . I
countr> site erosivity nieasure- niearl literature
nient annual
period rain
I N h l [a] [nim 1
Ibor? Coast Ab1tll~111 3 1 80 37 3 100 Roo\c ( I075 )
Roosc cY: Rerr~-:uld
( I07 I I"?
Roo\e & .ladin
( lo(>o).,:2
Koosc ( 1975 I
Wenncl- ( 1977 1 '.3
- -
Ko r h o ~ o
Eldorrr
K~\ LI I I I LI
Ki1;lle
1 -0db iII-
Malindi
Momba\a
Ni~irobi (Kahct c)
Nakuru
Nanyuhi
Niu-oh
Voi
Bcfandriana
Allokoto l
Calab~u-
I - . ~ L I ~ L I
I hadan
I hor n
Y \ L I ~ ~ ~ I
On i t s h ~~
0uen. i
Port- Harco~lrt
Llnludi kc
Rurare
La1 ( l 076h)
/\I-mon ( I OX4 ' I
Salako ( I 0 8 8 )
Armon ( 108I). ' - 1
Salako ( 19x8
-
- -
- -. GLI ~ LI LI
- Giscnyi
- - tiamemhe
- - Kigali (airport)
lii~hcrigcri
Senegal B~u~l hcy
Erosivity Ihr 4inglc sitcx
countr~ site erosivity measure- mean literature
ment annual
period rain
INJhI la1 Imml
Zanihia Chlp'itu 6x5 10 1016 Pauuelyn et al ( 1988)
Zimbabwe
Kabompo
Kahn e
Kiil.11~1 Polder
Kawma
Muinilunga
Nclol:~
Seshche
Rci~bl-idgc
Cliil,inp;~
'hi\unjhanjc
I k t t
Enheldoorn
Fort Victoria
Ciohnc
1nya11g:i
Karol
I>upane
Siili\hury
T,jolo!jo
Tul 1
- ' I
In: S:iliiko ( 1988)
" 2 i n: Ruosc ( 1975)
3 in: Moore ( 1979)
Annex 1.2
Annex 1.2 Erosivity regressions
See if your own site is listed in Tlible 12-1 Anrzex or if it is close to a
site listed there. There is no evidence as to how far these regressions can be
used apart from the specific sites for which they were calculated. However,
the quality of rain will generally not change in the same geographic area
within some kilometers. A qualitatively similar rainfall is presumed which
can differ in volume.
If regressions in Tciblt. 12-1 Annex are given for single storms, the
mean annual is calculated by summation of the storm erosivity for
several years. It is not possible to use these regressions with annual rain
volumes.
Tclhle 12-1 Arzizex: Regression,y.fi)r the calculntio~z of erosiliihl
(P~,,l,l = nzeu17 aiznu~11 rain volut?ze [mm],
ElLjo = lnearz cirznu~il erosivih; El-,o, = erosillihx c?f'ci S ~ O I - I ? ~ i;
PI = ruin llolume ( f a .srorm i)
No. country1 area location regression remarks literature
El3,, [ N h l
I Hurkina Faso Bobo- El = 0.0 ISX'?Pl - I .24 for i ~ n e l r i t ormi :
Galahert Kc
D~oul ; ~\ \ o regrei \ ~ori \ 101 No . I and MI I I o~o
2 were \erq i ~ m ~ l a r to (19731 111.
eclllatlon No 3 Dclwaullr DcI u; I LI I I ~
thesefol-e proposed to u\e ( 1973)
No, 3 for the uhcl reglor]
hetueen 440 and 1 160
111111
2 " - Dori i cc No . I - -
3 - " - Ga~npel a and Ell,,, = 0.01 58 'PI ,,,I- 1.2 foi- \ ~ngl e \ t or~i l \ : hoth I ) el ~al ~l l c.
Gon\C neat I-? > O. UX, n = 7 year\ of \ i t e\ Ilad very \i11111~11 I J I I I ( 1077)
0 u g a d o ~ 1 ~ 0 ~ 1 \ ~ r ~ g l e itol-liii clistrihur~o~i and ucl-c
c\ ;~lu;rteci together \% 1111
Allohotol N~gcr
4 ('amcroon a11 LI,,, = ( I 1+0 012 P;lnn)'
I X \t;ltloni \ + ~ t h 3 to I4 131r\cIi
rz = 0 Ol year\ cC~ch I 10c)3 I
Erosivity regressions
inland > 1250 Elio = 0.269 'Pann+ 1 13
m altitude
Uganda pla- El,(, = 0 833.'.Pann-396
teau
No. country1 area location regression remarks literature
EI,,, [ Nh l
5 Kenya coa\tal 7one El = I I49"'Pann-840 calculated for Lamu. - " -
tn 50 hm Mal ~ndi . Mombnsa. Dar ec
inland Salaam and Z a n ~ i bar
6 - ' , - inland < 1250 El;,, = 0 57 1 'Pann-80 calculated for Lodwar. - " -
m altitude Makindi. Voi. I hdoma,
Ki go~na. Mwan/a and
Tabora I
7 - , ' - calculared for Eldoret.
Equator. Kitale. Nai rob~
Airport. Dagoretti. Kabete
and Wilson. Nakuru.
Lyamungu and Mbeya
" calculated li)r K~surnu.
Entehbe. Foprt Portal.
Ciulu. Ji~?ja. Kahalc.
Kampala. Ka\e\e.
Masi nd~, Mbarara and
Tororo
0 - , ' - Katumanit Elio, = 0.9;>P,-97.4 regres\ion for \ ~ngl e Ill\aker &
Mach;rho\ r2 = 0. 9; n = 35 \terms hased o n 35 event\ On \ ~a d
( 1984)
I 0 - " - Kat ~~rri ~i ri i / El,,,, = 0. 020hPI~o, -3. 9 - - - -
Machakos r2 = 0.99. n = 35
I I Niger Allohoto El7(,, = 0.01 58"PI ,,,,- I .:! \ee No. 3 Dclwaulle
( 1973)
I:! Nigeria Alore El,,,, = (0.12+U.I8P,)' ('or single \torn]\ Nill ( 1993)
r' = 0.87: n = 240
13 - " - Ibadan El io, = ( lU48i'P,- 1059)*0.017 fol- single s t o r m: La1 ( 1976b)
r? = 0.67 regression for 5torm5 of 3
years
14 Rwanda a11 El3,, ca. (0.433':'Pann) biisrd on 6 stations
Kyumugahc
and I 0 years & Berding
(1992)
15 Sahel Ello = 0. 87 * Pann Roose ( 1977)
I 6 Tanzania \cc No. 5 to 8 u5e regre.;sion\ No. 5 to 8 \ee No. 5 t o 8 Moore ( 1979)
Uganda pla- u\ e r egr e~si on No. 8 see No. 8 17 Uganda
- -
Annex 1.1
No. country1 area location regrcssio~i remarks literature
El.,,, [N/h1
I8 Zambia ;dl t l , , , , = 0. 0230 :PI " 1 tol- \ ~ngl e \torlns Paun el! n ct
r2 = 0. 7 1 . n = 2348 al t IWX)
Kahwr tl ,I,, = 0. 0235 P, ' x7 -
r.2 = 0 (17. n = 2h 1
27 Zimbabwe all El ?, , = 0.2 l 5 Panrl-5 1 . 1
r 2 = 0 h l . n = O h
78 - " -
Ea\rern cl~\trlct\ I I ,,, = 0 I83 ' Parlrl-54.2
r? = iI,X7. 'I = I0
I Ke ~r c \ \ l o ~l \ 111 hlool-e (1070: T;lhlc 3 ) 101 KI I ~, 1 25 u c r r r ~l t c ~- c d Irlto rryl-c\\iorl. El;,, [ t t Lon\ 1111~1crc a1 =
0.02Y KE>?_5-26. 1 2 = 0.C) which W;I\ cal c~~l ; ~t ccl 11-o111 1 1 s t ; ~ t ~ o ~ i \ 111 Kcn) : ~ (Wrnnc.1-. 10771 ,11111r nul r ~pl ~r t l I>!
1 735 In ol-cicr- to tran\l;i~-ln to SI unlrs
Annex 1.3 National iso-erodent maps
Burundi (Simonart ct al., 1993)
Annex 1.3
Cameroon (Bresch, 1993)
GABON
CONGO
National i\o-el-odent maps
Marocco (Arnoldus, 1977)
ALGERIA
SPANISH SAHARA
Annex 1.3
South Africa (Smithen 8L Schulze, 1982)
ZIMBAWE
Zambia (Lenvain et al., 1988)
Annex 1.3
Zimbabwe (Stocking & Elwell, 1976)
Annex 1.4 Regional iso-erodent maps
West Africa (Rouse, 1977)
GABON
Annex 1.4
Africa north of the Sahara (Arnoldus. 1978)
National ~rainl'all ( l i \ r 1 . i h u t i ~) 1 1 111; qx
Annex 1.5 National rainfall distribution maps
Angola (Grif'f'i th. 1972)
1250
Atlantic Ocean
No information
NAMIBIA
Equatorial Guinea (Griffiths, 1972)
I
GABON
Atlantic Ocean
Ethiopia (Griffiths, 1972)
SOMALIA
KENYA
Ghana ( von Gnielinski, 1986)
BURKINA FASO
IVORY
National I-ainfall distribution map4
Ivory Coast (Wiece, 1988)
BURKINA FASO
LIBERIA 2000
Atlantic Ocean
Annex 1.5
Liberia (Griffiths, 1972)
National rainl'all tiislr-ihution maps
Madagascar (Sick, 1979)
Annex 1.5
National I-ainfall cii\tl-ihution map\
Mozambique (Nelson, 1984)
TANZANIA
0
b
ZIMBABWE
Indian Ocean
SOUTH AFRICA
Nigeria (British West African Meteorological Service, 1954)
CAMEROON
Atlantic Oman
Annex 1.5
Sao Tom6 and Principe (Servico Meteorol6gico)
Atlantic Ocean
Atlantic Ocean
V
National rainfiill di x~ri but i on map\
Sierra Leone (Griffiths, 1972)
UGANDA
National rainfall di\tribution map\
Tunisia (Schlicphake. 1984)
Remada
. -
Rail1 volume iund distribution for single sites
P I P - - a r t - r r r - - - a - t r - r t r , - r , z l r , c r ~ ~
X CI - r- Cr -
Cl r , - - P I -
- = - C C , ~ ~ ~ J C I - S X X -
PI - PI CI CJ r1 r. -t r.
Anncx 1.6
*
2
5
2
~ , , , 1 , , , , , 1 , , , , , 1 , ~ ~ , 1 ~ , ~ ~
c : : : : : : : : : : : : : : : : : : : : : : : : :
1 1 1 / , 1 1 / 1 , 1 , 1 , I / / I I I I I I I I
Rain volume and distribution for single sitea
Ra i n volurnc and distribution lot- sitlglc hi l t . \
r
a
i
n

v
o
l
u
m
e

a
n
d

d
i
s
t
r
i
b
u
t
i
o
n

[
m
m
]

c
o
u
n
t
r
y

s
i
t
e

a
n
n
u
a
l

j
a
n

f
e
b

m
a
r

a
p
r

m
a
y

j
u
n

j
u
l

a
u
g

s
e
p

o
c
t

n
o
v

d
e
c

r
e
c
o
r
d

l
i
t
e
r
a
t
u
r
e

i
n

y
e
a
r
s

G
a
b
o
n

B
i
t
a
r
n

1
7
4
6

4
9

7
2

1
8
2

1
9
1

2
3
1

1
0
1

2
6

3
7

2
7
7

2
9
7

2
0
8

7
5

1
2

S
u
c
h
e
l

(
1
9
7
2
)

-

-

L
i
b
r
e
v
i
l
l
e

2
5
9
2

2
0
6

2
9
1

2
6
4

3
9
5

2
4
4

4
0

0

1
1

1
0
6

3
5
9

4
1
6

2
6
0

1
8

G
r
i
f
f
i
t
h
s

(
1
9
7
2
)

-

-

M
a
k
o
k
o
u

1
7
5
4

8
3

1
2
4

2
5
5

2
4
5

1
8
2

5
1

5

1
8

1
3
2

3
3
1

2
1
0

1
1
8

?

-
"
-

-

-

M
e
d
o
u
n
e
u

1
9
6
7

1
5
.
5

1
6
3

2
3
5

1
9
9

1
6
0

5
9

4

1
4

1
6
8

3
3
7

3
0
1

1
7
2

?

-
"
-

-

-

M
e
k
a
m
b
o

1
6
6
1

7
7

1
3
0

1
8
2

1
7
4

1
7
8

8
8

2
8

7
0

1
5
3

2
6
8

2
0
9

1
0
4

?

-
"
-

-

-

M
i
n
v
o
u
l

1
5
2
9

4
2

7
9

1
3
9

1
5
5

1
9
4

1
2
2

4
7

5
5

2
0
0

2
6
3

1
7
5

5
8

1
3

S
u
c
h
e
l

(
1
9
7
2
)

-

-

M
i
t
z
i
c

1
8
4
2

1
1
8

1
1
0

2
2
6

2
0
7

2
2
2

4
6

1
0

1
4

1
5
0

3
4
6

2
4
7

1
4
6

3
0

G
r
i
f
f
i
t
h
s

(
1
9
7
2
)

G
h
a
n
a

A
c
c
r
a

7
8
7

1
6

3
7

7
3

8
2

1
4
5

1
9
3

4
9

1
6

4
0

8
0

3
8

1
8

?

"
4

-

-

A
d
a

8
6
4

7

1
8

7
2

9
4

1
7
4

2
0
8

5
7

1
2

6
1

9
9

4
7

1
5

?

v
o
n

G
n
i
e
l
i
n
s
k
i

(

1
9
8
6
)

-

-

A
k
u
s
e

1
1
1
8

2
3

4
6

1
0
5

1
2
8

1
6
3

1
8
0

6
5

3
9

9
6

1
3
1

1
0
2

4
0

?

-
"
-

-

-

A
x
i
r
n

2
1
2
9

5
1

6
2

1
2
9

1
4
2

4
2
0

5
3
5

1
5
6

5
4

8
7

2
0
5

1
9
2

9
6

?

-
"
-

-

-

E
n
c
h
i

1
6
5
1

4
1

6
5

1
3
8

1
6
0

2
1
0

2
7
0

1
4
3

7
5

1
6
1

2
1
0

1
2
5

5
3

?

-
"

-

-

-

H
o

1
4
2
1

3
6

7
8

1
3
9

1
4
5

1
7
7

1
8
3

1
1
1

8
3

1
4
9

1
9
1

7
9

5
0

?

-
"

-

-

-

K
e
t
e

K
r
a
c
h
i

1
4
0
4

1
9

3
7

8
2

1
2
7

1
6
7

1
9
2

1
5
8

1
2
4

2
2
3

1
8
5

6
8

2
2

?

-

"

-

-

-

K
i
n
t
a
m
p
o

1
5
6
7

8

4
0

1
0
2

1
6
0

1
8
6

2
3
9

1
4
4

1
1
7

2
7
7

2
1
3

6
8

1
3

3
5

G
r
i
f
f
i
t
h
s
(
1
9
7
2
)

-

-

K
u
m
a
s
i

1
4
8
1

1
7

5
9

1
3
7

1
4
5

1
8
2

2
3
4

1
2
6

7
4

1
7
6

2
0
2

9
8

3
1

?

v
o
n

G
n
i
e
l
i
n
s
k
i

(

1
9
8
6
)

-

-

N
a
v
r
o
n
g
o

1
0
7
3

2

4

1
9

5
0

1
1
6

1
2
6

1
9
0

2
6
3

2
3
1

6
3

7

2

?

-
"
-

-

-

T
a
k
o
r
a
d
i

1
1
8
6

3
3

3
7

8
0

9
5

2
4
5

2
8
0

8
9

3
6

5
0

1
2
0

8
3

3
8

?

-
"
-

-

-

T
a
m
a
l
e

1
0
5
3

2

9

5
1

8
8

1
2
1

1
3
2

1
2
8

1
8
9

2
1
7

9
8

1
3

5

?

-
"
-

-

-

W
a

1
1
1
1

5

1
0

4
2

7
5

1
3
2

1
4
5

1
4
4

2
1
5

2
3
4

8
5

1
9

5

?

-

"
-

-

-

W
e
n
c
h
i

1
3
5
4

8

4
2

9
7

1
5
0

1
7
8

2
0
4

9
3

6
9

2
0
0

2
1
9

7
5

1
9

1

-
"
-

-

-

Y
e
n
d
i

1
1
8
6

2

7

3
2

9
5

1
3
3

1
4
6

1
5
3

1
8
8

2
5
9

1
3
2

2
1

8

?

-
"
-

r
a
i
n

v
o
l
u
m
e

a
n
d

d
i
s
t
r
i
b
u
t
i
o
n

[
m
m
]

c
o
u
n
t
r
y

s
i
t
e

a
n
n
u
a
l

j
a
n

f
e
b

m
a
r

a
p
r

m
a
y

j
u
n

j
u
l

a
u
g

s
e
p

o
c
t

n
o
v

d
e
c

r
e
c
o
r
d

l
i
t
e
r
a
t
u
r
e

i
n

y
e
a
r
s

G
u
i
n
e
a

B
e
y
l
a

4
0
9
9

5
0

1
4
7

2
9
3

3
6
0

6
3
1

5
5
5

3
9
2

4
9
0

4
9
5

2
9
0

2
7
4

1
2
2

2
9

G
r
i
f
f
i
t
h
s

(

1
9
7
2
)

C
o
n
a
k
r
y

K
o
u
r
o
u
s
s
a

M
a
m
o
u

A
b
i
d
j
a
n

B
o
u
a
k
e

F
e
r
k
e
s
s
e
d
o
u
g
o
u

S
a
s
s
a
n
d
r
a

T
a
b
o
u

E
q
u
a
t
o
r

G
a
r
i
s
s
a

K
i
s
u
m
u

M
a
g
a
d
i

M
o
m
b
a
s
a

M
o
y
a
l
e

N
a
i
r
o
b
i

P
e
r
k
e
r
r
a

-

-

P
o
r
t

V
i
c
t
o
r
i
a

-

-

V
o
i

L
e
s
o
t
h
o

M
o
k
h
o
t
l
o
n
g

L
i
b
e
r
i
a

M
o
n
r
o
v
i
a

L
y
b
i
a

B
e
n
g
a
s
i

-

-

D
e
r
n
a

-

-

T
r
i
p
o
l
i

-

-

-

-

-

-

*
4

G
r
i
f
f
i
t
h
s

(

1
9
7
2
)

-

-

-

-

*
4

G
r
i
f
f
i
t
h
s

(
1
9
7
2
)

-

-

S
u
t
h
e
r
l
a
n
d

&

B
r
y
a
n

-

-

G
r
i
f
t
i
t
h
s

(

1
9
7
2
)

-

-

"
4

-

-

-

-

-

-

Annex 1.6
Annex 1.6
Rain volume and distribution I'or single sites
= ~ c c a 3 m w , m n = t - - m n r 1 ~ 1 1 ~ , - m ~ , c c
a * F i r , - m at- m
T, X 2
Annex 1.6
Rain volume and distribution for 5ingle sites
Annex 1.7 Estimation of the erosivity of the 10 year
storm (EI,,/lO)
Data for the EI,,,/IO will be deficient in many cases. For sorne areas
close correlations exist between single storm volume (Pi [mrn]) and single
storm el-osivity (EI,,,, [Nlh]). Frorn these correlations an estimate of the
El,,,/ I O can be made if the volume of the I O year storm is known (c.f. Figure
75-2). However, the regressions in Figure 17-1 Annex show that these are
rather large differences between climatic zones which recommend a cautious
use of such regressions. The regressions in Figure 17-IAnnex are mean
relationships of several regressions from Cameroon (Bresch, 1993) and
Zambia (Pauwclyn el al., 1988):
coast (Cameroon)
Douala: EI,oi = ( 1.45 + 0.095'I'P, )2, r2 = 0.75, n = 830
inland (Cameroon)
Y oko: El ,,,, = (0.14 + 0.1 39'!'P, )', r' = 0.80. n = 352
Batouri: EI ,,,, = (0.37 + 0. 133'I'P,)', 1-2 = 0.74, n = 324
Yaoundk: El3(), = (0.07 + 0.1 53'!'P, )'. r' = 0.8 1 . n = 553
highland (Cameroon)
Ramenda: EI,O, = (-0.08 + 0. I 52'::P, )2, 1-2 = 0.82. n = 423
Nkoundja: El,(), = (0.02 + 0.148'!'PI)', r' = 0.73, n = 469
north (Cameroon)
Maroua: EI ,,,, = (0.08 + 0. I56"'PI)2. r' = 0.82, n = 252
Garoua: El ,,,, = (0.13 + 0.1 5Uq'P, )', r' = 0.84, 11 = 132
Poli: EI ,,,, = (0.26 + 0. I 49'I'P, ) 2, r' = 0.83, n = 175
Ngaoundh-6: Elqo, = (0.20 + 0.15 1 "'P, )'. r2 = 0.78, n = 573
I ~ I - Zambia a regr-ession was given by Pauwelyn et al. ( 1988):
The curve f.ot- the coa\l i r ~ Cameroon \hould give I-eawnable
e\t~mate\ for \ite\ along the We\t African Coast wrth pronounced influence
of the monwon. The inland curve should be applicable for tho Central
E \ t l ~ ~ l a t ~ o n 01 thc I0 ycnl \ t o m cl o\ l Ll t >
Al'rican Lone between I 000 and 1 500 1nm of rainfall. The regre5sion f'oi-
Northern Cameroon can probably extcndcd to I'urther areas of West Africa in
the 7une between 600 to I000 mm.
single rain volume [mm]
-t- coast + inland --m- highland
-t- north ++ Zambia
Annex 2 Slope length and gradient
Lkvice for rnea\uring >lope-len~tli and ~riiclic'nt
Annex 2.1 Device for measuring slope length and gradient
10 m transparent hose
with - 10 mm internal diametre
Fig1rt.c 2 1- 1 At l l rr. ~: Wcrfut. I o~~c~l , / i ) t . tllccr.sr~t-c~t~rc~tlt ( ? f ' ~ I o ~ ~ c ~ - l ~ ~ t l g t / ~ ~ I I I ~ ~I . CI C/ I CI I I
-
-
-
-
7
-
The two scaled bars arc placed o n level ground and the hosc is filled
with hatel. up to the zero riiark on the bars. The stoppers need to be taken ol'f'
tlie hose ends bc1'ol.e starting the rne;tsut-cment. Make sure thai no air bubbles
are i n the hose! If the hose c1iamctc1- becotncs too small. i t is dif'f'icult to
eLracuate air bubbles from thc hose. A slilall cluai~tity of hoi~sehold detergent
may help in this case. For the tneasuremcnt. one person keeps one o f the bars
upright while a second person moves down-slope until tlie string bctween the
bars is completely streched out. The distance between the two bars should
now be 5 111. The vertical distance between the two bars can be read 011 the
scale. If. for cxatnplc. the vertical c-list;uice is 40crn. the water in the liose of
the lower bar SI ~OLI I CI be beside the 20cm 11iai.k w1ic1-eas o n the higher bar i t is
2Ocnn below the zern mark. The gradient ( s ) can be calc~rlated by:
For practical purpose\ i t is easier to double the scale on the bar\ (e.g. 0. 2
In = 0.4 m) in order to receive the vertical distance right away.
scaled bars
2 m high
5 m string with knots every metre
- -- --
-
-
-
-
-
-
-
stopper to
close hose
for transport
Annex 2.2 Conversion of slope gradient in degrees to
percent
gradient
degrees [%I
1 2
2 3
3 5
4 7
5 9
6 1 1
7 12
8 14
9 16
10 18
1 1 19
12 2 1
13 2 3
14 2 5
15 2 7
16 2 9
17 3 1
18 32
19 3 4
20 36
2 1 38
22 40
23 42
24 45
2 5 47
26 49
2 7 5 1
28 53
29 55
30 58
3 1 60
32 62
33 65
34 67
35 70
degrees I% 1
36 73
37 7 5
38 78
39 8 1
40 84
41 87
42 90
4 3 93
44 97
45 100
46 1 04
47 107
48 1 1 1
49 115
50 119
5 1 123
52 128
53 133
54 138
55 143
56 148
57 154
58 1 60
59 166
60 173
6 1 180
62 188
6 3 196
64 205
65 214
66 225
67 236
6 8 248
69 26 1
70 275
Nurnber of day in the yeas and con-espot1d1n.g date
Annex 3 Cover and management factor
Atlrle.k 3. 1 N~~r?zhc)t- t !/ ' t l t l ~, irz r l l ~ orit it. trrztl c.or.ro.\porztlirzg tklto.
0 I 1
O J 2
0 3 1 1 3
0 4 - 1 1 4
05-Jan 5
0 0 1 1 h
07-Jim 7
O X - I X
O O J I 0
0 - I I 0
I I I I I I
[?-Jan I I
13. l 11 13
14-Jan I 4
15-Jan 15
10-Jan 0
7 - I 7
IX-Jan IX
I I I 9
70-Ian 20
11-Jan 21
22-.I;ln 22
23-Jan 23
24-.la11 24
75-Jan 25
26-Jan 2(,
27-J:ln 27
28-.1;111 28
- 1 29
30-.la11 30
S I 31
01-Feh 32
02-I:eh 33
()3-Fc13 4
04-Feh 35
05-Fcb 36
Oh-Fch 37
07-I'ch 18
()X-r:yt~ 7')
09-l:ch 40
10-Feh 4 I
I 1 -FL>h 42
12-I.ch 4 1
I 7-kt,h 44
t ~ h 45
15-Fch 46
I 6- F~t h 47
17-Fch 48
I X-l,el> Jc)
I 9 ~~h 50
?()-I . ch 5 l
2 I.Ft.ll 57
72.i.+h 53
23-1.~13 54
24-l.et, 55
75-keh ,jh
70-f . eh 57
27.reh 58
2X-Fch 59
01-Mas (10
O M h l
03-Mi l r 02
04-Mas 03
05-Mas A4
Oh-M; I ~ 6.5
07-Mar 1 0
OX-Mat 67
O M 6X
10-Mar 60
I I - Mi l s 70
1 ?-Mar 71
13-Mar 72
14-Mal- 7.1
15-Mar 74
- M I 75
l 7 1 ; 1 1 7h
IS-Mar 77
IC)-Mar 78
O M 70
21-Mas SO
22-Mar XI
23-Mar. 82
24-Mar X3
15- Mar 84
26-Mal- X5
27-Mar Xh
28-Milt- 87
20-Mill- 88
0 - M a r SO
31-M;IS 0
Ol - Apr 1 I
01- i Zp~ 92
3 3
04-Apr 4
05-Apl - 05
00- Apr 0
07- A~I - 97
()X-Apt- 08
0 qC)
10-Apl I 00
I I - Aps 101
12-Apr 102
I I I 0
14-Apr 0
15-Apr. 105
I (I-Apr 106
17-Apr 107
IX-Apr. 108
LC) ,jpl- 0 )
20-Apr I I 0
2 I - Apl - 1 1 1
22-,A,pl 1 12
73-Apr 113
24- Ap1 114
25-Apl- 1 1.5
A I I
2 7 . ~ ~ ~ - 1 17
2X-i\pr 1 1s
o . ~ ~ - 1 1 9
i ( ) . ~ p ~ . 2 )
day in the year
Ol -Jul 1 X2
02-Jul I X3
03-Jul I 84
O L I 185
05-Jul 1 Xh
O J I 1x7
07-Jul I 88
OX-Jul ISC)
09-Jul I 90
10-Jul 101
I I -Jul I 92
12-Jtrl I 93
13-Jul I 04
4 J 1 l I 95
15-Jul I 00
I h-.l~rl I 97
17-Jul IOX
I S-Jul I 09
J I 200
20-Jul 201
I - I 202
22-Jul 203
2 3 - I 204
24-Jul 205
25-Jul 206
I - I 207
27-Jul 208
28-JuI 20C)
20-5111 210
30-Jul 211
31-Jul 712
Ol - Aug 2 13
0 2 - i Zu ~ 2 14
0 - 1 2 5
04-Aug 2 10
05-Auf 217
00-Auf 2 1 X
07- Aug 2 19
OX-Aug 120
09- Aug 221
O A I 222
I I - Aug 27.1
12-Aug 224
13-Auf 225
I 4- Aug 226
A 227
It,-Aug 228
17- ALI ~ 220
18-Aug 2.30
10-Auc 23 1
20-ALI~ 232
21 - Aug 233
22-Aug 234
23-Aug 235
2-1-Aug 236
25-Al l y 737
26-.4ug 238
27- ALI ~ 230
28-Aug 240
20- ALI ~ 241
A I 242
3 1- ALI ~ 243
date/ no. of
Ol -Ma! 12 1
02-M;I) 122
03-Mi11 123
0 - 1 ) I
05-Ma) 125
Oh-Ma) I
O M 127
OX-Ma) 128
09-M;i) 129
10-Ma) I 30
1 I - Ma) I
1 2-Ma! I 32
13-Ma) 133
14-M;I) I
15-Ma) 135
I h-M;l) 136
17-Ma) 137
I X-Ma) 1.38
I Y-Ma) I
20-M<I) 1-10
21-Ma) I
??-Ma) 142
23-Ma) 4 3
?&Ma) 144
25-Ma) 1-15
26-May 4
27-Ma) 137
28-Mil) I4X
71)-M;1) 140
I ) I 50
M I I 51
OJ L I I 152
02- J~I I 153
- I 5 4
0 4 - I 155
05-Jun 156
O J 157
07-Iu11 15s
OX-Jull 150
OC)-.lull I 60
10-Jun 101
I I - J ~r n 162
12-JLIII 103
13-Sun It14
I - L I I 165
15-Jun 0
I h-Jurl 1117
17-Jun IOX
18-JUII I 00
- J l 1 170
20-.lulr 17 1
7 1 -Sun 172
22-Jun 17'1
23-Jun 174
24-Jun 175
25-Jun 176
J I I 177
27-Jurl 178
2X-Jull 170
- I 180
( I 1 I
Ol -Sep 244
02-Sep 2-15
03-Sep 2 4
04-Scp 2-17
05-Sep 248
Oh-Scp 219
07-Sep 250
OX-Scp 25 1
OC)-Scp 252
10-Sop 5
I I -Scp 5
17-Scp 255
13-Sep 250
14-Scp 257
15-Sep 258
16-Sc11 259
17-Sep 260
IS-Sep 201
1')-Scp 262
20-Sep 0 3
21-Scp I
22-Sep 205
23-Sep 260
24-Scp 267
25-Sep 268
20-Sep 26')
27-Sep 270
28-Scp 271
2C)-Scp 7
30-Scp 273
01- 0ct 274
02-0c.t 275
03-Oct 276
0 O C t 277
05- 0ct 278
06-Ocr 279
07- 0c1 280
OX-Oct 2X I
09-Ocl 7x7
10-0c.t 283
I I - 78-1
12-0ct 2115
13-0c.1 280
1-1-0c1 287
15-Oct 2x8
Ih-Oc( 789
17-(Jet l C)0
IX-(-)cl 2'11
1')-Oct 202
20-Ocl 2<)3
2 1 -(-)el 2 4
22-OCt 295
23-Oct ?')(I
24-Oct 207
75-(Jet 298
20-0ct 2')q
27-Ocl 300
28-Ocl .{()I
2 9 - 0 ~ 1 302
70-Ocr 303
3 I-0c.t 304
OILNO\ 305
0 2 0 306
3 - 3o7
O N 308
05-No\ 300
Oh-NO\ 3 10
07-No\ 3 l I
OX-NO\ 3 17
00- KO\ 3 1 3
0 - 3 I 4
1 I 3 I 5
12-Yo\ 3 I t ]
13-No\ 3 17
14-No\ 318
I ~ - N O \ 3 I 0
I t ) - No\ 320
17-No\ 32 1
I X-No\ 322
10-No\ 373
N O 374
N O 325
22-NO\ 726
23-No\ 327
24-No\ 328
25-N0\ 324
? ~ - K o \ 1 0
27-K0\ 331
28-KO\ 332
N O 333
30- No\ 334
OlLDec 335
07-1)t.i. i i ( 7
O3-l)ei. .i37
04-LILY 338
05-1)c.c 1 lC)
Oh-l)cc 340
07-1)ec 34 1
OX-llci. 342
0') I)cL. 34.3
1 0 - 1 1 ~ ~ 444
I lLl)c.c 3-15
I ~-L)cL' 340
13-l)cc 447
14-l)cc 348
15-1)cc 340
10-DCC, 350
17-1)cc 35 1
18-l)cc. i S 1
IO-DCL' 351
70-l)ec 75-1
2 1 - 355
27-I)CC .;SO
~ ~ - D c c 357
24-l)cc 35X
75Dec 35')
~( ~- I ) L . c 300
27-L)t.i 7f1 l
?X-L)ec. 302
?' )-[ kc 363
30-Drc 1(34
'1 I - I ) CC 305
Ficlds rnethocls for- the ~ncnsur-rmt . n~\ ...
Annex 3.2 Field methods for the measurement of mulch
cover and canopy cover
I . Mulch cover measurement by the meterstick method
Put a meter\tick on thc ground and count on one side all the cln-
marks which are in contact with mulch tilaterial (Figul-e 32-]Annex). The
mulo11 cover ( MC) i \ given by:
number of cm mark\ in contact
MC ( %) =- letiglI3 of nletcl-\tic]\ ( cm)
1 00
Ex:irnple: 38 cm-rnarks are i n contact with one side of a 2 111 long rneterstich
MC is 38/200 = 19 (2.
This method 1s well sulted for \~iiall plots. Thc nur-nbcr ot
measul-ernents depend\ on the unifonnnc\s of the cover. In own
mea\urenicnt\ 12 rcpllcation\ with a 2 m long metel-\tick were taken o n 500
m' plot\ equivalent to a random 23 rn transsect. It 15 iniportant that the {tick
i \ randomly placed i n the plot. Random placement can be a\\ured by
throwlng the meterstick into the plot.
Annex 3.2
2. Mulch cover measurement by the cord and knot method
This method is similar to the meterstick method. Marks or knot5 are
attached to a 10 to 20n1 long cord which is streched out on a field (Figure 32-
2Annex). The number of knots in contact with mulch material ('I: 100) divided
by the total number of knots on the cord gives the percent nlulch cover.
Field method$ for [he me;ljLIrcmcnt
3. Mulch and canopy cover measurement by visual estimation
Form a quadrat of i x I m using 4 wooden sticks of I m length or two
f'oldable 2 rn-sticks to mark out a 1 m' area in the field (Figure 32-3Annex).
If the observed area is well delimitated, cover is easier to estimate than on an
undefined area. The size of the area can be srnaller than 1 m' but should not
be larger because visual estimation becomes more difficult with increasing
size of the area. Cover is estimated visually i n the delimitated area.
Calibration of the eye can be facilitated by the examples ill Figure 32-
4Annex. Estimations become also easier if wires are streched at regular
distances on the wooden frame. Estimations are reasonably precise after some
routine.
Fig~lsc] 32-3 AIIII( ~. Y: Mtrrkirlg o ~ l t r l r l trr-err wyitl~ cliflorclzt cle\~ic.e.c ( t r. cc'ooclcil
fj-il111e w9it11 c~,it-o-ilet, I?. t ~~>i g . $ , c. tn70 2 r~~-. \ t i (, h, $)
4. Measurement of canopy and mulch cover by a sighting frame
This method was proposed by Elwell 81. Wendelaar (1977). Ten hollow
pipe\ ale attached to a frame (Figure 32-SAnnex). By peering through onc of
the pipes a ~ ~ n a l l area can be observed. Mulch or ciinopy cover in this area i \
either rated as 'yes' or 'no' ol- rated on a scale bctwccn 1 and 10. In thc first
case. the number of points observed with cover (= yes) divided by the
n~lmbel- of all points ob\crvcd gives the coverage.
Annex 3.2
Figure 32-4 Annex: Selected coverages for the calibration of the eye
(the stick in the pictures is 1 m long)
Field rllelhod\ 1.01- rhe n~ea~ur - c r nc . ~~~
1- 1 OOcm -I - 3.92cm
F i ~ l / s ( ~ -32-6 A I I I ~ O ~ : 0/1\01-\,(itio11 O~ I . OI - ( / \ ~ I ~ ~ / L / ( ~ I I ( ~ o ~ / 1719 /)/(111t /loigll/
60-
50-
-
8
U
C
L
canopy height [cm]
FieIci met hods f or (he nl ca\ u~-zmenr ...
In the second ca\e the \urn of rattngs for all ten pipes gives the coverage
for the 1 rn transsect. 100 obcervations in a regular cropstand were pt-ecle
within 2 C/ r coverage (Cackett. 1964) whereas 300 observations here
necescary for a 5 % accuracy in an irregular croprand (Elwell & Gat-dt~es.
1975). Quansah et al. (1990) used an average of 5.4 observationslm'. In own
nleasurcment\ 180 polnts on a 500m2 plot (0.3 observat1on\/m2) proced
adequate (Nill. 1993). This version of sighting frame can be u\ed for mulch
and durlng the f'irst 2-4 weeks of plant growth while the plants are st111 \mall
becau\e the observation errol- increases rapidly with increa\ing plant he~ght
(Figure 32-6Atitiex).
A modified version of the sighting frame avoids this obsel-vation
error- by sliding a graduated stick through the pipes (Figure 32-7Annex). In
this case the nurnber of contacts of the stick with leaves or mulch at-c
counted. This version allows at thc same time to measure thc mean canopy
height above gt-oi~r~cl.
An alternative i n tall crops is the use of a min-or on tlie \~ghtitls
frame which allows an observation of' thc canopy cover outlined again\t the
\ hy (P~gi l re 32-8Annex). The sighting I1-:~tiie\ are ea\y and cheap t o construct.
Cursor
gunsight
End Elevation on Cursor
Showing the Gurml@t
CONSTRYCTION DETAILS FOR SIGHTING FRAME , , , _. , ,
-- -
Fielcl merhods I'or the mea~ur' t ' ment ...
Stop for Cursor
bar Support arms
Isometric View on Mirror
'C bracket and
horizontal
Sectional Elevation A - A
CONSTRUCTION DETAILS FOR MIRROR
Annex 3.3 Growth curves for mono- and mixed crops
Growth curves for the following mono- and mix-crops are given for:
Balmbara nut
canavalia
cassava
cotton
cowpea
groundnut
ma z e
maizelcassava mixcrop
pigeon pea
rice
sorghum
soya
suntlower
tea
tobacco
Figure 33- 1 An1ze.r
Figure 33- 1 Aiznex
Figure 33-2 Annex
Figure 33-3 Annex
Figure 33- 1 A1111e.r
Figure 33-4 An1ze.u
Figure 33-5 A I I I I ~ . ~
Figure 33-2 Anne. .u
Figure 33-2 An1ze.r
Figure 33-5 Anr~ex
Figure 33-5 A1ule.r
Figure 33-4 Ar1ne.x
Figure 33-3 Arlrle-x
Figure 33-6 Arzrze-x
Figure 33-3 A1zize.r
Growth curies for mono- ancl r ni xed cl-op4
1 00
90
80
- 70
ae
Y
,- 60
>
00 so
> a
g 40
m
30
20
10
0 1
0 20 40 60 80 100 120 140 160
days after emergence
L
100
90
80
~
& 60-
>
a
o 40--
- --
'i
- -.
1 l l 1 l l l l l 1 l l r l l l l 1 l 1 1 1 1 >
0 20 40 60 80 100 120 140 160
days from emergence
100
go---
-- - -- - -- - -------- -
- - - - -- - - - - ---- - -- -
--- - --
0 l l l l l l l l l l r l l l l l l l l l l l l l i l l l l l l l
0 20 40 60 80 100 120 140 160
days from emergence
Growth curve$ l or mono- ancl rn~xed crop\
100 -
_ __L_
o l l l - l l l l r l l l l l l l l l l l l l l l l l l l l l l l l '
0 20 40 60 80 100 120 140 160
days from emergence
8 0 ~
70---
60
-
8
SO-.
-. - -----..-"- -.
-. - - ,
-
1 r 8 1 1 1 1 1 1
800 1000 1200 1400 1600
days after emergence
Annex 3.4
Annex 3.4 Detailed C factors
T~thlt. 34-1 Anrze.~: Detciilc-'d C ji~c.torLs,for,fc)rrrst, hush and gr0.s.s vegetcrtiotl
n
o
.

6

7

8

9

1
0

1

1

1
2

1
3

1
3

l
a
n
d
-
u
s
e

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

c
r
o
p

f
a
l
l
o
w

w
i
t
h

m
a
i
z
e

r
e
s
i
d
u
e
s

0
.
0
9

-

M
a
l
a
y
s
i
a

S
u
l
a
i
m
a
n

e
t

a
l
.

(

1
9
8
3
)

r
e
s
i
d
u
e
s
"
3

-
"
-

f
a
l
l
o
w

w
i
t
h

m
a
i
z
e

0
.
0

1

-

-

-

-

r
e
s
i
d
u
e
s

a
n
d

m
u
l
c
h

-
"
-

f
a
l
l
o
w

w
i
t
h

m
u
n
g

b
e
a
n

0
.
2
5

-

-

-

-

r
e
s
i
d
u
e
s

-
"
-

f
a
l
l
o
w

w
i
t
h

g
r
o
u
n
d
n
u
t

0
.
2
8

-

-

-

-

r
e
s
i
d
u
e
s

-
'
*
-

f
a
l
l
o
w

w
i
t
h

c
o
w
p
e
a

0
.
0
3

-

-

-

-

r
e
s
i
d
u
e
s

r
e
s
i
d
u
a
l

f
a
l
l
o
w

e
f
f
e
c
t
s
*
'

b
u
s
h

f
a
l
l
o
w

w
e
l
l

d
e
v
e
l
o
p
e
d
:

1
s
t

y
e
a
r

0
.
8

0
.
7
8

t
o

0
.
8
1

1

C
a
m
e
r
o
o
n

Y
a
o
u
n
d

N
i
l
l

(
1
9
9
3
.

p
.

1
6
3
)

a
f
t
e
r

c
l
e
a
r
i
n
g

-
-
-

2
n
d

y
e
a
r
s

a
f
t
e
r

c
l
e
a
r
i
n
g
"
2

0
.
9
"
2

-

-

-

-

-

-

g
r
a
s
s

f
a
l
l
o
w

I
m
p
e
r
a
t
a

g
r
a
s
s
;

1
s
t

y
e
a
r

0
.
3

0
.
3
3

t
o

0
.
4
9

1

-

-

N
a
c
h
t
i
g
a
l

-

"

-

a
f
t
e
r

c
l
e
a
r
i
n
g

-
"
-

2
n
d

y
e
a
r

a
f
t
e
r

c
l
e
a
r
i
n
g
*
2

0
.
7

1

-

-

-

-

-

-

l
a
n
d
-
u
s
e

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

P
a
n
i
c
u
m

r
i
g
h
t

a
f
t
e
r

c
u
t
t
i
n
g
.

8

-

0
.
0
5
8

3

-
"
-

-

-

-

-

m
a
x
i
m
u
m

1
4
%

c
a
n
o
p
y

c
o
v
e
r

D
i
g
i
t
a
r
i
a

s
l
o
p
e

1
2
%
;
p
l
a
n
t
e
d

3

0
.
0
0
7

3

B
r
a
z
i
l

A
l
a
g
o
i
n
h
a

L
e
p
r
u
n

e
t

a
l
.

(

1
9
8
6
.

d
e
c
u
m
b
e
n
s

y
e
a
r
s

o
l
d

p
a
s
t
u
r
e

(
C
=

p
.

2
2
6
)

m
e
a
n

o
f

3

y
e
a
r
s
)

B
r
a
c
h
a
r
i
a

I

s
t

y
e
a
r

0
.
3
8
7

-

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
l
.

d
e
c
u
m
b
e
n
s

(

1
9
8
4
)

-

-

f
u
l
l
y

e
s
t
a
b
l
i
s
h
e
d

0
.
0
0
2

-

-
-

-

-

-

-

p
l
a
n
t
e
d

p
a
s
t
u
r
e

(
C
=

m
e
a
n

0
.
0
0
3

6

B
r
a
z
i
l

B
r
a
s
i
l
i
a

L
e
p
r
u
n

e
t

a
l
.

(
1
9
8
6
.

1

I

o
f

s
i
x

y
e
a
r
s
)
:

5
.
5
%

s
l
o
p
e

p
.

2
2
8
)

"
1

S
u
r
p
r
i
s
i
n
g
l
y
,

s
a
v
a
n
n
a

f
a
l
l
o
w

h
a
d

h
i
g
h
e
r

r
e
s
i
d
u
a
l

e
f
f
e
c
t

w
h
i
c
h

m
i
g
j
t

b
e

d
u
e

t
o

g
e
n
e
r
a
l
l
y

l
o
w
e
r

o
r
g
a
n
i
c

m
a
t
t
e
r

l
e
v
e
l
s

a
n
d

s
l
o
w
e
r

t
u
r
n
o
v
e
r

"
2

r
e
s
i
d
u
a
l

e
f
f
e
c
t
s

f
o
r

2
n
d

y
e
a
r

a
s
s
u
m
e
d

a
s

5
0

%

o
f

t
h
e

I

s
t

y
e
a
r

*
3

O
n
l
y

t
h
e

c
r
o
p

r
e
s
i
d
u
e
s

a
r
e

l
e
f
t

i
n

t
h
e

f
i
e
l
d

a
n
d

t
h
e

s
p
o
n
t
a
n
e
o
u
s
l
y

g
r
o
w
i
n
g

v
e
g
e
t
a
t
i
o
n

a
f
t
e
r

h
a
r
v
e
s
t

m
e
a
n

e
x
t
r
e
m
e
s

T
a
b
l
e

3
4
-
2

A
n
n
e
x
:

D
e
t
a
i
l
e
d

C

f
a
c
t
o
r
s

f
o
r

b
a
n
a
n
a

m
o
n
o
c
r
o
p

8
7
~

s
l
o
p
e
.

l

s
t

y
e
a
r
:

l
e
a
v
e
s

0
.
5
6

0
.
1
4

t
o

1
.
0
8

p
l
a
c
e
d

a
r
o
u
n
d

t
r
u
n
k
s
;

2
n
d

y
e
a
r
:

l
e
a
v
e
s

p
l
a
c
e
d

o
n

c
o
n
t
o
u
r
;

s
p
a
c
i
n
g

5

x

3

m
.

A
l
t
e
r
n
a
t
i
v
e
l
y

f
o
r

a

y
o
u
n
g

p
l
a
n
t
a
t
i
o
n

(

1
s
t

y
e
a
r
)

-

-

a
s

a
b
o
v
e

b
u
t

s
p
a
c
i
n
g

2

x

3

m

0
.
1
6

0
.
0
4

t
o

0
.
3

-

-

a
s

a
b
o
v
e

b
u
t

s
p
a
c
i
n
g

3

x

3

m

0
.
3

0
.
0
8

t
o

0
.
5
8

-

-

a
s

a
b
o
v
e

b
u
t

s
p
a
c
i
n
g

4

x

3

m

0
.
4
2

0
.
1

t
o

0
.
8
3

-

-

w
i
t
h

m
u
l
c
h

0
.
0
0
0
2
9

0
.
0
0
0
2
9

t
o

0
.
0
0
0
3
6

n
o
.

[
a
1

2

B
u
r
u
n
d
i

M
a
s
h
i
t
s
i

R
i
s
h
i
r
u
m
u
h
i
r
w
a

(

1
9
9
2
.
p
.

9
0
)

l
a
n
d
-
u
s
e

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

2

-
"
-

-

-

2

-
"
-

-

-

2

-
"
-

-

-

I
v
o
r
y

A
d
i
o
p
o
d
o
u
m

R
o
o
s
e

(

1
9
7
5
,

C
o
a
s
t

p
.

3
0
1
3

1
)

6

7

8

9

-
L
.
-

a
s

n
o
.

3
.

s
p
a
c
i
n
g

3

x

3

m

0
.
0
0
0
9

0

t
o

0
.
0
0
7

2

B
u
r
u
n
d
i

-

"

-

-

-

p
l
u
s

c
o
m
p
l
e
t
e

m
u
l
c
h

c
o
v
e
r

-
.
*
-

0
.
0
4

L
e
w
i
s

(

1
9
8
6

i
n
:

Y
o
u
n
g
.

1
9
8
9
.

p
.

3
3
)

i
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

b
e
a
n
s

0
.
1

-

-

-
.
+
-

w
i
t
h

s
o
r
g
h
u
m

0
.
1
4

-

-

Detailed C factor\
I

I
m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

n
o
.

l
a
n
d
-
u
s
e

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

1
6

*

1

e
g
u
s
i

m
e
l
o
n

-

C
o
1
o
c
~
j
~
n
t
h
u
.
s

c
,
i
t
r
~
~
l
l
u
,
s

L
.

*
2

1
0
-
1
2

a
r
e

r
a
t
h
e
r

s
m
a
l
l

v
a
l
u
e
s
:

h
o
w
e
v
e
r
,

i
t

w
a
s

n
o
t

s
p
e
c
i
f
i
e
d

i
f

t
h
e

r
e
s
i
d
u
e
s

w
e
r
e

k
e
p
t

w
i
t
h
i
n

t
h
e

f
i
e
l
d

"
3

t
h
e
r
e

w
a
s

a

s
l
i
g
h
t

t
r
e
n
d

o
b
s
e
r
v
a
b
l
e

t
h
a
t

t
h
e

s
u
b
f
a
c
t
o
r

f
o
r

i
n
t
e
r
c
r
o
p
p
i
n
g

i
n
c
r
e
a
s
e
s

w
i
t
h

i
n
c
r
e
a
s
i
n
g

p
i
n
e
a
p
p
l
e

d
e
n
s
i
t
y
.

T
h
e

r
a
t
i
o
s

i
n
-

t
e
r
c
r
o
p
/
m
o
n
o
c
r
o
p

w
e
r
e

0
.
7
7
,

0
.
9
4

a
n
d

0
.
7
9

f
o
r

c
o
w
p
e
a

i
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

4
5
.
0
0
0
.

6
0
.
0
0
0

a
n
d

7
0
.
0
0
0

p
i
n
e
a
p
p
l
e

p
l
a
n
t
s
.

W
i
t
h

m
e
-

l
o
n

t
h
e

r
a
t
i
o
s

w
e
r
e

0
.
4
3
,

0
.
5
3

a
n
d

0
.
5
4
.

r
e
s
p
e
c
t
i
v
e
l
y
.

s
u
b
f
a
c
t
o
r
s

-

-

~
n
c
r
e
a
\
i
n
g

d
e
n
\
~
t
y

b
y

0
.
7
5

0
.
7
1

t
o

0

7
8

1

-

-

-

-

1
5
0
0
0

p
l
a
n
t
s
:

c
a
l
c
u
l
a
t
e
d

1
7
*
3

1
8
*
3

f
r
o
m

d
e
n
s
i
t
i
e
s

a
b
o
v
e

-

-

0
.
8
4

0
.
7
9

t
o

0
.
9
3

1

-

-

i
n
t
e
r
c
r
o
p
p
i
n
g

i
n
t
e
r
c
r
o
p
p
i
n
g

w
i
t
h

-

-

c
o
w
p
e
a

(
2
0
0
0
0

p
l
a
n
t
s
l
h
a
)

-

-

-

-

-
"
-

i
n
t
e
r
c
r
o
p
p
i
n
g

w
i
t
h

e
g
u
s
i

0
.
5

0
.
4
4

t
o

0
.
5
3

1

-

-

m
e
l
o
n
*

1

(
5
0
0
0

p
l
a
n
t
s
l
h
a
)

Annex 1.3
T(lhltj 14-4 Aizr1e.r: Detailed C,firctor* ,for cn.s.sm)n
Detailed C' factor\
m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

n
o
t
i
l
l

s
p
a
c
i
n
g

I

x

l

r
n
;
o
n
l
y

0
.
0
1
5

0
.
0
1
5

-
-

-
-

-

-

-

-

t
i
l
l
a
g
e

o
f

p
l
a
n
t
i
n
g

h
o
l
e
s
:

7

t
o

2
0
%

s
l
o
p
e
s

n
o
,

i
n
t
e
r
c
r
o
p

c
a
s
s
a
v
a

i
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

0
.
0
8

0
.
1
0
9

t
o

3

G
h
a
n
a

K
w
a
d
o
s
o

B
o
n
s
u

&

O
b
e
n
g

m
a
i
z
e

(
7
5
x

2
5

c
m
)

a
n
d

0
.
0
2
3

(
1
9
7
9
.
p
.

5
1
5
)

c
o
c
o
y
a
m
:

p
l
o
w
e
d

a
n
d

h
a
r
r
o
w
e
d
;

1
9
.

3
,

1
5

k
g
/
h
a

N
P
K
;

7
.
5
%

s
l
o
p
e

-

-

a
s

a
b
o
v
e

b
u
t

3
%
s
l
o
p
e

0
.
0
6
6

T

I

-
"
-

E
j
u
r
a

-

-

-

-

i
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

m
a
i
z
e
;

0
.
2
4

0
.
4
3
t
o

0
.
0
5

1

K
i
g
e
r
i
a

I
b
a
d
a
n

A
i
n
a

e
t

a
l
.

(

1
9
7
9
)

1
s
t

s
e
a
s
o
n

C

=

0
.
3
3
;

2
n
d

s
e
a
s
o
n

C

=

0
.
0
5

c
a
s
s
a
v
a

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

i
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

m
a
i
z
e
;

0
.
0
0
7

1

-
'
.
-

A
l
o
r
e

S
a
b
e
l
-
K
o
s
c
h
e
l
l
a

r
e
s
i
d
u
e
s

i
n
c
o
r
p
o
r
a
t
e
d
;

o
n

6
0

(

1
9
8
8
)

I

I

c
m

h
e
a
p
s

o
r
i
e
n
t
e
d

a
c
r
o
s
s

6
%

s
l
o
p
e
;

d
i
s
t
a
n
c
e

b
e
t
w
e
e
n

h
e
a
p

t
o
p
s

0
.
7

x

I

.
4

m
;

3

r
e
p
e
t
i
t
i
o
n
s

n
o
.

3
0

7
1

7
7

-
-

3
3

3
3

5

2
6

c
a
s
s
a
v
a

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

-

-

-

~
n
t
e
r
c
r
o
p
p
i
n
g
m
a
i
z
e

+

r
i
c
e

0
.
5
8
8

-

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
l
.

+

C
;
l
S
S
a
V
a

(

1
9
8
4
)

-
-
-

i
n
t
e
r
c
r
o
p
p
e
d

~
v
i
t
h

\
o
y
a

0
.
1
8
1

-

-

-

-

_
.
.
_

w
i
t
h

g
r
o
u
n
d
n
u
t

0
.
1
9
5

-

-

-

-

-

-

a
s

n
o
.

9

b
u
t

i
n
t
e
r
c
r
o
p
p
e
d

0
.
0
2

I

U
S
A
1
H
;
l
w
a
i
i

M
o
l
o
k
a
i

E
l
-
S
w
a
i
t
)

e
t

a
]
,
(

1
9
8
8
,

w
i
t
h

s
t
y
l
o
s
a
n
t
h
e
s

a
n
d

p
-

9
)

D
e
s
m
o
d
i
u
m

t
r
i
t
l
o
r
u
m

-

.
.

-

c
a
s
s
a
v
a

o
n

l
e
v
e
l

g
r
o
u
n
d
:

0
.
1
7

0
.
1
5

t
o

0
.
I
9

-

-

-

.
.

-

-

-

-

-

s
p
a
c
i
n
g

1
.
7
x

0
.
6

m

a
c
r
o
s
s

s
l
o
p
e
:

3

r
o
w
s

o
f

P
h
a
s
e
o
i
u
s

o
r

i
o
w
p
e
a

b
e
t
w
e
e
n

c
a
s
s
a
v
a
;

s
p
a
c
i
n
g

0
.
3
5

b
e
t
w
e
e
n

r
o
u
s

-

-
.
-

i
n
t
e
r
c
r
o
p
p
i
n
g

o
f

m
a
i
z
e

+

0
.
3
5
7

-

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
1
.

r
i
c
e

+

c
a
s
s
a
c
.
a

w
i
t
h

r
e
s
i
d
u
e

(

1
9
8
4
)

m
u
l
c
h

i
n
t
e
r
c
r
o
p

+

i
n
t
e
r
c
r
o
p
p
e
d

w
i
t
h

r
i
c
e

+

0
,
0
7
9

-

-

-

-

l
n
u
l
c
h

m
a
i
z
e

p
l
u
s

6

t
l
h
a

o
f

r
i
c
e

\
t
r
a
w

n
~
u
l
c
h

2

3

\

?

A

m

2
.

-

m

:<

n

T

$
.

5
.

-

-

-

Q

Annex 3.4
T~rblr 34-4 Arznrx, c.ontivzuc,
- 0
C 3
3
L
bCI
k i'
%
8
4
2
G
a
4
. TJ -
4
a"
2
2
2
ri
0,
-
P a
QI
a
u
L
u
.- -
e
.- CI
u
-
+a
u
1
8
I
E u a
z g . z a I
2 E & -
b">
L.
u
U
2
C 0
.-
-
.%
L
b!
4
s
6
C
- - -
"8 00 d " o^
2 E 6
-
% 0
- 0 2 , z3.
d -. e
-
cf - C . f - - Z -
p, -
I-- 'U
Y a p - c f
'" c
3
- . -
u z
- c f - J2 5 2 2 2 s
g .go 2 2 ~ 2 2 . g z
g 2 %
k s a c Q - J z 5
a m 0 a
&i L ~ K < = Z E S ~
E
3 3
W . -
a
d
cf
&
. -
-a e
d 4
s
, $ 4 k i
-
z
. . . - a
m
u LC V:
. -
X
'U
J;
X
2 2 o
cf
a
4 2 , %
2 2 I t:
: : s
- - , , , ?
C
I-- -
v l ? ?
-
i ; ; I I I ,
c4
7 r!
3 3
"!
C1 CI w Tt
g s a 0
7 5 2 2
3 0
-
y 3
O C *
# -g
- 3 s 3
2 .. h, 3 ~ z ~ G . s C k
0
0 cf
L go? KO) P L
; ~ z e $ , x g E - =
g ~ ~ $ w , * - - O 2
x 8; - 2. E z :
* o m
g c ~ d F ...; e b .s .= 6 S 5 .$
o I- b ~ Q. s
- ~ ~ ' U C g $ g ~ c . - e,
g 2 i 5 c z 2 g e
z,o "
* W O 3 . . , 3 - a , 0 , , J ; v : c r
a
2
0 C I
3 -
I I , I
. * 4 -
E - 1
. . . .
, I I , I
- P4 F, -T m at--
Annex 3.4
T
a
b
l
e

3
4
-
6

A
n
n
e
x
:

D
e
t
a
i
l
e
d

C

f
a
c
t
o
r
s

f
o
r

m
a
i
z
e

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

p
l
o
w
*
2

s
p
a
c
i
n
g

0
.
9

x

0
.
5

m
;

s
l
o
p
e

0
.
1
6

0
.
0
2

t
o

0
.
1
2

3
'
"

1

Z
i
m
b
a
b
w
e

D
o
m
b
o
s
h
a
w
a

V
o
g
e
l
(
1
9
9
2
.
.

p
.

1
3
)

n
o
.

3
.
5

7
~
;

c
o
r
r
e
c
t
e
d

f
o
r

c
o
n
t
o
u
r
i
n
g

(
0
.
5
)

.
.

$
:
?

-

-

s
p
a
c
i
n
g

0
.
9

x

0
.
6
2
m
:

s
l
o
p
e

0
.
1
8

4
.
5

%
.
.

c
o
r
r
e
c
t
e
d

f
o
r

c
o
n
t
o
u
r
i
n
g

(
0
.
5
)

-

-

0
.
5
1

t
r
e
a
t
m
e
n
t

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

0
.
0
4
t
o
0
.
1
7

3
*
1

-

-

M
a
k
a
h
o
l
i

0
.
5

U
S
A
/

M
o
l
o
k
a
i

E
l
-
S
w
a
i
f
y

e
t

H
a
w
a
i
i

a
l
.

(

1
9
8
8
.
.

p
.

4
)

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
l
.

(

1
9
8
4
)

M
a
l
a
y
s
i
a

S
e
r
d
a
n
g

M
o
k
h
t
a
r
u
d
d
i
n

&

M
a
e
n
e

(

1
9
7
9
)

-

-

S
u
l
a
i
m
a
n

e
t

a
l
.

(
1
9
8
3
)

1
2

%

s
l
o
p
e
;

a
l
o
n
g

s
l
o
p
e

0
.
1
1

-

3

B
r
a
z
i
l

A
l
a
g
o
i
n
h
a

L
e
p
r
u
n

e
t

a
l
.

(
1
9
8
6
.
,

p
.

2
2
6
)

6

-

-

5
.
5

%

s
l
o
p
e
:

h
a
n
d

h
o
e

.

-

t
i
l
l
a
g
e
;

0
.
5
6

-

-

s
p
a
c
i
n
g
:

0
.
6

m

b
e
t
w
e
e
n

0
.
8
2

-

3

r
o
w
s
;

2
0
0
0
0

t
o

5

1
0
0
0

p
l
a
n
t
s
l
h
a

d
e
p
e
n
d
i
n
g

o
n

r
a
i
n

v
o
l
u
m
e
;

3
0

-

6
0

k
g
/
h
a

N
.
.

3
0

B
r
a
s
i
l
i
a

-

-

K
e
n
y
a

K
a
t
u
m
a
n
i

U
l
s
a
k
e
r

&

K
i
l
e
w
e

(

1
9
8
4
.
.

p
.

2
3
3
)

Drlailed C factors
n
o
.

2
9

3
0

3
1

3
2

3
3

3
4

3
5

3
6

3
7

3
8

t
r
e
a
t
m
e
n
t

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

-

.b

-
:
5
4

s
p
a
c
i
n
g

0
.
9

x

0
.
5
;

w
i
t
h

0
.
0
1
2

-

3

Z
i
m
b
a
b
w
e

D
o
m
b
o
s
h
a
w
a

V
o
g
e
I
(

1
9
9
2
,
,
p
.

1
3
)

m
a
i
z
e

r
e
s
i
d
u
e
s
;

c
o
r
r
e
c
t
e
d

f
o
r

c
o
n
t
o
u
r

p
l
a
n
t
i
n
g

-
-

-
*
4

s
p
a
c
i
n
g

0
.
9

x

0
.
6

m
;

w
i
t
h

0
.
0
1

1

0
.
0
1
6

t
o

3

-

-

M
a
k
a
h
o
l
i

-

'
*

-

m
a
i
z
e

r
e
s
i
d
u
e
s
:

c
o
r
r
e
c
t
e
d

0
.
0
1
7

f
o
r

c
o
n
t
o
u
r

p
l
a
n
t
i
n
g

-
"
-

7
.
5

Y
e

s
l
o
p
e
;

s
p
a
c
i
n
g

7
5

x

0
.
0
3

1

0
.
0
0
6
3

t
o

3

G
h
a
n
a

K
w
a
d
o
s
o

B
o
n
s
u

&

O
b
e
n
g

2
5

c
m
,
.

f
e
r
t
i
l
i
z
e
r

1
9
.
.

3

a
n
d

0
.
0
7
7

(
1
9
7
9
.
.
p
.

5
1
5
)

1
5

k
g
l
h
a

o
f

N
P
K

I

-
-
-

a
s

a
b
o
v
e
;

b
u
t

3

C
/
c

s
l
o
p
e

0
.
0
3
5

-
*
5

-

-

E
j
u
r
a

-

-

r
e
d
u
c
e
d

s
p
a
c
i
n
g
:

0
.
7
5

x

0
.
2
5

m
:

7
.
5

0
.
0
2

0
.
0
3
8

t
o

3

-

-

K
w
a
d
o
s
o

-

"

-

t
i
l
l
a
g
e

5
%

s
l
o
p
e
;

1
9
.
,

3

a
n
d

1
5

0
.
0
0
5
7

k
g
l
h
a

o
f

N
,
,

P

a
n
d

K
:

p
l
o
w
e
d

w
i
t
h
o
u
t

h
a
r
r
o
w
i
n
g
:

t
w
o

h
a
n
d
w
e
e
d
i
n
g
s

0
.
0
4
1

-
*
5

1

-

-

E
j
u
r
a

a
s

a
b
o
v
e
;

3

5
%

s
l
o
p
e

-

"

-

-

-

s
p
a
c
i
n
g

0
.
7
5

x

0
.
2
5

m
;

0
.
0
0
2
9

0
.
0
0
5
6
3

t
o

3

-

-

m
u
l
c
h

K
w
a
d
o
s
o

-
"
-

f
e
r
t
i
l
i
z
e
r

1
9
.
.

3

a
n
d

1
5

k
g
l
h
a

0
.
0
0
1
3

o
f

N
,
,

P

a
n
d

K
;

p
l
o
w
e
d

a
n
d

h
a
r
r
o
w
e
d
:

m
u
l
c
h

q
u
a
n
t
i
t
y
'
?

a
s

m
u
l
c
h

p
l
o
t

a
b
o
v
e
;

3

%

0
.
0
0
4

-
*
5

1

-

-

E
j
u
r
a

-
"
-

-

-

s
l
o
p
e
:

m
u
l
c
h

q
u
a
n
t
i
t
y
?

-
"
-

w
i
t
h

i
m
p
e
r
a
t
a

m
u
l
c
h

0
.
0
2

-

M
a
l
a
y
s
i
a

-

S
u
l
a
i
m
a
n

e
t

a
l
.

(
1
9
8
3
)

i
n
t
e
r
c
r
o
p
p
e
d
1

w
i
t
h

g
r
o
u
n
d
n
u
t

p
l
u
s

3

t
/
h
a

0
.
1
2
8

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
l
.

m
u
l
c
h

s
t
r
a
w

m
u
l
c
h

(
1
9
8
4
)

m
e
n
t

p
e
r
i
o
d

n
o
.

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

r
i
d
g
e

s
p
a
c
i
n
g
:

0
.
7
5

x

0
.
2
5

m
:

3

'3
-

0
.
0
2
6

0
.
0
5
7

t
o

3

G
h
a
n
a

K
w
a
d
o
s
o

B
o
n
s
u

&

O
b
e
n
g

(

1
9
7
9
,
,

s
l
o
p
e
;

1
9
.
.

3

a
n
d

1
5

k
g
l
h
a

o
f

0
.
0
0
8
7

p
.

5
1
5
)

N
.
,

P

a
n
d

K
:

p
l
o
w
e
d

a
n
d

t
r
e
a
t
m
e
n
t

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

h
a
r
r
o
w
e
d
;

r
i
d
g
e
s

a
c
r
o
s
s

s
l
o
p
e

-

-

r
i
d
g
e

o
n

c
o
n
t
o
u
r
;

7
.
5

5%
s
l
o
p
e

0
.
0
5
4

-
*
5

I

-

-

E
j
u
r
a

-

-

-

-

s
p
a
c
i
n
g
:

I
0
0

x

2
0

c
m
.

o
n

0
.
6
7

0
.
2
5

t
o

0
.
9
5

I
v
o
r
y

A
d
i
o
p
o
d
o
u
m

R
o
o
s
e

(

1
9
7
5
,
,

r
i
d
g
e
s

a
l
o
n
g

s
l
o
p
e

C
o
a
s
t

p
.

3
0
1
3

I
)

t
i
e
d
-
r
i
d
g
i
n
g
-

s
p
a
c
i
n
g

0
.
9

x

0
.
5
:

r
i
d
g
e
s

o
n

0
.
0
1
8

0
.
0
0
6

t
o

3

Z
i
m
b
a
b
w
e

D
o
m
b
o
s
h
a
w
a

V
o
g
e
I
(
1
9
9
2
.
,

p
.

1
3
)

/
n
o
t
i
l
l
*
5

c
o
n
t
o
u
r
;

2
5

c
m

h
i
g
h

c
o
n
t
o
u
r

0
.
0
2
6

r
i
d
g
e
s

w
i
t
h

1

%

l
a
t
e
r
a
l

s
l
o
p
e
;

t
i
e
s

a
t

i
n
t
e
r
v
a
l
s

o
f

1
.
5

m

a
n
d

a
p
p
.

1
5

c
m

h
i
g
h

-

-

"

*
5

s
p
a
c
i
n
g

0
.
9

x

0
.
6
;

r
i
d
g
e
s

o
n

0
.
0
0
4

0
.
0
0
3

t
o

3

-

-

M
a
k
a
h
o
l
i

-

-

c
o
n
t
o
u
r
,

2
5

c
m

h
i
g
h

c
o
n
t
o
u
r

0
.
0
6

r
i
d
g
e
s

w
i
t
h

1

'3
%

l
a
t
e
r
a
l

s
l
o
p
e
;

t
i
e
s

a
t

i
n
t
e
r
v
a
l
s

o
f

1
.
5

m

a
n
d

a
p
p
.

1
5

c
m

h
i
g
h

*

1

O
n
l
y

d
a
t
a

o
f

t
h
e

2
n
d

t
o

4
t
h

e
x
p
e
r
i
m
e
n
t
a
l

y
e
a
r

w
e
r
e

c
o
n
s
i
d
e
r
e
d
.

R
e
s
u
l
t
s

o
f

1
s
t

y
e
a
r

s
k
i
p
p
e
d

d
u
e

t
o

r
e
s
i
d
u
a
l

e
f
f
e
c
t
s

o
f

n
a
t
u
r
a
l

f
a
l
l
o
w

"
2

P
l
o
w
i
n
g

w
i
t
h

o
x
e
n

d
r
a
w
n

s
i
n
g
l
e

f
u
r
r
o
w

m
o
u
l
d
-
b
o
a
r
d

p
l
o
w

r
i
g
h
t

a
f
t
e
r

h
a
r
v
e
s
t

t
o

2
0

-

2
5

c
m

d
e
p
t
h

$
3

R
i
p
p
i
n
g

b
e
t
w
e
e
n

f
o
r
m
e
r

m
a
i
z
e

r
o
w
s

w
i
t
h

a
n

o
x
e
n

d
r
a
w
n

s
i
n
g
l
e

r
i
p
p
e
r

t
i
n
e

t
o

2
0

-

2
5

c
m

d
e
p
t
h

:
k
4

A
s

"
-
3
b
u
t

r
e
s
i
d
u
e
s

o
f

f
o
r
m
e
r

m
a
i
z
e

c
r
o
p

l
e
f
t

"
5

S
o
i
l

w
a
s

p
l
o
w
e
d

t
o

2
0

-

2
5

c
m

d
e
p
t
h

a
t

b
e
g
i
n
n
i
n
g

o
f

e
x
p
e
r
i
m
e
n
t
s
.

2
5

c
m

h
i
g
h

c
o
n
t
o
u
r

r
i
d
g
e
s

(
I

%

s
i
d
e

s
l
o
p
e
)

w
e
r
e

f
o
r
m
e
d

w
i
t
h

o
x
e
n

d
r
a
w
n

m
o
u
l
d
b
o
a
r
d

r
i
d
g
e
r
.

I
n

s
u
b
s
e
q
u
e
n
t

y
e
a
r
s
.
.

n
o

p
l
o
w
i
n
g

w
a
s

c
a
r
r
i
e
d

o
u
t

b
u
t

r
i
d
g
e
s

w
e
r
e

r
e
f
o
r
m
e
d

a
f
t
e
r

h
a
r
v
e
s
t

a
n
d

6

w
e
e
k
s

a
f
t
e
r

p
l
a
n
t
i
n
g
.

1
5

c
m

h
i
g
h

c
r
o
s
s

t
i
e
s

w
e
r
e

a
l
s
o

f
o
r
m
e
d

t
w
i
c
e

a

y
e
a
r

a
t

1

m

i
n
t
e
r
v
a
l
s

i
n

t
h
e

f
u
r
r
o
w
s
.

P
l
a
n
t
i
n
g

w
a
s

o
n

t
o
p

o
f

t
h
e

r
i
d
g
e
s
.

*
6

P
l
a
n
t
i
n
g

h
o
l
e
s

w
e
r
e

o
p
e
n
e
d

w
i
t
h

h
a
n
d
-
h
o
e

(
b
a
d
z
a

h
o
l
i
n
g

o
u
t

i
n

Z
i
m
b
a
b
w
e
)

P

\
O

Annex 3.4
Detailed C I'aclon
a
1 cI
L
*
.- -
G
.-
Y
2
-
+.
u
G
3
8
I
2 , -
2 E . E 7
Z E g -
r/l
L L
u
C,
CJ
e
U
E
.-
u
a
.C
L
2
-e
a
L
6
-
F,
g
-
-
.'
-
.y
0
E
. -
rJ
-
3
"
. -
x
, a'
-
S i
6
30
?
y
-
"
6\
2 ;
Eo :
. -
Z 8
- PI
30
p!
C
9
2
-
- -
rc,
6
-
- 2 g
-
X
9 $
0
-
-
-
- -
- a - a
*
p z Q ar
C
Z S
-.
-
-
F, 2
-
G 2
- 0
Lg 2
= -
2
. -
$ 2
h, 1 2 . z 3
-
u p , : 0 7 C ! 3
m - ~ ~ c l , gvl a 0 s L
".
r
. -
. -
C
J.
E
2 , I
-
, < ., 1 , 2
-S 3
. - . -
J.
- a, = 5,
.-
a s=
-
- a
, 5 5 s
' 'C:
3
-
0
!
C
I P
-
rc,
!
C
LC
b
E l ,
I-.
3\ 5-
-
x g
-22
c
2
2 2
f: d
3
. - -
I
z, : ,
r, Cc' '
I I I
.-
g % s
g G G
5 s
" '3
. 3 - ; C !
z 5 - I
5 :, :,
I
0 0 0 - - -
?-. -
-
- 3 c
a
+
5 6
, g ,I g +
*
g - , Q g
cr,
- 2
. .
blJ -
+
5 6
a ' 3
$ 2 ,
J. r
'" 'c
-
n
o
.

1
7

1
8

1
9

2
0

2
1

2
2

2
3

2
4

c
r
o
p

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

-
-
-

p
l
o
w
e
d
:

m
a
i
z
e

r
e
s
i
d
u
e
s
:

6
%

0
.
0
0
6

0
.
0
0
2

t
o

2

-
"
-

-

-

N
i
l
1

(
1
9
9
3
)

s
l
o
p
e
;

s
p
a
c
i
n
g

7
5

x

2
5

c
m
;

0
.
0
1
4

2

r
e
p
e
t
i
t
i
o
n
s

-
a
-

n
o
t
i
l
l
:

m
a
i
z
e

r
e
s
i
d
u
e
s
;

0
.
0
1

0
.
0
0
0
4

t
o

2

-

"

-

I
b
a
d
a
n

L
a
1

(

1
9
7
6
a
,
p
.

3

1
)

p
l
a
n
t
e
d

a
l
o
n
g

s
l
o
p
e

0
.
0
2

-
-
'
-

n
o
t
i
l
l
:

m
a
i
z
e

r
e
s
i
d
u
e
s
:

6
%

0
.
0
0
4

0
.
0
0
2

t
o

2

-
"
-

A
l
o
r
e

S
a
b
e
l
-
K
o
s
c
h
e
l
l
a

s
l
o
p
e

0
.
0
0
6

(

1
9
8
8
)

-

.
'

-

n
o
t
i
l
l
:

m
a
i
z
e

r
e
s
i
d
u
e
s
;

6
%

0
.
0
0
0
3

0
.
0
0
0
2

t
o

2

-

"

-

-

-

N
i
l
1

(
1
9
9
3
)

s
l
o
p
e
:

s
p
a
c
i
n
g

7
5

*

2
5

c
m

0
.
0
0
0
4

a
c
r
o
s
s

s
l
o
p
e

I
r
i
s
h

p
o
t
a
t
o
e
s

-

0
.
2
2

-

R
w
a
n
d
a

-

L
e
w
i
s

(

I
9
8
6

i
n
:

Y
o
u
n
g

1
9
8
9
,

p
.

3
3
)

l
e
m
o
n

g
r
a
s
s

0
.
4
3
4

-

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
l
.

(

1
9
8
4
)

p
a
p
a
y
a

w
i
t
h
o
u
t

c
o
v
e
r

c
r
o
p

2
.
1

-

-

A
b
d
u
l

R
a
s
h
i
d

(
I
9
8
1

i
n
:

S
u
l
a
i
m
a
n

e
t

a
l
.

1
9
8
3
)

s
o
y
a

0
.
3
9
9

-

I
n
d
o
n
e
s
i
a

-

A
b
d
u
r
a
c
h
m
a
n

e
t

a
l
.

(
1
9
8
4
)

n
o
t
i
l
l

(
p
r
o
b
a
b
l
y

w
i
t
h
o
u
t

0
.
1
0
3

r
e
s
i
d
u
e
s

l
e
f
t

b
e
c
a
u
s
e

n
o
t
i
l
l
-

s
u
b
f
a
c
t
o
r

=

0
.
6
7
)
:

s
l
o
p
e

2

3
-

7

'
4

P

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

-

2
5

2
6

2
7

n
o
.

c
r
o
p

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

-
"

-

0
.
1
1

0
.
0
2

t
o

0
.
1
9

1

N
i
g
e
r
i
a

I
b
a
d
a
n

A
i
n
a

e
t

a
l
.

(

1
9
7
9
)

-

.
-

p
l
a
n
t
e
d

o
n

c
o
n
t
o
u
r

o
n

4

o
u
t

0
.
3
8

1

I
n
d
o
n
e
s
i
a

-

K
e
e
r
s
e
b
i
l
c
k

(

1
9
9
0
.

p
.

5
7
0
)

o
f

8

d
i
f
f
e
r
e
n
t

I
n
d
o
n
e
s
i
a
n

s
o
i
l

t
y
p
e
s

-

'b

-

h
a
n
d

t
i
l
l
a
g
e
;

s
l
o
p
e

5
.
5
%
;

0
.
1
5
4

B
r
a
s
i
l
i
a

6

B
r
a
z
i
l

L
e
p
r
u
n

e
t

a
l
.

(

1
9
8
6
.

p
.

2
2
8
)

2
9

3
0

3
1

U

?

4

b

-
2

n

0

5
.

%

5
.
5
6

0
.
2
3

-

R
w
a
n
d
a

-

L
e
w
i
s

(

I
9
8
6

i
n
:

s
w
e
e
t

Y
o
u
n
g

1
9
8
9
,

p
.

3
3
)

p
o
t
a
t
o
e
s

0
.
5

-

-

t
o
b
a
c
c
o

2
n
d

c
y
c
l
e

R
o
o
s
e

(
1
9
7
5
,

p
.

4
0
)

0
.
4
5

-

R
w
a
n
d
a

-

L
e
w
i
s

(
1
9
8
6
i
n
:

t
o
b
a
c
c
o

Y
o
u
n
g

1
9
8
9
.

p
.

3
3
)

.

Annex 3.4
Detailed C factors
n
o
.

6

7

8

9

1
0

1
1

1
2

n
o
t
i
l
l
a
g
e

d
e
s
c
r
i
p
t
i
o
n

C

f
a
c
t
o
r

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

-
"
-

s
e
e
d

s
l
i
t

r
i
p
p
e
d

b
e
t
w
e
e
n

0
.
8
1

0
.
6

t
o

1
.
1
8

4

Z
i
m
b
a
b
w
e

D
o
m
b
o
s
h
a
w
a

V
o
g
e
l

(

1
9
9
2
,
p
.

1
3
)

f
o
r
m
e
r

m
a
i
z
e

s
t
u
b
b
l
e

(
a
c
r
o
s
s

s
l
o
p
e

o
f

4
.
5
%
)

-
"
.

a
s

a
b
o
v
e

0
.
5
1

0
.
3
7

t
o

1
.
0

4

-

"

-

M
a
k
a
h
o
l
i

-

"

-

w
i
t
h

r
e
s
i
d
u
e
s

c
a
l
c
u
l
a
t
e
d

f
r
o
m

c
a
s
s
a
v
a
;

0
.
1
7

0
.

l

I

t
o

0
.
3
7

2

y
e
a
r
s

C
o
l
o
m
b
i
a

S
a
n
t
a
n
d
e
r

d
e

R
e
i
n
i
n
g

(
1
9
9
I
,

p
.

1

I

1
)

s
l
o
p
e
s

7

-

1
3
%

w
i
t
h

3

Q
u
i
l
i
c
h
a
o

r
e
p
s

-
"
-

a
s

a
b
o
v
e

b
u
t

s
l
o
p
e
s

1
3

-

0
.
1
0
4

0
.
0
6

t
o

0
.
2
7

2

y
e
a
r
s

-

"

-

M
o
n
d
o
m
o

"

2
0
%

w
i
t
h

2

r
e
p
s

-
"
-

m
a
i
z
e

r
e
s
i
d
u
e
s

l
e
f
t

i
n

f
i
e
l
d
;

0
.
4
1

0
.
3

t
o

1
.
1
8

4

Z
i
m
b
a
b
w
e

D
o
m
b
o
s
h
a
w
a

V
o
g
e
l

(

1
9
9
2
,

p
.

1
3
)

s
l
o
p
e

7

-

1
3
%

-
+
&
-

a
s

a
b
o
v
e

b
u
t

s
l
o
p
e

1
3

-

2
0
%

0
.
2
9

0
.
1

I

t
o

0
.
7
7

4

-

"

-

M
a
k
a
h
o
l
i

-

"

-

-
"
-

c
a
l
c
u
l
a
t
e
d

f
r
o
m

w
h
e
a
t
-

0
.
1
4

6

B
r
a
z
i
l

G
u
a
i
b
a

L
e
p
r
u
n

e
t

a
l
.

(

1
9
8
6
.

p
.

m
a
i
z
e

r
o
t
a
t
i
o
n

p
l
a
n
t
e
d

o
n

2
3
0
)

1
2
%
s
l
o
p
e

Annex 4 Protection and management
Annex 4.1 Detailed support and management (P) factors
Detailed \upport and management ( P) tactor5
Table 41-2Annex: Detailed P factors for ridges
T
a
b
l
e

4
1
-
3
A
n
n
e
x
:

D
e
t
a
i
l
e
d

P

f
a
c
t
o
r
s

f
o
r

m
o
u
n
d
s

n
o
.

d
e
s
c
r
i
p
t
i
o
n

P

f
a
c
t
o
r
*
'

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

I

c
a
s
s
a
v
a

1
.
1
3

i
I
v
o
r
y

C
o
a
s
t

A
d
i
o
p
o
d
o
u
m

R
o
o
s
e

(

1
9
7
5
.
p
.

3
9
)

m
o
u
n
d
s

2

p
i
n
e
a
p
p
l
e

0
.
2
9

-

-

-

-

m
o
u
n
d
s

T
a
b
l
e

4
1
-
4
A
n
n
e
x
:

D
e
t
a
i
l
e
d

P

f
a
c
t
o
r
s

f
o
r

b
u
n
d
s

n
o
.

m
e
t
h
o
d

d
e
s
c
r
i
p
t
i
o
n

P

f
a
c
t
o
r
*
'

m
e
a
s
u
r
e
-

c
o
u
n
t
r
y

l
o
c
a
t
i
o
n

l
i
t
e
r
a
t
u
r
e

m
e
n
t

p
e
r
i
o
d

m
e
a
n

e
x
t
r
e
m
e
s

[
a
]

I

s
t
o
n
e

b
u
n
d
s

3
%

s
l
o
p
e
;

3
0

c
m

h
i
g
h

s
t
o
n
e

0
.
2
7

1

N
i
g
e
r

A
l
l
o
k
o
t
e

R
o
s
e

&

b
u
n
d
s

e
v
e
r
y

8
0

c
m

v
e
r
t
i
c
a
l

B
e
r
t
r
a
n
d

d
i
s
t
a
n
c
e
,

o
n

c
o
n
t
o
u
r
;

I

s
t

y
e
a
r

(
1
9
7
1
.

p
.

1
2
7
6
)

3

-
-
-

a
s

a
b
o
v
e

b
u
t

2
n
d

a
n
d

3
r
d

y
e
a
r

0
.
0
5

-

-

3

e
a
r
t
h
e
n

b
u
n
d
s

l
i
k
e

s
t
o
n
e

b
u
n
d
s

n
o
.

I

b
u
t

w
i
t
h

0
.
0
0
2
2

-

-

l
e
s
s

s
t
o
n
e
s

w
h
i
c
h

c
o
v
e
r

a
n

e
a
r
t
h
e
n

c
o
r
e

v
e
g
e
t
a
t
e
d

b
y

n
a
t
u
r
a
l

h
e
r
b
e
s

4

-
.
.
-

a
s

a
b
o
v
e

b
u
t

2
n
d

y
e
a
r

0
.
0
4

1

-
"
-

-

-

-

-

1)e~ailrd support and rn:uiagcment (1') [actor-\
T(rhlu 4 1-5 Alrilc..~: Dcr(ril~d P, / i rc~ot . sfi )r. bidfor- srr'iljs
Some usefill species for soil and water conservation
Annex 4.2 Some useful species for soil and water conservation
Tuhle 42- 1 Aizrze.~: U.st$ul speri e. ~ , f i r ero.sioi~ c.ontt-01, greerz manurirzg,
rnulcl? or- (.over c-rap
Annex 4.2
b
o
t
a
n
i
c
a
l

c
o
m
m
o
n

r
e
m
a
r
k
s

l
o
n
g
e
v
i
t
y

e
c
o
l
o
g
y
*
'

n
a
m
e

n
a
m
e

b
i
o
t
e
m
-

r
a
i
n
f
a
l
l

p
H

d
r
o
u
g
h
t

w
a
t
e
r
l
o
g
-

s
a
l
t

s
h
a
d
e

l
i
t
e
r
a
-

p
e
r
a
t
u
r
e

[
m
m
]

g
i
n
g

t
u
r
e

M
u
c
u
n
a

c
a
-

v
e
l
v
e
t

b
e
a
n

l
e
g
m
i
n
o
u
s

c
o
v
e
r

a
n
d

a

'?

h

'?

7

7

7

3

1

p
i
t
a
t
a

g
r
e
e
n

m
a
n
u
r
i
n
g

c
r
o
p

P
a
n
i
c
u
m

G
u
i
n
e
a

g
r
a
s
s

g
o
o
d

o
n

r
i
s
e
r
s
;

f
o
d
d
e
r

p

>

2
0

6
0
0
-

4
.
5
-
6
.
0

1

m
a
x
i
m
u
m

4
0
0
0

P
a
s
p
a
l
u
m

B
a
h
i
a

g
r
a
s
s

c
o
v
e
r

c
r
o
p

a
n
d

e
r
o
s
i
o
n

p

>
I
2

8
0
0
-

5
.
1
-
>
8
.
1

1

n
o
t
a
t
u
m

c
o
n
t
r
o
l

1
0
0
0

P
e
n
n
i
s
e
t
u
m

K
i
k
u
y
u

g
r
a
s
s

e
r
o
s
i
o
n

c
o
n
t
r
o
l

i
n

P

8
-
2
4

6
0
0
-

5
.
1
-
6
.
0

1

c
l
a
n
d
e
s
t
i
-

h
i
g
h
l
a
n
d
s

4
0
0
0

n
u
m

P
e
n
n
i
s
e
t
u
m

e
l
e
p
h
a
n
t

f
o
d
d
e
r

a
n
d

m
u
l
c
h

P

>

8

6
0
0
-

4
.
5
-
5
.
5

1

p
u
r
p
u
r
e
u
m

g
r
a
s
s

4
0
0
0

S
t
y
l
o
s
a
n
-

s
t
y
l
o

l
e
g
m
i
n
o
u
s

c
o
v
e
r

a
n
d

p

>

3
0

1
0
0
0
-

4
.
5
-
5
.
0

-

1

t
h
e
s

g
u
i
a
-

g
r
e
e
n

m
a
n
u
r
i
n
g

c
r
o
p

4
0
0
0

n
e
n
s
i
s

V
i
g
n
a

u
n
-

c
o
w
p
e
a

l
e
g
u
m
e

a

>
1
3

6
0
0
-

4
.
5
-
>
8
.
1

1

g
u
i
c
u
l
a
t
a

3
0
0
0

E
u
p
h
o
r
b
i
a

l
i
v
e

h
e
d
g
e
.

p
r
o
p
a
g
a
t
e
d

p

1

b
a
l
s
a
m
i
f
e
r
a

b
y

c
u
t
t
i
n
g
s

o
r

s
e
e
d
s

J
a
t
r
o
p
h
a

p
o
u
r
g
h

r
e

l
i
v
e

h
e
d
g
e

s
e
e
d
e
d
:

s
l
o
w

p

7

c
u
r
c
a
s

e
s
t
a
b
l
i
s
h
m
e
n
t
:

c
u
t
t
i
n
g
s

s
e
v
e
r
e
l
y

a
t
t
a
c
k
e
d

b
y

t
e
r
m
i
t
e
s

b
o
t
a
n
i
c
a
l

c
o
m
m
o
n

r
e
m
a
r
k
s

n
a
m
e

n
a
m
e

A
g
a
v
e

s
i
s
a
-

s
i
s
a
l

l
i
v
e

h
e
d
g
e
:

s
o
m
e
t
i
m
e
s

l
a
n
a

n
o
t

a
c
c
e
p
t
e
d

c
l
o
s
e

t
o

h
o
m
e
s
t
e
a
d
s

(
h
i
d
e
-
a
w
a
y

f
o
r

s
n
a
k
e
s
)

A
n
d
r
o
p
o
-

G
a
m
b
a

g
r
a
s
s

e
r
o
s
i
o
n

c
o
n
t
r
o
l
;

f
o
d
d
e
r

g
o
n

g
a
y
a
n
u
s

l
o
n
g
e
v
i
t
y

e
c
o
l
o
g
y
"

b
i
o
t
e
m
-

r
a
i
n
f
a
l
l

p
H

d
r
o
u
g
h
t

w
a
t
e
r
-

s
a
l
t

s
h
a
d
e

l
i
t
e
r
a
-

p
e
r
a
t
u
r
e

[
m
m
]

l
o
g
g
i
n
g

t
u
r
e

P

>

2
3

<

3
0
0
-

5
.
6
-
>

8
.
1

+

3

-

B
r
a
c
h
a
r
i
a

C
o
n
g
o

g
r
a
s
s

e
r
o
s
i
o
n

c
o
n
t
r
o
l
:

f
o
d
d
e
r
;

p

>

I

7

,
?

-

3

r
u
z
i
z
i
e
n
s
i
s

q
u
i
c
k

e
s
t
a
b
l
i
s
h
m
e
n
t
.

r
e
g
r
o
w
t
h

v
e
r
y

k
a
r
i
a
b
l
e

S
t
y
l
o
s
a
n
-

e
r
o
s
i
o
n

c
o
n
t
r
o
l
:

r
a
p
i
d

p

7

>

?

?

1

,?

'?

2

t
h
e
s

h
a
m
a
t
a

r
e
g
r
o
w
t
h

T
e
p
h
r
o
s
i
a

g
r
e
e
n

m
a
n
u
r
e

o
n

c
l
a
q
e
)

p

=
.
2
1

1
0
0
0
-

1
.
5
-
7
.
5

-

3

v
o
g
e
l
i
i

s
o
i
l
s

1
0
0
0

P
h
a
c
e
l
i
a

t
a
-

p
h
a
c
e
l
i
a

f
o
d
d
e
r
.

c
o
\
e
r

a
n
d

g
r
e
e

a

>

7

)

I

9

,?

'?

3

n
a
c
e
t
i
f
o
l
i
a

m
a
n
u
r
i
n
g

c
r
o
p

B
o
t
h
r
i
o
-

T
u
r
k
e
s
t
a
n

r
e
c
u
l
t
i
v
a
t
i
o
n

o
f

e
r
o
d
e
d

p

8
-

I
0

1
0
0
0
-

6
.
1
-
7
.
0

+

3

c
h
l
o
a

b
l
u
e
s
t
e
m

p
a
s
t
u
r
e
s

1
1
0
0

1
i
s
e
h
a
e
m
u
m

I

P
u
e
r
a
r
i
a

k
u
d
z
u

l
e
g
u
m
i
n
o
u
s
,

c
r
e
e
p
i
n
g

p

>

2
2

1
0
0
0
-

4
.
5
-
7
.
0

-

+

+

3

p
h
a
s
e
o
l
o
i
d
e
s

c
o
\
e
r

a
n
d

g
r
e
e
n

3
0
0
0

m
a
n
u
r
i
n
g

c
r
o
p
.

f
o
d
d
e
r

Some u\ et ' ~~I \pccics f or soil and warcr conscr-vation
lJ\el'ul trees
Further information on suitable speci es can be found in:
200 to 500
Acacia albida
Acacia radiana
Acacia senegal
Annona senegalensis
Balanites aegyptiaca
Roscia salicifolia
Commiphora africana
Conocarpi~s lancifolius
Dobera glabra
Eupliorbia balsamifera
Mael-va crassili)lia
Parkinsonia aculeata
I'rosopis juliflora
Zi ~i phus spp.
Young ( 1 989)
Hudson ( 1975)
von Maydell ( 1983)
Merlier & Montegut ( 1982)
ICRAFIGTZ Multipurpose TI-ce & Shrub Database
annual precipitation [mm]
500 to 900
Adansonia digitata
Anacardiuln digitata
Azadil-achta indica
Bauhinia spp.
Cassia siamea
Combrctum spp.
Eucalyptus
camaldulensis
Ficus syco~norus
Haxoxylon persicurn
Parkia biglobosa
Salvadora persica
Sclerocarya birrea
Tamarix articulata
Terminalia spp.
900 to 1200
Al bi ~i a lebbeck
Anocgeissus
leiocarpus
Borassus aethiopum
Butyrospermum parki i
Casuarina
equisetifolia
Cordia abyssinica
Dalbergia
melanoxylon
Erythrina abyssinica
Mat-khamia spp.
Tamarindus indica
Literature
Literature
Abdul Rashid, F. D. (1975): Studies on the correlation between the
erodibility of a standard sand and five Malaysian soils. Unpublished
R. Agr. Sc. Project Paper, University of Malaya, Kuala Lumpur,
Malaysia.
Abdurachman, A.; Abujamin, S. & Kurnia, U. (1984): Soil and crop
management practices for erosion control. Pernbl-. Pen. Tanah Dan
Pupuk 3: 7- I?.
Aina, P. 0; Lal, R. & Taylor, G.S. (1979): Effects of vegetal cover on soil
erosion on an Alfisol. In: L A , R. & Greenland, D.J. (eds.): Soil
physical properties and crop production in the tropics. John Wiley &
Sons, Chichester: 50 1-508.
Allison, F.E. (1973): Soil organic matter- and its role in crop production.
Develop-nients in Soil Science 3, Elsevier, Amsterdam: 637 pp.
Ambassa-Kiki, R. bi Lal, R. (1992): Surface clod size distribution as factor
influencing soil erosion under silnulated rain. Soil Tillage Res. 22:
3 1 1-32?.
Armon, M.N. (1984): Soil crosion and degradation in south-eastern Nigeria
in relation to biophysical and socio-economic factors. PhD-thesis,
Depart~i~ent of Agronomy, University of Ibadan, Nigeria.
Arnoldus, H.M.J. (1977): Methodology used to dctermine the maximum
potential average annual soil loss due to sheet and rill erosion in
Marocco. In: Assessing soil degradation. FA0 Soils Bull. 34: 39-48.
Arnoldus, H.M.J. (1978): An approxiniation of the rainfall factor in the
Universal Soil Loss Equation. In: DeBoodt, M. & Gabriels, D. (eds.):
Assessment of erosion, John Wilcy, New York: 127-1 32.
Asoegwu, S.N. & Obiefuna, J.C. (1990): Preliminary evaluation of
pineapple mixed cropping systems for protecting reclaimed gullys in
the tropics: an experiment in south-east Nigeria. Land Degradation &
Rehabilitat~on 2, 3: 237-241.
Auerswald, K. & Eicher, A. (1992): Comparison of German and Swiss
rainfall simulators: Accuracy of mcasurement and effect of rainiall
sequence on runoft' and soil loss rates. Z. Ptlanzenernihr. Bodcnk.
155: 191-195.
Auerswald, K.; Kainz, M.; Schriider, 1). & Martin, W. (1992a):
Cornpal-ison of German and Swiss rainfall simulators - Expcril-nent:~l
setup. Z. Pflanzcnerniihr. Bodenk. 155: 1-5.
Literature
Auerswald, K.; Kainz, M.; Wolfgarten, H.J. & Botschek, J. (1992b):
Comparison of German and Swiss rainfall simulators - Influence of
plot dimensions. Z. Pflanzenernahr. Bodenk. 155: 493-497.
Auerswald, K.; Sinowski, W. & Hausler, W. (1992~): Gew" asserver-
sauerung und Gewiissereutrophieri~ng - iikologische Folgen lateraler
Stofffliisse in Biiden. Landwirtschaftliches Jahrbuch 2: 87-96.
Barfield, B.J. & Albrecht, S.C. (1982): Use of a vegetative filter zone to
control fine- grained sediments f'rom sul-face mines. Symposium on
Surfaces Mining Hydrology, Sedimentology and Reclamation, Univ.
of Kentucky, Lexington, Kentucky, Dec. 5- 10, 1982; 48 1-490.
Becher, H.H. (1990): Cornparison of German and Swiss rainfall simulators -
Relative drop energies and their distribution in time and space for
simulated rains. Z. Pflanzenerniihr. Bodenk. 153: 409-4 14.
Bennett, H.H. & Chapline, W.R. (1928): Soil erosion a national menace.
US Dep. Agric., Circ. 33.
Rlaikie, P.M. (1985): The political economy of soil erosion in developing
countries. Longman, London.
Bonsu, M. (1980): Assessment of erosion under different cultural practices
on a savanna soil in the northern region of Ghana. In: Morgan, R.P.C.
(ed.): Soil conservation, problems and prospects. Proc. Int. Conf. Soil
Conservation, Silsoe, U.K., John Wiley 8r Sons, New York: 247-253.
Bonsu, M. & Obeng, H.B. (1979): Effects of cultural practices on soil
erosion and maize production in the serni-deciduous rainforest and
forest-savanna transitional zones of Ghana. In: Lal, R. 8r Greenland,
D.J. (eds.): Soil physical properties and crop production in the tropics.
John Wiley & Sons, New York: 509-5 19.
Bouwman, A.F. (1989): Modelling soil organic matter decomposition and
rainfall erosion in two tropical soils after forest clearing for
permanent agriculture. Land Degradation & Rehabilitation, 1 : 125- 140.
Bresch, J. (1993): Die Erosivitiit der Niederschliige Kameruns. M .Sc. thesis.
Lehrstuhl fiir Bodcnkunde, Technische Universitat Miinchen,
Gerl ~~any: 92 pp.
British Westafrican Meteorological Services (1954): Preliminary note on
the climate of Nigeria. Manuscript: 4 pp.
Bundesamt fur Seeschiffahrt und Hydrographie (1991 a): Handbuch der
Westkiiste Afrikas. NI-. 2040. Hamburg, Germany: 520 pp.
Bundesamt fur Seeschiffahrt und Hydrographie (1991 b): Handbuch der
Siid- ilnd Ostkiiste Afrikas. Nr. 2046. Hamburg. Germany: 344 pp.
Rundesamt fur Seeschiffahrt und Hydrographie (1992):
Literature
Mittelmeerhandbuch. 111. Teil (1989). Nachtrag Nr. I . Nr. 2029.
Hamburg, Germany: 363 pp.
Cackett, K.E. (1964): A simple device for measuring canopy cover. Rhod. J.
Agric. Res. 2: 56-57.
Caine, N. (1980): The rainfall intensity-duration control of shallow
landslides and debris flow. Geografiska Annaler. 62A: 23-27.
Carrara, P.E. & Carroll, T.R. (1979): The determination of erosion rates
from exposed tree roots in the Piceance Basin, Colorado. Earth
Surface Processes 4: 307-3 17.
Carter, C.E.; Greer, J.D.; Braud, H.J. & Floyd, J. M. (1974): Raindrop
characteristics in south central United States. Transactions of' the
ASAE 17, 6: 1033- 1037.
Charreau, C. & Nicou, R. (1971): L'amelioration du profil cultural dans les
sols sableux de la zone tropicale skche Ouest-Africaine et ses
incidences agronorniques. Agron. Trop., 26. 9: 903-978.
Chauvel, A.; Pedro, G. & Tessier, D. (1976): R6le du fer dans
I'organisation des matkriaux kaoliniques. Science du Sol 2: 10 1 - 1 13.
Chin & Tan (1974): Effects of simulated erosion on the performance of
maize (Zea mays) and groundnut (Arachis hypogeae): a preliminary
assessment. Syrnp. on Classification and Management of Malaysian
Soils, Kota Kinebalu, Malaysia.
Chorley, R.J. (1978): The hillslope hydrological cycle. In: Kirkby, M.J.
(ed.): Hillslope hydrology. John Wiley & Sons, New York: 1-41.
Chromec, F.W.; El-Swaify, S.A. & Lo, A.K.F. (1989): Erosion proble~ns
and research in Hawaii. In: Crouch, R.J. & Collison, J.E. (1989): An
air injected, single nozzle rainfall simulator, designed for use on gi~lly
sides. Australian J. Soil Water Cons. 2, 3: 37-39.
CIEH (1985): Isohyktes des pluies journalikres de frequence dkcennale.
Cornit6 Interafricain d'etudes hydrauliques, B.P. 369, Ouagadougou,
Burkina Faso.
CTFT (1973): Difense et ristauration des sols. Gampela 1967-72. CTFT
(Ouagadougou), Burkina Faso.
Croouch, R.J. & Collison, J.E.: An air injectcd. single nozzle rainfall
simulator, designed thr use on gully sides. Austr. J. Soil Water Cons..
2. 3: 37-39.
Dangler, E.W. & El-Swaify, S.A. (1976): Erosion of selectcd Hawaii soils
by sirnu-lated rainfall. Soil Sci. Soc. Am. J. 40: 769-773.
&lwaulle, J.C. (1973): Resultats de six ans d'obversations s i ~r I'krosion uu
Niger. Bois et ForCts des Tropiclues 150: 15-37.
Ilissmeyer, G.E. & Foster, G.R. (1980): A guide for predicting shcet and
rill erosionon forest land. U.S. Dept. Agric., Forest Serv. Tech. Publ.
SA TP- I I : 40pp.
Duke, J.A.; Terrell, E.E. (1974): Crop diversification matrix: Introduction.
Taxon, 23, 5/6: 759-799.
Dumas, J. (1965): Relation entre 1'6rodibilitk des sols et leurs
caractkristiques analytiques. Cah. Orstom, Ser. Pedal. 3, 4: 307-333.
Dunne, T. (1975): Sediment yields of Kenyan rivers. Unpublished report.
Dunne, T. (1978): Field studies of hillslope processes. In: Kirkby, M.J. (ed.):
Hillslope hydrology. John Wiley & Sons Ltd.: 227-293.
Dunne, T.; Dietrich, W.E. & Brunengo, M.J. (1978): Recent and past
erosion rates in semi-arid Kenya. Z. Geomorph. N. F., Supp1.-Bd. 29:
130- 140.
Durand, P. (1983): Lutte contre l'krosion. Synthese des essais apres la
catnpagne 1983 5 Rushubi. Manuscript: '?pp.
Ekwue, E.I. (1991): The effects of soil organic matter content, raintidl
duration and aggregate size on soil detachment. Soil Technology 4:
1 97-207.
El-Swaify, S.A. (1990): Research needs and applications to reduce erosion
and sedimentation in the tropics. Proceed. Fiji Syrnp., June 1990,
IAHS-AISH Publ. 192: 3- 13.
El-Swaify, S.A. & Dangler, E.W. (1977): Erodibility of selected tropical soils
in relation t o structural and hydrologic parameters. In: Soil erosion:
Prediction and control. Soil Cons. Soc. Am. Spec. Publ. 2 1 : 105- 1 14.
El-Swaify, S.A. & Dangler, E.W. (1982): Rainfall erosion in the tropics: A
state of the art. In: Soil erosion and conservation in the tropics. Soil
Sci. Soc. Am. Spec. Publ. 43: 1-25.
El-Swaify, S.A.; Lo, A.; Joy, R.; Shinshiro, L. & Yost, R.S. (1988):
Achieving conservation-effectiveness in the tropics using legume
intercrops. Soil Technology I: 1 - 12.
Elwell, H.A. (1980a): A soil loss estimation technique for Southern Africa.
In: Morgan, R.P.C. (ed.): Soil conservation, problems and prospects.
Proc. Int. Conf. Soil Conservation, Silsoe, U.K., John Wiley & Sons,
New Y ork: pp. 28 1-292.
Elwell, H.A. (1980b): Design of safe rotational systerns. Department of
Conservation and Extension, Zimbabwe, Manuscript: 50 pp.
Elwell, H.A. (1984): Soil loss estimation: a inodelling technique. In: Hadley,
R.F. & Walling, D.E. (eds.): Erosion and sediment yield - Some
methods of measurement and modelling. University Press, Cambridge,
pp 15-30.
Elwell, H.A.; (1993): Summary of crop covcr mesurements for erosion and
runoff prediction and control 1959 to 1993 and 1984 analysis results.
The Institi~te of Agricultural Engineering, Harare, Zimbabwe: 32 pp.
Elwell, H.A. & Gardener, S. (1975): Comparison of two techniques for
measuring percent canopy cover of row crops in erosion research
programmes. Department of Conservation and Extension, Salisbury,
Zimbabwe, Research Bulletin 19.
Elwell, H.A. & Stocking, M.A. (1976): Vegetal cover to estimate \oil
ero\ion hazard in Rhodesia. Geoderrna 15: 6 1-70.
Elwell, H.A. & Wendelaar, F.E. (1977): To initiate a vegetal cover data
bank for \oil loss e\tirnation. Dcpart~iient of Conservation and
Exten\ion, Salisbury, Zimbabwe, Ke\ea~-ch Bulletin 23: 42 pp.
Engelund, F. & Hansen, E. (1967): A nomograph on \ediment tran\port In
alluvial streams. Teknick Verlag, Copenhagen, Denmark.
Falayi, 0. & Lal, R. (1979): Efl'ect of aggregate size and mulching on
cl-odibility, crusting, and crop ernergence. In: Lal, R. Nr Greenland,
D.J. (cds.): Soil physical properties and crop production in the tropics.
J ohn Wiley & Sons, Ncw York: 87-93.
F A0 (1990): The conservation and rehabilitation of African lands. Koline,
Italy: 38 pp.
F A0 (1991): Network on erosion-induced loss i n soil productivity. Report of
a Workshop, Centre for Soil & Agroclimate Research Bogor,
Indonesia, March 199 1 : 52 pp.
FA0 (1993): The conservation and rehabilitation of African lands. FAO.
Rorne, ARCl9014: 38 pp.
Firth, C.R. & Whitlow, R. (1991): Patterns of gullying in Zimbabwe.
Geojournal 23, 1 : 59- 67.
Flanagan, D.C.; Foster, G.R. & Moldenhauer, W.C. (1988): Storm pattern
effect on infiltration, riunoff, and erosion. Transactions of the ASAE
3 1.2: 414-420.
Fones-Sundell, M. (1992): Survival for whom? Farmer, government and
donor perspectives on land degradation in sub-saharan Afl-ica. In:
Tato, K. & Hurni, H. (eds.): Soil conservation for survival. Soil Water
Cons. Soc.: 1 18- 124.
Foster, G.R. (1982): Modeling the erosion process. In: Haan. C.T.: Johnson.
H.P. & Rrakcnsiek, D.L. (eds.): Hydrologic rnodeling of small
watersheds. ASAE Monograph 5: 297-380.
Foster, G.H. & Highfill, R.E. (1983): Effect of ter-races o n soil loss: USLE
P factor valucs lijr terraces. J . Soil and Water Cons. 38. 1 : 48-5 1 .
Foster, G.H.; I,ombardi, F. & Moldenhauer, W.C. (1982): Evaluation of'
rainfall-runoff erosivity factors for individual storms. Transactions of'
the ASAE 25, 1 : 124- 129.
Foster, G.R. & Meyer, L.D. (1972): Mathematical simulation of upland
erosion by fundamental erosion mechanics. Present and Prospective
Techn. for Predicting Sediment Yields and Sources. Proc. Sediment
Yield Workshop, USDA Sediment Lab., Oxford Miss., Nov. 28-30.
1972, ARS-S-40: 190-204.
Foster, G.K.; Moldenhauer, W.C. & Wischmeier, W.H. (1982):
Transferability of U.S. technology for prediction and control of
erosion in the tropics. Am. Soc. of Agronomy and Soil Sci. Soc. Am.
(eds.): Soil erosion and conservation in the tropics: 135- 149.
Foster, G.R.; Weesies, G.A.; Renard, K.G.; Yoder, D.C. & Porter, J.P.
(1992): Support practice factor (P). In: Renard, K.G.; Foster, G.R.;
Weesies, G.A.; McCool, D.K. & Yoder, D.C. (eds.): The Revised
Universal Soil Loss Equation. Draft Version 1992. US Dept. of
Agricul ture.
Fournier, F. (1962): Carte du danger d'erosion en Afrique au sud du Sahara.
Cornmunaut6 Economique Europienne avec Commission de
Coopiration Technique en Afrique, Paris.
Fournier, F. (1967): La recherche en krosion et conservation des sols dans le
continent afi-icain. Sols Africains 12, 1 : 5-5 1 .
Galabert, J. & Millogo, E. (1973): Indice d'6rosion de la pluie en Haute
Volta. CTFT, (Ouagadougou), Burkina Faso: 34 pp.
Gasser, W. & Ziibisch, M.A. (1988): Erdrutschungen und Mafinahmen der
Hang-sicherung. Selbstverlag des Verbandes der Tropenlandwirte,
Witzenhausen, Germany: 165 pp.
Ghadiri, H. & Payne, D. (1981): Raindrop impact stress. J . Soil Sci. 3, 2:41-
49.
Griffiths, J.F. (1972): In: Landsberg, H.E. (ed.): World Survey and
Climatology. Elsevier, Amsterdam 10: 197- 192.
Griggs, G.B. & Gilchhurst, J.A. (1977): The earth and landuse planning.
University of California, Santa Cruz, Duxbury Press.
Gunn, R. & Kinzer, G.D. (1949): The terminal velocity of fall for water
droplets in stagnant air. J. of Meterology 6: 243-248.
Hettler, J. (1994): Hiichste Gefahrenstufe fiir die Wiilder im Wasser.
Frankfurter Rundschau, 19. 07. 1 994: 6.
Herkendell, J. & Koch, E. (1991): Bodenzerstiirung in den Tropen.
Beck' sche Reihe 436, Germany: 184 pp.
Hijkoop, J.; Van der Poel, P. & Kaya, B. (1991): Une lutte de longue
haleine. Amenagements anti-erosifs et gestion de terroir. Systemes de
production rurale au Mali: Volume 2. Institut d'Economie Rurale,
Bamako, Mali et Institilt Royal des Tropiques, Amsterdam, Pays-Bas:
154 pp.
Hoeblich, J.M. (1992): Le lavaka malgache: Une fhl-me d'krosion parfois
utilisable. Reseau Erosion Bulletin 12: 255-268.
Horton, R.E.; Leach, H.R. & Van Vliet, R. (1934): Laminar sheet-flow.
Amer. Geophys. Union Trans. 2: 393-404.
Hudson, N.W. (1961): Raindrop size distribution, kinetic energy and
intensity of' sub- tropical rainfall: 127- 133.
Hudson, N.W. (1963): Raindrop size distribution in high intensity storms.
Rhod. J. Agric. Rcs. 1 : 6- 1 1 .
Hudson, N.W. (1975): Field engineering for agricultural development.
Clarendon Press. Oxford. UK: 225 pp.
Hudson, N.W. (1986): Soil conservation. Batsford Lirni ted, London, UK:
324 pp.
Hudson, N.W. (1991): A study of the reasons for success or failure of' soil
conservation projects. FA0 Soils Bull. 64: 65 pp.
Hudson, N.W. & Jackson, D.C. (1959): Results achieved in the
measurement of erosion and runoff in southern Rhodesia. Third Inter-
African Soils Conference, Dalaba, Africa Soils Bureau 2: 575-584.
Hurni, H. (1980): A nomograph for the design of' labour-intensive soil
conservation measures in rain-fed cultivations. In: Morgan, R.P.C.
(ed.): Soil conservation, problems and prospects. Proc. Int. Conf. Soil
Conservation, Silsoe, U.K., John Wiley & Sons, New York: 185-210.
ICRAFIGTZ (1991): The multi-purpose tree and shrub database. Version 1 .
Kainz, M. (1989): Runoff, erosion and sugarbeet yields in conventional and
mulched cultivation. Results of the 1988 experiment. Soil Technology
Serie 1 : 103- 1 14.
Kainz, M.; Auerswald, K. & Viihringer, R. (1992): Comparison of German
and Swiss rainfall sirnulators - Utility, labour demands and costs. Z.
Pflanzenerniihr. Bodenk. 155: 7- 1 1 .
Kamphorst, A. (1987): A small rainfall si~nulator for the determination of
soil erodibility. Netherlands Journal of Agricultural Sciences 35: 407-
4 15.
Kaye, C.A. (1959): Shoreline features and quaternary shoreline changes.
Puerto Rico. U.S. Geological Survey Professional Papers No. 3 17-B:
49- 1 40.
Keersebilck, N.C. (1990): Erosion control in the tropics. In: De Boodt. M.F.;
Hayes, M.H.B. & Herbillon, A. (eds.): Soil colloids and their
associations in aggregates. Plenum Press, New York: 567-576.
Kilewe, A.M. & Mbuvi, J.P. (1987): Evaluation of soil erodibility factors
using natural runoff plots. East Afr. Agric. For. J. 53, 2: 57-63.
Kirkby, M..J. (1980) (ed.): Hillslope hydrology. John Wiley & Sons, New
York: 43-72.
Kittler, G.A. (1962): BodenflulIS. Eine von der Agrarriiorphologie
vernachlassigte Erscheinung. Bundcsanstalt fiir Landeskunde und
Raumforschung, Selbstverlag, Bad Godesberg, Germany.
Kollmannsperger, F. (1979): Erosion - eine globale Gefahr. Deutsche
Gesellschaft fiir Technische Zusammenarbeit, Eschborn: 63 pp.
Kanig, D. (1992): Thc potential of agroforestry methods for erosion control
in Rwanda. Soil Technology 5: 167- 176.
Kowal, J.M. & Kassam, A.H. (1977): Energy load and instantaneous
intensity of rainstorms at Sa~naru. northern Nigeria. In: Greenland.
D.J. & LaI, K. (eds.): Soil conservation and management in the humid
tropics. John Wiley & Sons, New York: 57- 70.
La Marche, V.C., Jr. (1968): Rates of slope degradation as determined from
botanicril evidence. White Mountains, Califoi-nia. US Geological
Survey Protssional Paper 352, 1 : 34 1-377.
Laflen, J. M. ; Foster, G.R. & Onstad, C.A. (1985): Simulation of
individual storm soil loss for modeling the impact of soil erosion on
crop productivity. In: El-Swaify. S.A.: Moldenhauer, W.C. & Lo, A.
(eds.): Soil erosion and conservation. Soil Cons. Soc. Am.. Arikcny,
Iowa.: 285-295.
I,al, R. (1976a): Soil erosion proble~ns on an Alfisol i n westel-n Nigeria anti
their control: 1ntcrn;itional Institute for Tropical Agriculture.
Monograph No. 1 . Ibadan, Nigeria: 208 pp.
Lal, K. (1976b): Soil erosion on AISisols in Western Nigeria, 111. Effects of
rainfall characteristics. Geodel-ma 16: 389-40 1 .
I,al, R. ( 1 980): Soi I conservation: Preventive and control measures. I n:
Morgan. K.P.C. (cd.): Soil conservation. prohlerns and prospects.
Proc. Int. Conl'. Soil Conservation, Silsoe, 1J.K.. John Wiley & Sons,
New York: 175- 1 8 1 .
Lal, R. (1983): Soil erosion and its relation to pr-oductivity in tropical soils.
In: El-Swaify, S.A.; Moldenhauer, W.C. &L Andrew. L.O. (edh.1:
Proceedings oT the International Conference of Soil Erosion and
Conservation, Hawaii: 237-247.
Ial, R. (1990): Soil erosion in the tropics. Mc Graw Hill, lnc.. New York:
580 pp.
Lal, R. (1902): Restoring land degraded by gully erosion in the tropics. Adv.
Soil Sci., 17: 123- 152.
Lal, R.; Lawson, T.L. & Anastase, A.H. (1980): Erosivity of tropical rains.
I n: DeBoodt, M. & Gabriels. D. (eds.): Assessr-nent of erosion. John
Wiley. New York: 143- 12 1 .
Larsen, M.C. & Simon, A. (1993): A rainfiill intensity duration thr-eshold
for landslides in a hutnicl-tropical environment. Puerto-Rico.
Geografiska Annaler. 75 A, 1-2: 13-23.
Laws, J.O. (1941): Measurements of the hll-velocity of water-drops and
I-aindrops. Transacl. Am. Gcophys. Union: 709-722.
Laws, J.O. & Parsons, [).A. (1943): The relation of raindrop-size to
inten\ily. Transact. Am. Geopllys. llnion 24: 452-460.
[>envain, J.S.; Sakala, W.K. & Pauwelyn, P.L.L. (1988): Isoerodent map
of' Zambia Part 11: Erosivity prediction and rnapping. Soil Technology
1 : 25 1-262.
Leprun, J.C.; Silveira, C.O. da & Sobral Filho, R.M. (1986): Efficacitk
des praticlues culturales ;untikrosives testkes sous diff61-cnts climats
brdsiliens. Cah. Orstom, Ser. P@dol. 22.2: 223-233.
Levy, G.J. & Van der Watt, H.H. (1988): Effects of clay mineralogy and
soil sodicity on soil infiltration rate. S.-Afr. Tyclskt-. Plant Grond 5. 2:
92-96.
Lewis, L.A. (1986): Predicting soil loss in Rwanda. In: Beek. K..I.:
Burrough. P.A. & Mc Cormack. D.E. (cds. ): Quantified land
e\.aluation procedures. Pi-oc. of the Int. Workshop. Washington DC.
27 April - 2 May 1986: 137- 139.
Luk, S. (1983): Effect of aggregate size and microtopogsaphy on rainwash
kind rainsplash erosion. Z. Geomorph. N.F.. 27, 3: 283-295.
Lyon, T.C. & Buckman, H.O. (1922): The nature and properties of soils.
The Mac-millan Co., New York.
Maene, I,.M. & Chong, C.P. (1979): Drop size distribution and erosivity of'
tropical rainstorms under tlic oil palm canopy. Laporan Penyelidikan.
Jabatan Sains Tanah, University Pcrtanian. Malaysia, Serdang.
Sclangor: 8 1-93.
Mazour, M. (1992): Lcs Sac~eurs cle riscl~~c dc I'irosion en nappe dans Ic
bassin versant d'Issel- Tlerncen Algcric. Reseau Erosion Bulletin 12:
300-3 1 3.
Mbagwu, J.S.C.; Lal, R. & Scott, T.W. (1984): Effects of desurftlcing Of
All'i\ol\ and Ul t ~\ ol \ In wuthern Niger~a: I. Crop pel-formance. So11
Sci. Soc. Am. J . 48: 828-833.
McCool, D.K.;Krown, I,.C.;Foster, G.R.;Mutchler, C.K.& Meyer, L.1).
(1987): R e v i d \lope stecpnes\ factor l-'o~ the Univer\al So11 Lo\\
Ecli~ation Am. Soc. Agr. E,ng..Mlch~gan: 1387- 1396.
McCool, D.K.; Foster, G.K.: Mutchler, C.K. 8r Meyer, L.D. (1980):
Kev~\ ed \lope length factor 1'01- the Unlver\al Soil Lo\\ Equation.
Tran\act~on\ of the ASAk 32, 5: 157 1 - 1576.
McCool, D.K.; Foster, G.R. & Weesies, G.A. (1992): Slope length and
\tccpne\\filctors (1,s) I n: Kcnard, K.G.: Foster, G.R.; Wee\ ~e\ , G.A.;
McCool, D.K.&L Yoder, 1>.C.(ed\.): Predict~ng \oil ero\ron by water:
A guide to conscrvation planning with the Revised Universal Soil
Loss Equation (RUSLE), Draftversion 1992.
Merlier, H.; Montegut, J. (1982): Adventices tropicale\: Flore aux \tade\
plantale el adulte de 123 esp$ce\ africaines ou pantropicales.
Ministkre des Relations ExtCricure(;, Coopel-ation ct D6vcloppenient.
France.
Meyer, L.D. (1988): Rainfill1 si111uIato1-s f'oi- soil el-osion research. In: Lal. R.
(ed. ): Soil erosion research methods. Soil Water COIIS. Soc. Am.,
Ankeny, Iowa: 75-95.
Meyer, I,.D. & Harmon, W.C. (1985): Scdi~iient losscs from cropland
furrows of diffel-cnt gradient\. Transactions of the ASAE 28, 2: 348-353.
Moeyersons, J . (1989): La nature de 1'6rosion des versants au Rwanda.
Institut Nat. de Recherche Scicntifiquc, Hutare, Republique
Rwandaise. Pub. 33: 370 pp.
Mokhtaruddin, A.M. & Maene, I,.M. (1979): Soil erosion under ciifl'el-ent
crop\ and management practices. In: Proc. Intern. Conf. on Agric.
Engineel-ing in National Devcloprncnt, September 1979, Scrdang.
Malay\ia, lJniver\ity of Pertania, Malaysia.
Moore, T.R. (1979): Rainl-all ermivity in East Africa. Creogl-afisha Annalel-
6 1 A. 3-41 147- 156.
Morgan, K.P.C. (1986): Soil ero\ion and conservation. Longman, UK: 298
PP-
Mostafa, K.T. & Osama, A.E.K. (1992) (eds.): The world environriient
1972- 1992. TWO decades of challenge. Chapman &L Hall, London,
England.
Musgrave, G.W. (1947): The yuantitative evaluation of factor\ in watcr
el-o\iol~, a first approximation. J . Soil Water Cons. 2, 3: 133- 138.
Mtakwa, P.W.; I,al, R. & Sharma, R.H. (1987): An evaluation of the
Universal Soil Loss Equation and field techniques for assessing soil
erosion in a tropical Aif'isol in Western Nigeria. Hydrological
Processes I: 199-209.
Muller, J.P. (1977): Ida rnicrolyse plasniique et la differenciation des
kpipkdons dans Ies sols I~I-rallitiq~~es rouges du Centl-e-Ca~neroun.
Cali. Orstom, Ser-. Pkdol. 15. 4: 345-359.
Murphee, C.E. & Mutchler, C.K. (1981): Veril'ication of' the slope factor in
the Universal Soil Loss Equation 1'0s low slopes. J . Soil Water Cons.
36, 5: 300-302.
Mutchler, C.K. & Greer, J.D. (1980): Effect of slope length on erosion
from low slopes. Transactions of the ASAE 23, 4: 866-869, 876.
Mutchler, C.K. & McCregor, K.C. (1983): Erosion from low slopes. Water
Resources Research, 19, 5: 1323- 1326.
Mutchler, C.K. & Young, K.A. (1975): Soil detachment by raindrops. U.S.
Agric. Res. Serv. Kep. ARS-S-40: 1 13.
Nelson, H.D. (1984) (ed.): Mo~ambiquc - a country study. Headquartel-s.
i1epa1-trnent of' the Army, DA Par11 550-64, USA: 341 pp.
Ngatunga, K.I,.N.; ],a], R. & Uriyo, A.P. (1984): Effects of surfitce
management 011 r ~~nof f and soil erosion from some plots at Mlingano.
Tanzania. Creodcri~~a 33: 1 - 12.
Nill, D. (1993): Soil erosion from natural and si~nulated rain in forest-,
savannah- and highland areas of humid to sub-humid West Africa and
influence of management. PhD thesis, Lehrstuhl fiir Bodenkunde,
Technische Universitiit Miinchcn, Germany: 270 pp.
Nyagumbo, I. (1992): The influence of socio-econo~~lic fiictors o n potential
adoption of' no- till tied-ridging in four coriir-nunal areas of Zi~nbabwe.
In: Vogcl, H.: Conservation tillage Ihr sus1ainable crop production
syst e~~i s. Pro.ject Report no. 5, P.O.Box BW 41 5, Bol-rowdale, Harare.
Zimbabwe: 19-32.
Nyamulinda, V. (1991): Erosion sous cultures de versants et transports
solides dans les bassins de drainage des hautes teri-es de Ruhengeri au
Rwanda. Reseau Erosion Bulletin 1 1 : 38-63.
O'Laughlin, C.L. (1981): Tree I-oots and slope stability. Forest Research
Institute. No. 104, Christchurch, New Zealand.
Oostwoud Wijdunes, D.J. & Bryan, R.B. (1991): Gully devclopmcnt on the
Nje~nps Flats. Baringo, Kenya. Catena Supplc~nent 19: 7 1-90.
Othieno, C.O. (1975): Surface runoff and soil erosion 011 fields of young tea.
Trop. Agriculture 52, 4: 209-308.
Owens, I,.B. (1974): Rates ot' weathering ancl soil formation on granite.
Literature
Unpubl. report, Dep. Agric. Univ. o f Rhodesia.
Patil, P.Y. & Bangal, G.B. (1991): Effects of land slope and cultivation
practices on soil loss and runoff due to rainfall. J . of Maharashtra
Agric~lltural Universities 16, 2: 155- 158.
Pauwelyn, P.L.L.; Lenvain, J.S. & Sakala, W.K. (1988): Iso-erodent map
o f Zambia part I: The calculation of erosivity indices frorn a rainfall
data bank. Soil Technology I: 235-250.
Petri, R. (1992): Erosionsdisposition (C-Faktoren) traditioneller
Anbausysterne Kamesuns. M.Sc. thesis, Lehrstuhl fiir Bodenkunde,
Technische Universitat Miinchen, Germany: 80 pp.
Pierce, F.J.; I,arson, W.E.; Dowdy, R.H. & Graham, W.A.P. (1983):
Productivity of soils: A\se\\ing long-term changes due to erosion. J.
Soil Water Cons.: 39-44.
Pimentel, D.; Terhune, E.C.; Dyson-Hudson, R. & Rocherau, S. (1976):
Land degradation: Effects on food and energy resources. Science 94:
140- 155.
Poesen, J. (1984): The influence of slope angle on infiltration rate and
Hoi-tonian overland flow volume. Z. Geomorph. N. F.. Suppl.-Bd. 49:
117-131.
Pontanier, R.; Moukouri Kouh, H.; Sayol, R.; Seiny-Boukar, I,. &
Thebe, B. (1984): Comportemcnt hydriclue et \ensibilitk 5 IqCro\ion
de quelques 501s du Nol-d Cameroun. Ministhe de I'Enseignement
Supkrieure et de la Recherche Scientifique. Carneroun: 76 pp.
Quansah, C.; Raffoe-Bonnie, E. & Agyei, F. (1990): Runoff and soil lo\\
under four legume\. In: Ztibisch, M.A. (ed.): Land use and the
environment. Proc. I Ith Ann. Meeting of the Soil Science Soc. o f
Ghana: 67-75.
Raussen, T. (1990): Planung von Windschut~-Ptlan7unger1 in aridcn und
semiariden Gebieten. Der Tropenlandwirt, Beihef't 43: 232 pp.
Rehm, S. & Espig, C;. (1976): Die Kulturpllan7en der Tropen und
Subtropen. Eugen Ulmer, Stuttgart, Germany: 496 pp.
Reining, L. (1991): Charakterr\ierung und Verminderung der Bodenero5ion
durch Washer in kleinbuuerlichen Maniokanbausy\temen. Beitrage /us
Agrarwi\senschaften 4, Verlag w. Mahle, Wittenchlick. Bonn.
Germany: 266 pp,
Renard, K.G.; Foster, G.R.; Weesies, G.A. & Porter, J.P. (1991):
RUSLE- Reviscd U~liversal Soil Losg Equation. J. Soil Water Con\.
46, 1 : 30-33.
Rhoton, F.E.; Meyer, L.D. & McChesney, D.S. (1991): Depth-of-crogion
assessment using iron-manganese nodule concentrations in surface
horirons. Soil Sci., 152, 5 : 389-394.
Rishirumuhirwa, T. (1992): Ruissellernent et erosion sous bananier au
Burundi. Reseau Erosion Bulletin 12: 83-93.
Romkens, M.J.M.; Roth, C.R. & Nelson, D.W. (1977): El-odibility of
selected clay subsoils in relation to physical and chemical properties.
Soil Sci. Soc. Am. J. 41 : 954-960.
Roose, E.J. (1975): Erosion et ruissellement en Afrique de I'Ouest. Vingt
annke de mesures en pitites parcelles experirnentales. Orstom, B.P.
20, Abidjan, Ivory Coast: 72 pp.
Roose, E.J. (1977): Use of the Universal Soil Loss Eq~~at i on to predict
erosion in West- Africa. In: Soil erosion, prediction and control. Soil
Conserv. Soc. Am., Spec. Pub. 21, Ankeny, Iowa: 60-74.
Roose, E.J. & Bertrand, R. (1971): Contribution i I'ktude de la methode
des bandes d'al-ret pour lutter contre I'erosion hydrique eri Afrique de
I'Ouest. Agron. Trop. 26, 1 1 : 1270- 1283.
Roose, E. & Jadin, P. (1969): Erosion, ruisellement et drainage oblique sur
un sol :I cacao de nioyenne CGte d'Ivoire. ORSTOMIIFCC., Abidjan,
Ivory Coast: 77 pp.
Roose, E. & Piot, J. (1984): Runoff, erosion and soil fertility restoration on
the Mossi Plateau (central Upper Volta). Challenges in African
Hydrology and Water Resources, Proc. of the Harare Symposium,
July 1984, IAHS Publ. 144: 485-498.
Roose, E. & Sarrailh, J.M. (1989): Erodibilite de quelques sols tropicaux.
Vingt annkes de mesure en parcelles d'erosion sous pluies naturelles.
Cah. Orstom, Sel-. Pkdol. 25, 1-2: 7-30.
Roth, C.B.; Darrell, W.N. & Riimkens, M.J.M. (1974): Prediction of
subsoil erodibility using chemical ~nineralogical and physical
parameters. Report for Office of Research and Development. U.S.
Environ. Protect. Agency, Project No. 15030 HIX, Element I BB042,
Manuscript: I 1 1 pp.
Ryumugabe, J.B. & Berding, F.R. (1992): Variabititk de l'indice d'agressivitk
des pluies au Rwanda. Rescau Erosion Bulletin 12: 1 13- 1 19.
Sabel-Koschella, U. (1988): Field studies on soil erosion in the southern
Guinea Savanna of Western Nigeria. PhD thesis, Lehrstuhl f ur
Bodenkunde, Technische Universitat Miinchen, Germany: 180 pp.
Salako, F.K. (1988): Determination of potential sheet and rill erosion losses
in southeastern Nigeria as a guide for soil conservation planning.
Beitrage trop. Landwirtsch. Veterinarmed. 26, 2: 1 17-125.
Salako, F.K.; Obi, M.E. & Lal, R. (1991): Comparative assessment of
several rainSall erosivity indices in southern Nigeria. Soil Technology
4: 93-97.
Schauder, H. & Auerswald, K. (1992): Long-term trapping efficiency of a
vegetated filtcr strip under agricultural use. Z. Ptlanzenerniil~r.
Bodenk. 155: 489-492.
Schauer, T.H. (1975): Die Blaikcnbildung in den Alpen. Schriftenreihe des
Bayer. Landesamtes fiir Wasserwirtschaft, Miinchen, Germany, 8. 29.
Schliephake, K. (1984): Tunesicn. Thiemernann, Stuttgart, Germany: 600 pp.
Schuster, R.L. (1978): Landslides, analysis and control. Nat. Acad. Sci..
Washington, D.C., Special Report 176.
Schwertmann, U.; Vogl, W. & Kainz, M. (1987): Bodenerosion di~rch
Wasser. Eugen Ulmer Vel-lag, Stuttgart, Germany: 62 pp.
Segalen, Y. (1967): I.,es sols et la geornorphologie du Carneroun. Cuh.
Orstom. Ser. Pedol. 5, 2: 137- 187.
Servico MeteorolGgico (1972): Cartas das isoietas midias mensais e annais
das ilhas de S. Tom6 c Principe. S. Tomt5: 59 pp.
Shainberg, I. (1985): The effcct of exchangeable sodiuin and electrolyte
concentration on crust formation. Advances in Soil Science 1 : 10 1 - 122.
Shainberg, I.; Gal, M.; Ferreira, A.G. & Goldstein, D. (1991): El'fkct of
water quality and amcndrnents on thc hydraulic properties and erosion
fro111 several niecliterranean soils. Soil Technology 4: 135- 146.
Sheng, T.C. (1989): Soil conservation for small f'arrners in the humid tropics.
FA0 Soils Bull. 60: 104 pp.
Sheng, T.C. (1990): Runol'l' plots and erosion phenor-ncna on trol3ical
steeplands. Research needs and applications to reduce erosion and
sediinentation in tropical steeplands, Proc. Fiji Synlp.. June 19cNl.
IAHS-AISH Publ. 192: 154- 10 1.
Sick, W.D. (1979): Madagaskar. Tropischcs Entwicklungsland zwischcn den
Kontinenten. Wissenschaf liclle Ruchgcsellschaft Darmstadt.
Germany: 32 1 pp.
Sinlonart, T.; Duchaufour, H.; Bizimana, M. & Mikokoro, C. (1993): La
co~~servation dcs sols en milieu paysan burundais. Etude et
h i e r a r c l i t i o ~ des stratt5gies anti-erosives. Reseau Erosion, Bulletin
No. 13: 72-83.
Singer, M.J.; Rlackard, J. & Janitzky, P. (1980): Dithionite iron and soil
cation content as Sactors in soil erodibility. In: Dc Boodt, M. 81
Gabl-iels, D. (eds.): Assessment of erosion. John Wiley. New York:
259-267.
Smith, D.D. (1941): Interpretation of soil conservation data lhr field use.
Agr. Eng., 22: 173- 175.
Smithen, A.A. & Schulze, R.E. (1982): The \patla1 di\tl-ibution In southern
Africa ot rainh~ll ei-o\~\ ~ t y Ihr u\e In tlie Un~ver\al Soil Loss Equat~on.
Watcr SA 8, 2: 74-78.
Stocking, M.A. (1981): Caujes and prcdictlon of tlie advance of gul l ~e\ .
Proc. South-Ea\t Asian Rcg~onal Sylnp. on Problem\ of \ o ~l ero\lon
and \eci~mcntation, Banpkoh 27-29 Jan.: 37-47.
Stocking, M.A. (1988): Asscss111g vegetative coccr and management cfl'cct\.
In: Lal. R. (ed. ): Soil erovon rcscarch method\: 163- 185.
Stocking, M.A. & Elwell, H.A. (1976): Ka~ntall er o\ i \ ~t y o\ er Rhoclt.\~a.
Tranj. In\[. Br. Geogr. 1 , 2: 23 1-245.
Suchel, J.R. (1972): L a I-bpart~t~on dcs plules et le\ rkg~mc\ pluviorndt~-~qi~e\
ail Camcroi~n. Tracaux et Document\ de C;bogr,tphie Troptcale, n o . 5
CEGbT, T~tlence. France: 287 pp.
Sulaiman, W.; Maene, I,.M. & Mokhtaruddin, A.M. (1981): Runoff, \ o ~ l
and nutl-ier~t lo\\c\ Ir-oln an liltisol under dif1i.1-ent legu~nes. SouthEa\t
A\ian Regional Symp. on Pro171~1i1s ot Soil Ero\i011 and
Secl~mentation, Bangkok, Tlia~land, 27-2C) Jan. 198 1 : 275- 286.
Sulaiman, W.H.W.; Mok, C.K.; Maesschalck, G. & Jamal, 'I'. (1983):
Atlc;lncc\ 11) \ n ~l ~uid w,lter conwrvatlon ic\e:~rch In Malay\la.
Pel-tan~ha 6 ( Kc\ . si~ppl. ) : 1 15- 132
Sutherlancl, K.A. & Bryan, R.B. (1990): Runotf and el-o\lon from a \mall
semlal id catchment. Baringo I>i\trtct, Kenya. Appl~ed Geography 10.
91 - 109.
Suaziland (1990): Annual \tatist~cal bullctln. Central Stati\tical Off~co. P.O.
Box 336. Mbabane. Swa71land. 148 pp.
'l'ato, K. & Hurni, H. (1992) (eds.): So11 conser\:~tlon Ihr \urv~val. So11 ~ul cl
Water Con\. Soc.. Ankeny. Iowa: 420 pp.
l'enlple, P.H. (1972): Me:t\uremcnt\ ot runol'f and soil eroslon at an erosloll
plot xa l e with particular I-cferencc to Tanranla. Geografi\ha Alinaler
54A: 203-220.
Temple, P.H. & Rapp, A. (1972): Land\lides In tho Mgeta area. We\tern
Ulugun~ Mountains, Tanzania. C;cografi\ha Annaler, 54A.13-4: 157- 193.
'I'hornbury, W.D. (1985): Pnnc~pl es of geomol-phology. John W~l ey &
Sons. Ncw York: 594 pp.
Torrent, J. & Rarrhn, V. (1993): Laboratory riiea\urc~nent ol' \oil colour:
Theory and practice. In: Ripham, J.M. & Ciolho\z. E.I. (eds 1. So11
colou~.. Soil SCI. Soc. Anl. Spec. Pub. 3 1 : 2 1-33.
Ulsaker, I,.(;. & Kilewe, A.M. (1984): Kilnol'f and so11 erosion lor an Alii\ol
Literature
in Kenya. East African Agric. Forestry J. (Special Issue) 44: 2 10-23 1 .
USDA (1972): SCS National Engineeing Handbook. US Dept. Agric., Soil
Cons. Serv., Section 4, Hydrology.
USDA (1979): Proceedings of the rainfall simulator workshop, Tucson,
Arizona. ARM-W- 10. Northern Plains Soil and Water Research
Center, Sydney, Montana: 185 pp.
Valentin, C. & Janeau, J.L. (1989): Les risques de degradation structurale
de la surface des sols en savane hurnide (C6te d' Ivoire). Cah. 01-storn,
Ser. Pkdol. 25. 1-2: 4 1-52.
Van Liew, M.W. & Saxton, K.E. (1983): Slope steepness and incorporated
residue effects on rill erosion. Transactions of the ASAE 83: 1738- 1743.
Vanelslande, A.; Rousseau, P.; Lal, R.; Gabriels, D. & Ghuman, R.S.
(1984): Testing the applicability of a soil erodibility nomogram for
some tropical soils. Challenges in African Hydrology and Water
Resources, Proceedings 01' the Harare Symposium, July 1984, IAHS
Publ. 144: 463-473.
Vogel, H. (1988): Deterioration of a mountainous agro-ecosystem in the third
world due to emigration of rural labour. Mountain Research and
Development 8, 4: 32 1-329.
Vogel, H. (1992): An evaluation of five tillage systems for smallholder
agriculture i l l Zimbabwe. Projcct Report no. 5, P.O.Box BW 415,
Borrowdale, Harare, Zimbabwe: 1 - 18.
Von Gnielinski, S. (1986): Ghana. Tropisches Entwicklungsland an der
Oberguineakuste. Wissenschaftliche Buchgesellschafi Dal-mstadt,
Germany: 278 pp.
Von Maydell, H.J. (1983): Arbres et arbustes du Sahel. Schriftenreihe der
Gesellschaft fur Technische Zusamrnenarbeit No. 147: 53 1 pp.
Walling, D.E. (1984): The sedirnent yields of African rivers. Challenges in
African Hydrology and Water Resources, Proceedings of the Harare
Symposium, July 1984, IAHS Publ. 144: 265-283.
Walling, I1.E. (1988): Measuring scdiinent yield from river basins. In: Lal,
R. (ed. ): Soil erosion research r-nethods. Soil Water Cons. Soc. Am..
Ankeny, Iowa: 39-73.
Weber, F.K. & Stoney, C. (1986): Reforestation in arid lands. Volunteers in
Technical Assistance, 18 15 North Lyun Street, Suite 2000, Arlington,
Virginia 22209, USA: 167 pp.
Wenner, C.G. (1977): Soil conservation in Kenya. Land and Management
Division, Ministry of Agriculture, Nairobi, Kenya: 205 pp.
Wenner, C.G. (1989): Soil and water conservation in the farming areas of
Lesotho: A review and some proposals. Topics in Applied Resource
Management 1 : 57-86.
Wiese, B. (1988): Elfenbeinkiiste. Erfolge und Problcme eines Ent-
wicklungslandes in den westafrikanisclien Tropen. Wissenschaf'tliche
Buchgesellschaft, Ilarrnstadt, Germany: 303 pp.
Wilkinson, G.E. (1975): Rainfall characteristics and soil erosion in the
rainforest area of Western Nigeria. Expl. Agric. 1 1 : 247-255.
Wilson, R.C.; Torikai, J.D. & Ellen, S.D. (1992): Development of' 1-ainf'ull
warning thresholds for debris flow in the Honolulu district, Oahu. U.S.
Geol. Survey Openf'ilc Report 92-52 1 : 35pp.
Wischmeier, W.H. (1966): Relati011 of field-plot runoff to management and
physical fllctors. Soil Sci. Soc. Am. Proc. 30: 272-277.
Wischmeier, W.H. (1975): Estimating the soil loss equations cover and
n~anagcliient factor I'or undisturbed areas. U.S. Agr. Res. Ser. (Rep. ),
ARS-S-40: 285 pp.
Wischmeier, W.H. (1976): Use and rnisuse of the Universal Soil Loss
Equation. J. Soil Water Cons. 3 1 , 1 : 5-9.
Wischmeier, W.H. & Mannering, J. V. (1969): Relation of soil properties to
its erodibility. Soil Sci. Soc. Am. Proc. 33: 13 1 - 137.
Wischmeier, W.H. & Smith, D.D. (1958): Rainfall energy and its
relationship to soil loss. Transactions, American Geophysical Union
39, 3: 285-29 1 .
Wischmeier, W. H. & Smith, D.D. (1978): Predicting rainfall erosion losses
- a guide to conservation planning. IJ.S. Depar t l ~~et ~t of Agricultiirc,
Agricilltilre Handbook No. 537: 58 pp.
Wischmeier, W.H.; Johnson, C.B. & Cross, B.V. (1971): A soil erodibility
no~i ~ograph for fart-nland and construction sites. J. Soil Water Cons.
26. 5: 189- 193.
Yoder, D.C.; Porter, J.P.; Laflen, J.M.; Simanton, J.R.; Renard, K.G.;
McCool, D.K. & Foster, G.R. (1992): Cover management factor (C).
I n: Renard. K.G.; Foster. G.K.; Weesies, G.A.; McCool, D.K. &
Yoder. D.C. (coord.): Predicting soil erosion by water. A guide to
conservation planning with the Revised Univel-sal Soil Loss Equation
(RUSLE). Draft version 1992. US Dept. Agriculture.
Young, A. (1989): Agroforestry for soil conservation. CAB International,
Wallingford, UK: 276 pp.
Zingg, A.W. (1940): Dcgrcc and Icngth of land slope as it affets soil loss in
runoff. Agric. Eng., 2 1. 2: 59-64.
Zobisch, M.A. (1986): Erfassung und Bewertung von Bodenerosions-
prozessen auf Weideflichen im Machakos-Distl-ikt von Kenia. Der
Tropenlandwirt, Beiheft Nr. 27: 220 pp.
I nde \
Index
10 year. rtorm 45, 92. 146, 2 1 0
Aggregate 47, 104
Aggregate \i7e 37
Aggregation 36. 38, 4 1
Alficol 66. 108, 147, 160
Algeria 165. 194
Alurniniunn 38
Andi\ol 60
Angola 179. I94
Antecedent rno~\ture 38, 62. 70
Arid 37
Aridisol 66. 137
Arsersment 69
Bambara nut 250
Banana 233
Harcl'allow 28, 35. 75
Hare Icvel 6 1
E3eanr 250
Bed loaci 25
Rcdd~ng 33
Bedload 22. 83
Rcnch terrace 53
Heni n I 95
Brological actrvity 10. 3 1 . 42
Botswana I95
Brick 50
Buffer strip 26 1
Buffcl-\trip 153
Bulk dcnrity 104. 106
Rund 41.48.51. 156
Bunds 260
Burkina Faso 165. 168, I95
Burilncli 105, 17 1 . 195
c' f~lctor I I7
C'abhagc 250
Ca~netwon 100, 122, 165, 168, 172,
195. 210
Canary Island\ 197
Canopy 2 1 7
Canopy cover 124
Canopy height 123
Cassava 238
Cation exchange capacity 47
C;ii~\e
-Illiteracy I4
-Povc1-ty 12
Causes
-physical I I
Central Africa I98
Channel crorion 20
Chili 25 1
Clay 35, 36. 37. 47. 10 1
Climate 62
Compaction 20
Congo I99
Conservation I I
Conservation bench terrace 53
Contour ridging 33
Contour-ridging 145
Contour-ir~g 32. 135, 258
Coral riff 2 1
Coghocton wliccl 75
Cotton 25 1
Cover 40.72, 1 17. 2 17
Cowpea 25 1
Crop stage 1 17
Cropping cycle 37
Dam 20
Damage 45
Damage\ 15
-off-site 15. 20
-on-site 15
Deforestation 1 2, 20
Dcposition 31. 35, 47. 50
I>eprc\siion storage 22
Detachment 22
Detention \torage 22
Digue\ dever\ante\ 50
Digue\ filtrante\ 50
Disperjion 38
Ditch 158
Drainage ditch 5 1
Drop 69
Dunnc tlow 26
Dyhe 50
ti,a\t AI'rica 5 1
Economy 19
Egypt I 00
Encap\illated ail 38
Energy 32, 80
Equatorial Guinca 180. 199
f'ritrcu 199
Er o d ~b ~l ~t y 30, 37. 66. 70, 75, 87, 99
I-,ro\ion 16
-geologic 1 I
-\electrvt. I h
Eso\lon cycle 58
L~- o\ ~on nail\ 77
Er o\ ~\ ity 32. 92. 1 17. 168
b,r-osi\ ~ t y ~ndi ce\ 33
Ethiopia 181. 199
Exchangeable catloti\ 37
Fallow 37, 125. 230
Fal l o*~ 37
Fany~i Juu 5 1 . 158
Fel-tility 20
Fertili~er I9
Filter-strip 4 1
Filter-\trips 47
Flood 20. 50
Flow depth 23
F l o ~ belocity 23, 25
Flume 72
Fodder crop 232
Forest 1 I , 12. 125. 230
Foul-nier index 33
Furrow 45, 46. 146
Gabon 200
Geologic erosion 58
Geology 62
Ghana 182.200
Gradient 29, 43, 2 13
Gravel 43, 10 1
Gr o~~ndnut 120, 343
C;roundwater 20
Growth curve 1 19, 226
Growth \tagc 72
Guinca 20 1
Gully 19, 58
Hail 93
Handtillage 42, 145
Hard wtting 28
Heaping 33. 45
Heap\ 156
Hematite 66
Hlghland 66
tIill\ide ditch 5 1
Hill\idc-ditch 41
Horton tlow 26
Hu d wn index 33
Incepti\ol 06. 108. 147
Indicatol- 67
Infiltl-ation 22, 26, 36. 32. 48
Inwlberg 58
I~lctitution 13
Intensity 3 1 , 32. 33, 89
Intermittent terrace 55
Interrill 22, 39
Interterrace 53
Irish potatoes 352
Iso-el-odent n ~ a p 33, 34
I\o-erodent map5 9 1
Ivory Coast 166, 183, 20 1
Kaolinitc 35
Kenya 166, 169.20 1
Laboratory te\ts 70
Land tenure 14
Landscape 58
Land3lide 62
Leaching 18
Lemon grass 252
Lesotho 20 1
Level bench terrace 53
Liberia 184, 20 1
1,owland 66
LS factor 40, 1 1 1
Lybia 201
Madagascar 67, 166. 1 85, 202
Madeira 202
Magnesium 37
Maize 120, 245
Malawi 186, 203
Mali 203
Management 37. 40
Manning 23
Marocco 173, 203
Mauritania 203
Mauritius 204
Migration 12
Mincral 35
Minerals 43
Mocarnbique 204
Mollisol 66
Mounds 260
Mo~arnbique 187
Mudtlow 62
Mulch 42, 43, 1 17, 124, 2 17
Mu\grave equation 86
Namibia 204
Niger 166, 169. 204
Nigeria 166, 169, 188, 204
Not~llage 255
Orchard terrace 55
Organic carbon 17
Organic rnatter 16, 37, 47, 66, 10 1
Overgrazing 12, 63
Overland flow 23
Overpopulation 13
Oxide 35, 108
Oxides 70
Oxisol 35, 66, 147
P factor 145
Papaya 252
Parent material 15. 64, 108
Pediplain 58
Peneplain 58
Permeability 102
Pertneability class 102
Pineapple 235
Plastic foil 43
Ponding 22
Poverty 12
Pressures 3 1
Principe 190
Quintuples 46
R factor 32, 89
Rain drop 3 1
Rainfall 3 1
-Energy 3 1
Rain Fall sirllulator 69, 72
Refugees 13
Relative erosivity 1 19
Remobilization 84
Residual effect 125
Residue 42
Residues 23 1
Retention storage 22
Revised Universal Soil Loss
Equation 39, 146
Ridge 259
Rill 19, 22, 29, 39, 45
Riparian filter strip 154
River 20, 67, 84
Root 37, 68, 8 1 , 160
Runoff 15, 28, 42, 45.48. 70, 1 14
Runoff coefficient 28
Runoff plots 75
Runoff velocity 40. 48. 114. 154
RUSLE 40
Rwanda 166, 169, 189
Sabre growth 63
Sahel 169
Saline 38
Sand 36. 47
Sao Tom6 190, 205
Schist 62
Sealing 22, 26, 28, 29, 36
Seal 37
Season 3 1 . 38
Sediment delivery ratio 84, 159
Sedin~ent cnrichrnent ratio 17
Sediment traps 70
Seclilnont yield 84
Sei\rn~c activity 62
Selective r-enlovul 43
Serni-arid 37
Senegal 166. 205
Sheet erosion 22
Sheet flow 22
Sick \lope 43, 53
Side-slope 159
Sierra Leone 19 1 , 2C)5
Sighting frame 2 19
Silt 36
SI,EMSA XX
Slope 38. 62
-concave 1 14
-convex 1 14
-uniform 1 14
Slope length exponent 40
Slope length 29. 39, l l I . 158, 2 13
S~nectite 36
Socio-economy I I
Sodi u~~i 37. 42. 6 I
Soil
-bulk density 15
-depth 15
-fertility 18
-formation I 1 , 16 1
-function 15
-functions 160
-organic matter 37
-productivity 15
-properties 36
-texture 15
Soil classification 66
Soil colour 66
Soil loss 37, 87. 1 18
Soil loss ratio 1 18
Soil solution 38
Soil weathering 16 1
Somalia 206
Soudan 192. 206
South Africa 174. 206
Soya 252
Splash 22. 72
Splash erosion 22. 26. 37
Stl-eatn bank ct-osion 20
Streambank erosion 84
S tr11ct 11re 1 5
Structure class 10 1
Subk~ctor 123
Subsoil 10 1
Sub~urface flow 6 1
Surface roughness 22
Surface storagc 42
Suspended loaci 2. 25
Swa~iland 207
Sweet potatoes 253
Tanzania 169, 207
Tchad 165
Temperature 38, 42, 66. 70
Terminal velocity 3 1
Terrace 41, 5 1 . 53. 158
Texture 36
Tied r~dging 43
T~ed-ridge 259
Tied-ridging 145
Tillage 30, 4 1
Tobacco 253
Tolerance 18. 160
Topography 38
Tran\port capacity 20, 23, 154
Ti~nl\la 193, 207
Tul-bulence 25
Uganda 169, 208
Ulti\ol 66, 108, 147, 160
Ulti\ol\ 28
Un i t plot 75, 86
Llniver\al Soil Lo\\ Equation 86
Vegetation 28. 35, 37. 62
Verti\ol 36, 66. 147
Vertiwls 28
Visco\~ty 38
Volcanic a\h \oil\ 106
Water capacity 37, 160
Water layer 3 1
Water mulch 38-42
Water retention 16 1
Waterlevel 2 11
Waterlogging 48
Watershed 20, 26, 28, 62, 67, 84
Waterways 53
Weir 50
Wo\t Al'rica 40. 177
Wind 3?
Yam 253
Y ~el d 160
Zaire 208
Zambia 167. 170, 175. 208, 2 10
Zimbabwe 167, 170, 176
Acknowledgments
The author\ thank the Deut\uhe Gcsellschaft fiir Techni\che
Zusam~nenarbeit GmbH (GTZ) fol- \ponsci-ing the booh. We a150 ~ipprcciated
very much MI-\. Schuhba~~cr' s tedious worh on the irlcluded maps.