|
Published by the Sustainable Agriculture Programme of the
International Institute for Environment and Development
Indigenous Soil and
Water Conservation
in Africa
Chris Reij
IIED
GATEKEEPER SERIES No. 27
INTERNATIONAL
INSTITUTE FOR
ENVIRONMENT AND
DEVELOPMENT
The Gatekeeper Series of the Sustainable Agriculture Programme is produced by the Interna-
tional Institute for Environment and Development to highlight key topics in the field ofsustainable
agriculture. The Series is aimed at policy makers, researchers, planners and extension workers
in government and non-government organizations worldwide. Each paper reviews a selected
issue of contemporary importance and draws preliminary conclusions of relevance to develop-
ment activities. References are provided to important sources and background material. The
Swedish International Development Authority and the Ford Foundation fund the series.
Chris Reij is at the Centre for Development Cooperation Services, Vrije University, Van der
Boechorstraat 7, 1081 BTAmsterdam, The Netherlands.
1991
GATEKEEPER SERIES NO. SA27 1
2 GATEKEEPER SERIES NO. SA27
INDIGENOUS SOIL AND WATER
CONSERVATION IN AFRICA
Introduction
Many parts of Africa are experiencing annual population growth rates between 2 and 4 %,
degradation of the natural resource base, recurrent droughts and a growing dependence on food
aid as well as the import of cereals to cover food deficits. During the last two decades increasing
financial outlays for agricultural research in Africa have neither produced significant break-
throughs nor led to agricultural growth (Lipton, 1988).
Since the 1920's, numerous reports have warned against the disastrous effects of increasing
erosion, land degradation, desertification, mismanagement of natural resources due to increasing
demographic pressure, and as a result soil conservation emerged at the end of the 1930's as a
central concern in East Africa (Anderson, 1984). In many African countries considerable efforts
have been made during and since colonial times to conserve soil and water resources. Yet most
soil and water conservation projects in sub-Saharan Africa have failed. A major argument is that
what has been constructed often at great expense has seldom been maintained by the
"beneficiaries". Where adequate maintenance is lacking, conservation works quickly dilapidate
and accelerate erosion instead of reducing it.
The most important reasons for these failures in African soil and water conservation (SWC)
include a dominant top-down approach, the use of techniques which are complicated to design and
expensive, both in terms of labour and capital, to maintain and therefore are not replicable by
farmers, a neglect of farmer training, a heavy reliance on machinery for the construction of
conservation works and an indiscriminate use of food-for-work. Rather than importing soil and
water conservation strategies, we must rely more on indigenous SWC techniques:
although many indigenous farming systems in the region are now under severe pressure or
are disintegrating into environmental abuse, they are usually grounded in the necessary
detailed knowledge of the local environment. Thus they form the most viable basis for the
development of a workable soil and water conservation strategy for the future" (Reij et al,
1986a).
Partly as a reaction to the disappointing results of integrated rural development programmes
(IRDP's) with their strong emphasis on "transfer of technology", the 1980's have seen a growing
awareness of the importance of indigenous environmental knowledge (Richards, 1985), farmer
first models (Chambers, 1983; Chambers et al, 1989), fannrmer participatory research (Farrington
and Martin, 1988) and Rapid Rural Appraisal (McCracken et al, 1988). These all draw attention
to the environmental knowledge, capacities, skills and priorities of rural people. As part of this
trend, the awareness of the importance of indigenous SWC techniques has also increased. But an
increase in the awareness of its importance does not necessarily imply a corresponding increase
in our knowledge of indigenous SWC.
GATEKEEPER SERIES NO. SA27
The objective of this paper is to assess our current knowledge of indigenous SWC in Africa and
to identify research needs and policy requirements in the field of African indigenous SWC. Three
major issues are explored in the main text.
The first demonstrates that despite a growing awareness of its importance, African indigenous
SWC continues to be neglected. The second analyses present trends in indigenous SWC. Are
indigenous techniques increasingly abandoned and if so, why? Can we identify cases where
indigenous SWC techniques continue to be maintained and even expanded? The third briefly
examines the effect of project interventions. Some examples will be given of project interventions
damaging indigenous SWC and of others improving the efficiency of indigenous SWC tech-
niques. The annex contains details and diagrams of indigenous SWC from Cameroon, Morocco,
Somalia, Sudan and Tunisia.
The Neglect of Indigenous SWC
The 57 references listed at the end of this paper would seem to indicate a wealth of knowledge
of and a lively interest in indigenous SWC systems. However, our current knowledge of African
indigenous SWC is limited and often outdated. Indigenous SWC techniques continue to be
neglected by African and expatriate researchers, and by SWC specialists as well as by government
staff. The following examples clearly illustrate this point.
1. Some detailed descriptions and analyses of indigenous SWC systems can be found in various
documents. Ludwig (1968) studied indigenous SWC techniques on Ukara island (Tanzania), but
no other study seems to have been made since then. French researchers studied indigenous SWC
in the Mandara mountains of North Cameroon (Hallaire, 1971; Boulet, 1975; Boutrais, 1973), but
this research has essentially been carried out in 1965 or 1966 and detailed follow-up studies are
lacking. Netting's study of the Kofyar on the Jos Plateau (Nigeria) done in the first half of the
1960's is the only study I know which has been followed up with a detailed study in the first half
of the 1980's (Netting, 1968; Netting et al, 1989).
2. Several indigenous SWC techniques have not been studied at all. For example, no research has
been done on traditional stone lines in the Ader Doutchi Maggia in the Tahoua Department of
Niger. No publications can be found about pitting systems in the Djenn& Sofara region in Mali.
Indigenous water harvesting systems in the Central rangelands of Somalia have only recently
been "discovered" (Critchley et al, in press). Traditional terracing in the Rif mountains in North
Morocco has not been the object of serious research. There are dozens of similar examples.
3. Traditional stone lines in the Ader Doutchi Maggia, Niger, can be observed by anyone driving
on the main road from Konni to Tahoua. Important SWC projects have been implemented in this
region since the beginning of the 1960's. Scores of SWC experts have visited the area, but their
reports contain no remarks about traditional stone lines. A French researcher, who did a most
interesting six year study in the Maggia valley on the impact of different SWC techniques on
runoff, soil loss and yields remarked that one of the reasons for selecting the village of Allokoto
for the implantation of a research station was the fact that the villagers already had some notion
GATEKEEPER SERIES NO. SA27
of land rehabilitation as they had constructed contour stone bunds themselves .."de facon assez
sommaire bien sfir" (Delwaulle, 1973). Although Delwaulle characterized the traditional stone
bunds as rudimentary, he recognized that they could be a starting point for future land rehabili-
tation. Yet they were not studied.
4. Geographers and anthropologists have made numerous studies of the Dogon in Mali and these
indicate that the Dogon have developed intricate systems to conserve rainfall. They even
transport soils to bare rock in order to create new fields to cultivate rice or onions (Gallais and
Sidikou, 1978; Krings, 1988). Despite a recent reconnaissance survey and brief analysis of various
techniques (Kassogue and Ponsioen, 1990), a detailed analysis of the range of techniques
employed by the Dogon is still lacking.
5. Staff of the Planning Department of the Ministry of Agriculture of the Cape Verde recently
produced a challenging document for a major CILSS/Club du Sahel conference on decentralized
management of natural resources in the Sahel (at Segou, Mali, in May 1989). According to this
document, "the topography, the rainfall characteristics and population pressure, combined with
the fact that the population shows a lack of awareness of the problem, has created an erosion
process on the islands" (MDRP/GEP, 1989). No mention whatsoever is made in this document
of extensive bench terracing on the island of Santo Antao.
6. Heusch (1985) estimates that in Tunisia, Algeria and Morocco traditional irrigation terraces
cover about 2 million hectares. But at the same time he acknowledges that cost-benefit data for
traditional irrigation terraces are not available.
An Analysis of Indigenous SWC
Ethno-engineering
African cultivators apply a wide range of techniques such as crop rotation, crop mixtures, the
application of manure, the protection of nitrogen-fixing trees, terrace building, pitting systems,
drainage ditches, small dams in valley floors, and so on to conserve soil and water and to prevent
soil degradation. I do not in this paper deal with relevant agronomic and agroforestry practices,
although these are often used in combination with mechanical SWC. Today agronomic and
agroforestry practices are receiving increasing attention. For example, the International Council
for Research in Agroforestry (ICRAF) has worked on an Agroforestry Systems Inventory (Nair,
1989) and Young (1989) has summarised the state of knowledge in the field of agroforestry and
soil conservation. Yet inadequate attention has been given in the last 25-30 years to African
indigenous conservation works, and therefore I will concentrate in this paper in particular on what
Jodha (1990) has coined ethno-engineering, which comprises "such practices as terracing
mountain slopes, harnessing the runoff and developing small drainage systems".
Ethno-engineering practices are applied by African cultivators under widely varying climatic
conditions. Generally speaking, water harvesting techniques are used in regions with 100 mm to
GATEKEEPER SERIES NO. SA27
about 700 mm rainfall. In the context of this paper, water harvesting can be defined as the
collection and concentration of runoff for increased and more reliable plant production (Reij, et
al, 1986). The tabias and jessours in Tunisia (see annex) are a good example of floodwater
harvesting within a streambed in a region with 100-200mm rainfall. The meskat and mankaa also
in Tunisia and the teras system in East Sudan (see annex) are examples of water harvesting
systems using a long slope in regions with 200-400mm rainfall.
In regions with 500mm rainfall or more, the emphasis shifts from water harvesting techniques to
in-situ moisture conservation. For example, the Dogon in Mali build small earth ridges around a
number of sorghum or millet plants. These 20-30cm ridges have a rectangular or a beehive shape
and in principle they conserve every drop of rain falling within them. The use of in-situ moisture
conservation techniques in areas with about 500mm rainfall causes problems in years with below
average or irregular rainfall when total rainfall may not be enough to guarantee a crop or when
long dry spells during the rainy season cause crop damage. In particular in regions with about 500-
700mm in the West African Sahel, one can observe a largely project inspired shift from in-situ
moisture conservation techniques to water harvesting techniques.
As soon as the annual rainfall approaches a 1000mm or more, combinations of SWC (terracing
of steep slopes, stone bunds on low slopes) and drainage systems in farm fields, where
waterlogging is a risk, are more common. Examples of this can be found in the Mandara
mountains in North Cameroon (see annex).
An interesting but slightly different description of the diversity of traditional and modem
strategies for SWC in the Sudano-Sahelian region of West Africa has identified four subareas
where traditional strategies differ (Roose, 1990). The South Sudanian area with more than
1000mm rainfall is characterized by what is called drainage farming, such as Senoufo farmers in
North Ivory Coast and Southwest Burkina Faso cultivating on broad ridges. The North Sudanian
zone with 700 mm 1000 mm rainfall is characterized by rainfed farming, such as the Minianka
farmers of South Mali who build protection earth bunds at the bottom of the hills, but try to capture
all the rainfall, in the field by using tied ridges. In the South Sahelian area with 400-700mm
rainfall farmers practise runoff farming, such as the Mossi in the Yatenga region of Burkina Faso,
who use stone bunds, planting pits, mulching, etc. Lastly, there are farmers in the North Sahelian
area (rainfall less than 400 mm) who practise valley farming, which involves intensive cultivation
of valley floors, where runoff from the slopes is concentrated.
More than 50 examples of permanent farming systems with soil conserving practices in the
African tropics drawn from the literature are presented in Table 1. It is the only systematic
inventory of farming systems with soil conserving practices. In 29 out of 52 examples terracing
and irrigation farming are practised. For the compilation of this table, Ludwig has used books and
reports published between 1935 and 1961. The major value of this table is that it clearly shows
that African indigenous SWC techniques are not an exception; they are applied over large parts
of the continent. A further 18 examples of indigenous SWC systems using ethno-engineering
practices are shown in Table 2. Again Tables 1 and 2 together are not exhaustive. For example,
in Mali terracing systems can be found in the Monts Mandingues around Bamako and in the falaise
de Tambaoura (no information available) and in Ethiopia bench terracing for chat (Catha edulis)
is common (Thomas, 1988). In Zimbabwe, so-called Invanga terraces are widespread in regions
inhabited by the maShona (Nyamapfene, 1987).
6 GATEKEEPER SERIES NO. SA27
Table 1: Permanent Farming Systems with Soil Conserving Practices in the Africa
Tropics
Ethnic Group
Malinke
Baule
Kita
Dogon
Bobo
Gurensi
Nunuma
Mamprusi
Losso
Kabre
Mandara
Kamuku, Kanuri,
Chamba
Bauchi, Berron
Sokoto, Kano
Batta, Mundang
Mandji, Bamum
Dama, Musgu
Bana, Adamawa
Kuru, Bari
Konso
Tigre
Kipsigi, Kikuyu
Nandi, Suk, Keyu
Rundi
Ruanda
Kiga
Matengo, Makonde
Kinga
Sandawe, Iraque,
Fipa, Turu, Gogo
Mbugu, Shambala,
Pare, Meru, Teita
Wakara
Country
Senegal
Guinea
Ivory Coast
Mali
Dahomey
Dahomey
Dahomey
Dahomey
Ghana
Togoland
Togoland
Nigeria
Nigeria
Nigeria
Nigeria
Cameroons
Cameroons
Sudan
Ethiopia
Ethiopia
Kenya
Burundi
Ruanda
Uganda
Tanzania
Tanzania
Tanzania
Tanzania
Tanzania
Average
rainfall
(mm)
1,000
1270
1,000
800
800
800
800
800
800-2,000
1,500
800
800
800
800-1,000
800-1,000
800-1,500
1,000-1,400
1,000-1,200
600-1,000
1,400-1,800
1,000-1,400
1,000-1,400
800-1,500
1,000-1,200
1,000-1,400
800-1,500
Population SWC
density measures
per sq. km
a b c d
10
10-20
10
10-50
10-50
10-50
30-50
50-100
230
50-70
50-230
50-700
125-200
50-230
100-150
30-50
200
100-150
50-150
100-150
150-200
50-100
30-100
20-100
10-100
1,500-2,000 50-100
1,600 200
X
x
x
x
x
x x
X
X
x x
x
X X X
x x
x
X X X
x
X X X
X x
Main crops
millet, rice, maize
yams, banana, taro,
millet
millet, yams, banana
millet, yams, banana
millet, yams, banana
millet, yams, banana
millet, yams, banana
millet, groundnut, yams
millet, yams, rice
millet, beans
millet, yams, banana
millet
millet, groundnut,
cassava
millet, yams, banana
millet, yams, earth pea
millet
millet, cotton, maize
millet
X X millet, maize, cassava
X millet, banana, yams
X X banana, millet, yams
X millet, banana, beans
X maize, millet, casava
maize, millet
X millet, maize, beans
X x
XXX X
millet, maize, beans
millet, cassava, rice
a = Terracing b = Irrigation farming c = Manuring d = Stabling
Abandonment of SWC Practices
The overall impression obtained from available reports is that African indigenous SWC
techniques are increasingly being abandoned. Traditional terraces in the Djebel Marra mountains
in West Sudan are gradually abandoned (Miehe, 1986); traditional terraces in the Mandara
mountains are in some places not adequately maintained or completely abandoned, which leads
to their collapse (Riddell and Campbell, 1986); traditional earth dams (tabias) in South Tunisia
are in some cases no longer adequately maintained or are abandoned, which disrupts the terraced
GATEKEEPER SERIES NO. SA27
Table 2: Additional examples of indigenous SWC techniques in Africa with prominent 'ethr
Ethnic Rainfall Population
group (mm) density
persons/km2
Mossi
400-700
Indigenous SWC
Techniques
stone lines stone teraaces
planting pits (zay)
Burkina Faso (South)
Burkina Faso
(Southwest)
Burkina Faso
(Southwest)
Cameroon (North)
Cape Verde
(S. Antao Island)
Mali (Djenne-Safara)
Morocco (Anti-Atlas
mountains)
Niger (Ader Doutchi
Maggia)
Sierra Leone
Somalia
(Hiraan region)
Sudan (East)
Kassena 700-800
Birifor
80-100
1000-1100 35
Dagari 1000
20 ethnic
groups
35-80
800-1100 80-250
400-1200
(in uplands)
Bambara/
Sourhai
?
Haussa
>100
20-30
100-250 ?
300-500
Somali 150-300 ?
Hedwenda, 225-400
Shukriya,etc.
stone lines
stone bunds on slopes; network of
earth bands and drainage channels
in lowlands
contour stone bunds on slopes,
cdrainage channels
bench terraces (0, 5-3m. high) stone
bunds
bench terraces (rainfed, irrigated),
contour stone walls (murets),
floodwater control dams, river bank
protection walls (bardos)
pitting systems
terraces, stone banks and small
stone walls
stone lines, planting pits (tassa)
sticks and stone bunding on fields
and in gullies, drainage techniques
earth bunds with upslope wingwalls
(caug) and earth bunds dividing plots
of land into a grid (gawan)
earth bunds (straight) with upslope
wing-walls (teras), and water
spreading techniques
Sudan (Djebel Marra)
Tanzania (Uluguru
mountains)
Tchad (Ouddai)
Tunisia (Medenine)
Tunisia (Sousse)
Fur 600-1000 20-37 (1976)
Luguru 1500 100
250-650 5-6
100-200 15-25
200-300 ?
bench terraces
ladder terraces
various earth-bunding systems with
upslope wingwalls, in drier regions
with catchment area (water
harvesting)
earth dams (tabias/jessours) within
streambeds
earth bunds (meskat system)
8 GATEKEEPER SERIES NO. SA27
Country
Burkina Faso
(Central)
-engineering' practices
Major crops
sorghum, millet
sorghum, millet
sorghum
millet
maize
sorghum
millet
sorghum
peanuts
sugar cane
maize
sweet potatoes
pigeon peas
sorghum, millet,
barley
sorghum
millet
sorghum
cowpeas
sorghum
millet/sorghum
fruit trees,lentils
beans
olive trees
Publications
Savonnet (1958)
Reij (1983)
Savonnet (1976)
Pradeau (1975)
Hallaire (1972)
Boulet (1975)
Boutrais (1987)
Riddell and Campbell (1986)
Haagsma (1990)
Kloosterboer and Eppink
(1989)
Ayers (1989)
Kutsch (1982)
Reij, Martin (1986, 1990)
Millington (1984)
Critchley, et al (1990)
Critchley, et al (1990)
Randall (1963)
Ibrahim (1988)
Miehe (1986)
Temple (1972)
Sommerhalter (1986)
Bonvallot (1986)
Cel Amami (1983)
wadi system (Bonvallot, 1986). The reasons
for abandonment encompass political stabil-
ity, population density, efficiency and market
forces.
Many mountainous regions in Africa, such as
the Djebel Marra in Sudan, the Mandara
mountains in Cameroon, the Plateau Dogon in
Mali, the Jos Plateau in Nigeria and the Kabye
region in Togo, have had a refuge function in
the pre-colonial era. Population groups seek-
ing refuge in mountainous regions often had to
survive on limited land resources and there-
fore had to introduce intensive land manage-
ment practices. In these cases steep slopes
have been terraced. Increased political stabil-
ity and pacification during and after the colo-
nial period made it no longer dangerous for
people to move downhill into the plains.
In some cases governments, during the colo-
nial period as well as after independence,
made deliberate efforts to draw people out of
mountain regions with a high population den-
sity in order to settle them in areas with a low
population density and a higher potential for
agricultural production. Additional arguments
were that they could be better controlled by the
administration and provided with essential
services. The German colonial government in
Togo, for example, forced the Kabye to work
in South Togo. The French colonial govern-
ment continued this policy and forced the
Kabye to settle in Central Togo (Sauvaget,
1981). The efforts of the German and French
colonial governments in Cameroon to move
the people out of the mountains were not very
successful (Campbell and Riddell, 1984).
However, spontaneous migration to cities and
to more productive areas both within and beyond
the Mandara mountains is now an important
demographic variable.
Planned and spontaneous migration from the
mountain regions to the lowlands may have a
negative environmental impact on both the
GATEKEEPER SERIES NO. SA27
mountain regions and on the area of settlement. As the population density descends below a
minimum threshold, maintenance requirements of terraces cannot be assured and fertility
management practices cannot be carried out adequately (Boulet, 1975). As a result, terraces risk
collapse and yields are diminishing.
An exodus from the mountains may also have a negative impact on the areas of settlement. The
Kabye instantly abandon their intensive land use management practices when they settle in areas
of low population density and they destroy the natural vegetation (Sauvaget, 1981). The same
process occurs in North Cameroon (Boulet, 1975; Campbell and Riddell, 1984), although the
Ouldeme continue to construct terraces on the lower slopes of the mountains as long as stones are
abundant (Hallaire, 1971). The descent into the plains may also cause social conflicts between
cultivators and herders (Riddell and Campbell, 1986). A low density population of the plains does
not mean that these areas are empty and available to cultivators (Boutrais, 1987).
A low efficiency of indigenous SWC techniques can be another reason for abandonment. In the
Ader Doutchi Maggia, Niger, as well as on the Plateau Dogon, Mali, fields treated with stone lines
have been abandoned. The most likely reason for abandonment in these cases is that the
techniques were not efficient enough to be able to cope with a combination of recurrent droughts
and loss of soil fertility.
Market forces also play a role in the abandonment of indigenous techniques. The opportunities
for earning a cash income often are much higher in adjacent towns than in isolated mountain areas
with a high population density. In South Tunisia, credit is available for mechanisation, which
makes it attractive to move from the mountains to the lowlands (Bonvallot, 1986). In the Matmata
area of Tunisia, because of rising costs of manpower and employment opportunities in Libya,
most jessours are no longer maintained (Heusch, 1985).
Although many indigenous SWC techniques are able to support reasonable yield levels and high
population densities, (up to 250-300 persons/km2), labour investment in maintenance of terraces
as well as in fertility management is high and returns to labour are low. In regions of high
population density with growth rates in the order of 2%, it is impossible to accommodate significant
population growth when the average farm size is at or less than 2 ha. In some regions the situation
looks rather grim. Heusch (1985) indicates for the Mahgreb countries that within a generation
entire villages will be deserted. Adult male workers have already left. In many cases terrace walls
are no longer maintained and women do not have the resources to put back the heavy stones.
Maintenance and Expansion
The preceding section may have created the impression that African indigenous SWC is at peril
and has no future. However, there are many locations where indigenous SWC techniques continue
to be maintained and even expanded. In some instances, indigenous techniques, abandoned some
decades ago, have been revived suddenly.
GATEKEEPER SERIES NO. SA27
1. Stone lines in the Ader Doutchi Maggia (Niger)
This region is characterized by the presence of major fertile valleys and large barren plateaus.
Rainfall in the area varies from 250 450 mm. Stone lines (Haussa term: gandari) have been laid
out mainly by the people living on the plateaus to conserve water and trap windblown sand. They
are often laid out in straight lines (grid pattern), but in some cases efforts are made to follow a
contour line. It is not possible to estimate how many hectares have been treated with gandari as
they are dispersed over a large area. The size of the fields treated varies from a few hectares to
several hundreds of hectares for example on the Plateau Wandali not far from Koura Abdou in
the Badeguicheri valley.
The gandari seem to be used mainly for the rehabilitation of barren degraded lands with a hard
impenetrable crust (Haussa term: fako). These fakos have sandy-clayey soils. Once laid out,
farmers wait till the harmattan has deposited sufficient sand to permit cultivation. They usually
have to wait 5 or 6 years before the land treated with gandari can be cultivated.
Although numerous fields have been abandoned, elsewhere stone lines continue to be maintained
and fields can be found where stone lines have recently been laid out. All construction is done by
individual farmers on a voluntary basis without any project support. Farmers cover part of the soil
with millet stalks or some manure, which, according to them, helps trap sand.
In 1988 farmers in this region suddenly started making planting pits (tassa) to rehabilitate
degraded land. It is not yet clear where they got the idea from. Some say it is a revival of
indigenous techniques, but according to others, Haussa migrants from Niger who travelled on the
Jos Plateau in Nigeria picked up the idea there. An International Fund for Agricultural
Development (IFAD)-funded SWC project is now assisting these farmers with technical advice
(on increasing dimensions of the pits and using organic matter) and although applied now on a
small-scale, I expect a rapid adoption of this technique by the farmers in this region.
2. Planting pits in the Djenne Sofara region (Mali)121
According to Ayers (1989), 12% of the farmers she interviewed practised pitting to retain more
water. Forty-four per cent of the farmers practising pitting had introduced them since 1985. Some
farmers mentioned that they had observed the technique in other villages in the region, but others
claimed that pitting was an ancient technique which had been superseded by the plough.
Increasing problems of crusting and compaction apparently are a reason for farmers to revive
pitting. There are some indications that pitting of crusted barren land is increasing rapidly
(observations of CARE project manager).
3. Terracing in the Rif mountains (Morocco)
No systematic study has been made of traditional terraces in the Rif mountains of Morocco, but
field observations indicate that farmers continue to maintain and expand their traditional
terraces"'. In some areas expansion of the terraces is related to the cultivation of kif (hasj).
GATEKEEPER SERIES NO. SA27
4.Traditional small scale water harvesting in the Hiraan region (Somalia)
Traditional small scale water harvesting systems for crop production are common over an
extensive part of Hiraan Region in the Central Rangelands of Somalia. It has been estimated that
up to half of the families within the region practice some water harvesting (Caag and gawan
systems: see annex). Farmers continue to maintain the structures and examples have been found
where farmers experiment with the techniques (Critchley et al, 1991).
5. Traditional teras cultivation in the Eastern plains (Sudan)
Teras cultivation is a traditional form of water harvesting used extensively in the clay plains of
Eastern Sudan (see annex). The word teras itself refers to the earthern bund which surrounds three
sides of each cultivated plot, and impounds runoff from the plains. While some traditional areas
of teras cultivation have been replaced by irrigation schemes, others are becoming mechanised
and apparently may even have expanded in area (Critchley et al, 1991).
An Emerging Interest
African indigenous SWC has long been ignored by researchers and developers. It is striking that
interest in indigenous resource management systems is increasing simultaneously in Latin
America 4', Asia 51 and in Africa. It is therefore not a coincidence that the first resolution of the
6th International Soil Conservation Conference (Addis Abeba, Ethiopia and Nairobi, Kenya; No-
vember 6 18, 1989), is about learning from traditional techniques from small-scale farmers. It
states that:
In recognizing the immense wealth of traditional know-how, especially in agricultural
systems, we should learn from the principles behind their success or failure in the handling
of soil and water management.
In particular, it is necessary to integrate the successful and promising principles into modern
research and development efforts in meeting challenges of the present and future land use.
Such integration will require that present day sectorial institutions work together in order to
exploit this traditional potential since it is almost invariably of a multidisciplinary nature.
Slowly but surely, a certain consensus is emerging that indigenous SWC techniques could be used
have a role to play. This trend reflects a feeling of disappointment with or even despair about the
failure to narrow the gap between food needs and food production in Africa and the inability to
create conditions for sustainable rural development. Although I fully support the view that
indigenous SWC has a role to play, some caution is required. First of all we should avoid deifying
indigenous SWC techniques as they are in many instances no longer sufficiently efficient. A
marriage between indigenous and modern techniques may be required to increase the technical
efficiency (coping with degradation) as well as the returns to labour (higher incomes). Secondly,
indigenous techniques have been developed in a particular socio-economic context, but labour
GATEKEEPER SERIES NO. SA27
migration and market forces have profoundly changed that context 16. Thirdly, as indigenous
SWC techniques are location specific, a question to be researched is under what conditions can
they be transferred to other regions? Finally, at present we know too little about the current state
of most indigenous SWC in Africa, and as a result it is difficult to design project interventions
taking indigenous SWC techniques as a starting point.
The Role of Project Interventions
As indigenous SWC techniques have been neglected by researchers, a major risk is that
interventions are proposed which are insufficiently rooted in the priorities and perceptions of local
populations. What do we know about the present state of terracing and other techniques in North
Africa? Despite studies by Kutsch (1982), Bonvallot (1986), El Amami (1983), Despois (1956)
and others, the answer is very little. Another risk is that techniques, which are successful in
specific regions will be transferred to other regions without adequate testing or adaptive research.
Although only few cases can be found of deliberate interventions in African indigenous SWC,
some examples will now be given of project interventions with damaging and positive effects.
Project Interventions Damaging Indigenous SWC
In South Tunisia cultivators made lateral spillways in the earth dams constructed across valley
floors. About 50 years ago, government services started replacing lateral spillways by a central
spillway. One of the major inconveniences perceived by the djebalia mountain people is that the
central spillway is usually built long and low and therefore reduces the capacity to retain water.
The major weakness they perceive is that the central spillway is static. The height of the spillway
can only be raised progressively in proportion to the rate of sedimentation of the jessour (see
annex). As soon as a central spillway has been imposed, the farmer feels less concerned about the
jessour (Bonvallot, 1986).
In the Uluguru mountains of Tanzania, ladder or step terraces are the most widely employed and
generally accepted soil conservation method on the western side of the mountains (Temple,
1972). But in 1950 the Uluguru Land Usage Scheme introduced compulsory bench terracing in
shambas (farms) of medium gradients, with afforestation of steeper slopes, involving their closure
to cultivation of annual crops (Temple, 1972, quoting Duff, 1960). Basically large bench terraces
were superimposed on indigenous ladder terraces. The Uluguru Land Usage scheme was
abandoned in 1955 after serious riots.
Project Interventions Improving Indigenous SWC
On the Central Plateau in Burkina Faso, the Mossi used to employ various techniques to conserve
soil and water (Reij, 1983). One of those techniques was the construction of stone lines. The
efficiency of these stone lines in retaining soil and water was quite limited as contour lines were
not respected, and gaps were left between stones. The Projet Agro-Forestier (PAF) in the Yatenga
region introduced contour stone bunding in the beginning of the 1980s. Contour stone bunding is
GATEKEEPER SERIES NO. SA27
essentially an improvement of existing stone lines. Thousands of farmers have been trained in
using simple survey equipment based on water tube levels to determine contour lines and in basic
rules of construction. The annual rate of implementation increased from 150 ha in 1982-83 to
some 5,000 ha in 1987-88 (Reij, 1988). Contour stone bunds are constructed on cultivated fields,
but they are also used on a considerable scale to rehabilitate degraded soils with a hard
impenetrable crust. Farmers spontaneously introduced (or revived ?) planting pits, zav, in the
beginning of the 1980s. They systematically combine contour stone bunds and zav to rehabilitate
degraded soils. Yield increases are substantial. Farmers estimate that yields on already cultivated
fields increase with about 40%. In years with fairly regular rainfall, yields of 1,000-1,200 kg/ha
on rehabilitated fields are not exceptional.
The ASAL Programme in West Pokot and Northern Marakwet (Kenya) made some effort to
improve indigenous irrigation systems. The programme provided technical assistance and
material support to villagers to solve two problems they faced: (a) when furrows cross permeable
soils or a rocky surface and (b) where recently developed gullies cross furrows or where furrow
sections have been swept away by landslides (Dietz, 1990, quoting van Klinken, 1987).
The Need for Research on Indigenous SWC in Africa
Our current knowledge of indigenous SWC techniques in Africa is extremely limited and many
of these studies are more than 25-30 years old. For this reason we do not know whether techniques
are gradually abandoned or whether they continue to be maintained and expanded. Many reports
only mention the existence of indigenous techniques, but due to lack of research cannot provide
any detailed information. Many indigenous SWC techniques have not been studied at all, leaving
an important gap to be filled 71.
More research on indigenous SWC techniques is essential for two major reasons. First, many
SWC projects have failed and although we have at present very little factual evidence to support
the assumption, we believe that indigenous SWC could be used as a starting point for new
programs. Second, we are faced with a growing exodus from African fragile regions and
mountains to lowlands and urban areas. The consequences are serious for the zones of departure
where intricate terracing systems are no longer adequately maintained, collapse, increase erosion
and aggravate flood hazards in the lowlands. At the same time, the areas of settlement cannot
always cope with the arrival of considerable numbers of settlers from the mountains and in
particular not when they instantly abandon their intensive land use management practices.
To fill gaps in our current knowledge of indigenous SWC techniques and to get a better idea of
constraints and opportunities with regard to the development of indigenous SWC, the first step
could be to make a fairly large number of relatively quick studies. At the same time it would be
useful to initiate, wherever possible, small trials in close consultation with the local resource
users. The objective of such trials could be to improve the technical efficiency of indigenous SWC
techniques. Another objective could be to test techniques which are successful in similar
conditions. A good, but rare, example of this type of adaptive research can be found in Kitui
District, Kenya. Traditional pitting practices of the Kofyar, on the Jos Plateau, Nigeria, and the
GATEKEEPER SERIES NO. SA27
Matengo of southwest Tanzania are used as a source of inspiration for the development of modern
variants of pitting to assist in the rehabilitation and revegetation of eroded grasslands (Jones et al,
1989).
Some research on indigenous SWC systems has recently been or will soon be initiated in Sudan
on teras in the Kassala region, in Niger on traditional stone lines and planting pits (tassa), in Mali
on planting pits in the Djenne Sofara region and in Morocco on traditional terracing in the Rif
mountains. It is hoped that these first research efforts signal the beginning of what will evolve into
a new paradigm in African development research.
Policy Requirements
Even if, due to increased research efforts, our knowledge of indigenous SWC will have increased
substantially within a few years, the practical use of this knowledge may still be hampered by a
number of institutional constraints. Two important institutional constraints are the inappropriate-
ness of present project design and the tendency of governments and donor agencies to go for quick
and tangible results.
If we want to give indigenous SWC and other forms of local environmental knowledge a real
chance, then conventional project design should be thoroughly changed.
Many donor agencies field a number of missions for project identification, preparation and
appraisal. These missions often take 3-4 weeks in the field followed by a similar period for report
writing at headquarters, they involve several consultants and gaps of several months between each
mission are common. This design chain is highly inadequate. Project identification missions tend
to spend half their time in the field talking to public administrators, staff of various ministries
(agriculture, environment) and to some representatives of the target group- often village elite.
The rest of the time is spent in the capital on data collection (census data, price data, etc.) and on
discussing with ministries and donor agencies. This type of identification mission is usually not
in the position to identify and analyse local perceptions,priorities and environmental knowledge.
Assuming some continuity between what has been identified and what has been appraised, it is
important that identification missions get it right. Therefore the emphasis during design should
shift from appraisal to identification. A solution is to field small identification missions (2 or 3
consultants instead of 6 or 7), who know the region well and are prepared to stay 3 months in the
field.
The second constraint is that governments, donor agencies and the press want quick and
tangible results.
The easiest way to achieve this is to use ample food-for-work supplies and/or to rely on the use
of heavy machinery for earth moving or transport of stone. No consideration is given to whether
activities can be sustained in the post-project phase or whether they can be applied by the land
GATEKEEPER SERIES NO. SA27
users without or with a minimum of external support. Indigenous SWC techniques are not well
known and require some research. Experiments should be designed to improve their technical
efficiency, and several techniques should be tested at village level and evaluated by technicians
and villagers. It may take 3 5 years before the best and most acceptable technical package is
identified, hence tangible results can rarely be obtained before 5-10 years have elapsed. It is
essential that donor agencies and governments accept these time frames for projects.
Notes
1.This paper is a revised version of a paper written for a workshop on "Conservation in Africa:
indigenous knowledge and government strategies" held in Harare (December 2 -7, 1990). The
paper was commissioned by the American Social Science Research Council (Project on African
Agriculture). Without the support from different colleagues, this paper could not have been
written in the short period available to the author. Fred Zaal did background research on various
examples of indigenous SWC techniques. Stephen Turner stimulated me at a time when I risked
to "drown" in other work, Alie van de Wal collected documentation and produced with Fred
Zaal a bibliography. Will Critchley's shared interest in indigenous SWC has been a consider-
able incentive. However, only the author is responsible for omissions or errors of interpretation.
2.1 interviewed some young people in the village of Yala in September 1989. They mentioned
without a trace of hesitation that the pitting system had been introduced by their ancestors and
not by extension agents.
3.Observation during short field trip in June 1990.
4.Growing interest in traditional agricultural resource management in Latin America is illustrated
by Wilken (1987) and Browder (1989).
5.In June 1991 a workshop was held at ICRISAT, Hyderabad, about traditional SWC techniques
in India. Earlier, the Asian Soil Conservation Network organised a workshop on "Indigenous
Conservation Farming Practices" in Papua New Guinea (December 1990).
6.A participant in the workshop on Conservation in Africa: indigenous knowledge and govern-
ment strategies remarked: "If you adapt or combine techniques without an understanding of the
social systems in which they emerge you run the risk of undermining social structures which
help to sustain the operation of indigenous technology. We are not just talking about technique;
we are looking at part of a social, cultural and intellectual system, which interacts with other
parts."
7. The Centre for Development Cooperation Services and the Drylands Programme of IIED have
recently formulated a joint research effort on indigenous SWC in Africa. This project plans to
involve African researchers in case studies of indigeneous SWC techniques. Information can
be obtained from Chris Reij at CDCS, or Camilla Toulmin at IIED.
GATEKEEPER SERIES NO. SA27
References
Anderson, D. 1984. Depression, dust bowl, demography, and drought: the colonial state and soil
conservation in East Africa during the 1930's. African Affairs 83 (232), 321-343.
Ayers, A. 1989. Indigenous Soil and Water Conservation in Sub-Saharan Africa: the Circle of
Djenne, Central Mali. MSc thesis. University of Reading.
Bonvallot, J. 1986. Tabias et jessour du Sud tunisien; agriculture dans les zones marginales et
parade A l'6rosion. Cah. ORSTOM, s6r. Pedol., XXII (2), 163-171.
Boulet, J. 1975. Magoumaz; pays Mafa (Nord Cameroun). Paris, ORSTOM, Atlas des Structures
Agraires au Sud du Sahara 11.
Boutrais, J. 1987. Mbozo Wazan; Peul et montagnards au Nord du Cameroun. Paris, ORSTOM,
Atlas des Structures Agraires au Sud du Sahara 22.
Browder, J.O. 1989. Fragile Lands of Latin America: Strategies for Sustainable Development.
Boulder, U.S.A., Westview Press.
Campbell, D.J. and J.C. Riddell. 1984. Social and economic change and the intensity of land use
in the Mandara mountains of North Cameroon. Tijdschrift voor Econ. en Soc. Geografie 75 (5),
335-343.
Chambers, R. 1983. Rural Development: Putting the Poor First. London, Longman.
Chambers, R., A. Pacey and L.A. Thrupp (eds.). 1989. Farmer First: Farmer Innovation and
Agricultural Research. London, Intermediate Technology Publications.
Critchley, W.R.S., C.Reij and A. Seznec. 1991. Water Harvesting for Plant Production. Part II
Case-studies and Conclusions for Sub-Saharan Africa. To be published as a World Bank
Technical paper.
Delwaulle, J.C. 1973. Resultats de six ans d'observations sur l'6rosion au Niger. Revue Bois et
Forits des Tropiques 150, 15-36
Dietz, T. 1990. Interventions in indigenous irrigation and other aspects of environmental
management in Pokot and Marakwet areas, Kenya. In: Reader Workshop Environmental Man-
agement in Agricultural Development. Amsterdam, Royal Tropical Institute, 12 and 13 Novem-
ber, 1990.
El Amami, S. 1977. Traditional technologies and development of the African environments:
utilization of runoff waters: the meskats and other techniques in Tunisia. African Environment 3
(3/4), 107-120.
El Amami, S. 1983. Les amenagements hydrauliques traditionnels en Tunisie. Centre de
Recherche du Genie Rural, Tunis.
GATEKEEPER SERIES NO. SA27 17
Farrington, J. and A. Martin. 1988. Farmer Participation in Agricultural Research: A Review of
Concepts and Practices. Agricultural Administration Network, Occasional Paper No. 9. London,
O.D.I.
Gallais, J. and A.H. Sidikou. 1978. Traditional strategies, modem decision-making and manage-
ment of natural resources in the Sudan-Sahel. In: UNESCO. Management of natural resources in
Africa: traditional strategies and modem decision-making: 11-34. MAB Technical Notes no. 9.
Haagsma, B. 1990. Erosion and Conservation on Santo Antao (Cape Verde); No Shortcuts to
Simple Answers. Ribeira Grande.
Hallaire, A. 1971. Hodogway (Cameroun Nord). Paris, ORSTOM, Atlas des Structures Agraires
au Sud du Sahara 6.
Heusch, B. 1985. Conservation in North Africa: its status and related agricultural production. A
report to FAO in connection with the FAO Africa Study.
Heusch, B. 1986. Cinquante ans de banquettes de D.R.S. C.E.S. en Afrique du Nord: un bilan.
Cah. ORSTOM, s6r. P6dol. XXII (2), 153-162.
Ibrahim, F. 1988. Water harvesting and water spreading fur Flutanabau in der Sahelzone des
Sudan. Die Erde 119, 179-184.
IRA/ORSTOM, et. al. 1988. Utilisation et conservation des resources en sol et en eau (Nord
Cameroun). Rapport final.
Jodha, N.S. 1990. Mountain agriculture: the search for sustainability. Journal for Farming
Systems Research-Extension, 1 (1), 55-75.
Jones, R.K., E.M. Gichangi, J.M. Muthoto, S.K. Kitheka, D.M. Njarui and J.R. Simpson. 1989.
Development of "pitting" practices for rehabilitating eroded lands on smallholder farms in the
semi-arid tropics of eastern Kenya: a progress report. Paper present to VI ISCO Conference, Addis
Abeba, November 1989.
Kassogue, A. et T. Ponsioen. 1990. Le Plateau Dogon. Rapport d'etudes sur les techniques
traditionnelles de conservation des eaux et des sols. IIED/OXFAM. Unpubl. paper.
Klinken, M.K. van. 1987. The Pokot traditional furrow irrigation. The construction of arched
aquaducts in West Pokot District. ASAL Programme Kapenguria.
Kloosterboer, E.H. and L.A.A.J. Eppink. 1989. Soil and water conservation in very steep areas
- a case study of Santo Antao island, Cape Verde. In: Baum, E. et al., Topics in Applied Resource
Management 1, 1989, 111-142.
Krings, T. 1988. Die bedeutung des Bewisserungsfeldbaus futr die Subsistenzsicherung bei den
Dogon (Republik Mali). Die Erde 119, 185-192.
GATEKEEPER SERIES NO. SA27
Kutsch, H. 1982. Principal features of a form of water-concentrating culture on small-holdings
with special reference to the Anti-Atlas. Trier Geographische Studien Heft 5.
Lipton, M. 1988. The place of agricultural research in the development of Sub-Saharan Africa.
World Development 16 (10), 1231-1257.
Ludwig, H.D. 1968. Permanent farming on Ukara: the impact of land shortages on husbandry
practices. In: Ruthenberg, H.(ed.) Smallholder farming and smallholder development in Tanza-
nia. Miinchen, Weltforumverlag.
Martin, P. et C.Reij. 1990. Rapport d'une mission d'appui technique au programme special Niger
FIDA sous programme Conservation des Eaux et du Sol. Amsterdam, Centre for Development
Cooperation Services.
McCracken,J.A., J.N. Pretty and Conway, G.R. 1988. An Introduction to Rapid Rural Appraisal
for Agricultural Development. London, IIED.
MDRP/GEP. 1989. La preservation et la valorisation du system socio-ecologique Cap Verdien-
une demarche participative. Praia. Document provisoire.
Mensching, G. 1985. Slope erosion and its control in the traditional farm land of the Rif-Atlas
mountains in Morocco. In: Meckelein, W. and H. Mensching. Resource Management in
Drylands. Stuttgarter Geographische Studien 105.
Miehe, S. 1986. Acacia albida and other multipurpose trees on the Fur farmlands in the Jebel
Marra Highlands, Western Darfur, Sudan. Agroforestry Systems 4, 89-119.
Millington, A.C. 1984. Indigenous soil conservation studies in Sierra Leone. In: Challenges in
African Hydrology and Water Resources. IAHS Publication 144, 529 -538.
Nair, P.K.R. (ed.) 1989. Agroforestry Systems in the Tropics. Kluwer Academic Publishers in co-
operation with ICRAF.
Nyamapfene, K.W. 1987. A Short History and Annotated Bibliography on Soil and Water
Conservation in Zimbabwe. SADCC Report No. 12.
Pradeau, C. 1975. Adaptabilit& d'une agriculture tropical traditionnelle. Etudes Rurales 58, 7-
28.
Randall, J.R. 1963. Land use on the arid margin of the clays in Blue Nile and Kassala provinces.
In: Lebon, J.H.G.(ed) Proceedings 9th Annual Conference Philosophical Society of the Sudan,
1961.
Reij, C. 1983. L'&volution de la lutte anti-erosive en Haute Volta depuis l'independence: vers une
plus grande participation de la population. Amsterdam, Insitute of Environmental Studies, Vrije
Universiteit.
GATEKEEPER SERIES NO. SA27
Reij, C., S.D. Turner and T. Kuhlman. 1986. Soil and Water Conservation in Sub-Saharan Africa:
Issues and Options. Rome, IFAD. 82 p.
Reij, C., B. Dabire, P.Martin et A.H.Sidikou. 1986. R6publique du Niger. Programme de
Conservation des Eaux et du Sol. FIDA/Center for Development Cooperation Services, Vrije
Universiteit, Amsterdam.
Reij, C. 1989. The present state of soil and water conservation in the Sahel. OECD/Club du Sahel.
Paper presented to Club du Sahel/CILSS Meeting on La Gestion des Terroirs Villageois du
Sahel". Mali, Segou May 22 -27, 1989.
Richards, P. 1985. Indigenous Agricultural Revolution. London, Hutchinson.
Riddell, J.C. and D.J. Campbell. 1986. Agricultural intensification and rural development: the
Mandara mountains of North Cameroon. African Studies Review 29 (3), 89-106.
Roose, E. 1990. Diversity of Traditional and Modem Strategy for Soil and Water Conservation:
Ecological and Ethnical Impact in Sudano-Sahelian areas of Western Africa. Montpellier, OR-
STOM.
Sauvaget, C. 1981. Boua; village de Koude un terroir kabye (Togo septentrional). Paris,
ORSTOM, Atlas des Structures Agraires au Sud du Sahara 16.
Savonnet, G. 1958. Methodes employees par certaines populations de la Haute Volta pour lutter
centre l'erosion. Notes Africaines 78, 38-40.
Savonnet, G. 1976. Les Birifor de Diepla et sa region insulaires du rameau Lobi (haute Volta).
Paris, ORSTOM, Atlas des Structures Agraires au Sud du Sahara 12.
Sommerhalter, T. 1987. Boden- und Wasserkonservierende Masznamen im Ouaddai Geogra-
phique des Tschad. Unpublished mission report.
Westphal, E. 1981. L'agriculture autochthone au Cameroon. Miscellaneous papers 20. Agricul-
tural University Wageningen.
Temple, P.H. 1972. Soil and water conservation policies in the Uluguru mountains, Tanzania.
Geografiska Annaler 54A (3-4), 110-123.
Thomas, D.B. 1988. Conservation of cropland on steep slopes in eastern Africa. In: Moldenhauer,
W.C. and N.W. Hudson (eds.) Conservation Farming on Steep Lands. Ankeny, Iowa, Soil and
Water Conservation Society.
Wilken, G.C. 1987. Good Farmers: Traditional Agricultural Resource Management in Mexico
and Central America. University of California Press.
Young, A. 1989. Agroforestry for Soil Conservation. C.A.B. Intemational/ICRAF.
20 GATEKEEPER SERIES NO. SA27
ANNEX 1
EXAMPLES OF INDIGENOUS
SWC IN AFRICA
1. MOROCCO: ANTI-ATLAS (SOUTH MOROCCO)
In the Moroccan Anti-Atlas, where rainfall is of the order of 100-250 mm annually, there are three
indigenous techniques for water harvesting and concentration (Figure 1):
- Stone banks and small walls to hold back water running off slopes;
- Narrow troughs parallel to the slope (on particularly steep slopes);
- Terraces 25% of which reach to the crest of a hill or mountain; the other 75% are restricted to
the lower slopes.
Barley is the principal crop, and water harvesting techniques allow yields of 500 to 1000 kg/ha.
Recent trends include the deterioration of the drinking water supply in the larger basin landscape
may have detrimental effects upon agriculture (emigration, reduction of agricultural activities).
Three technical and agronomic measures have been suggested by Kutsch (1982) to improve this
water concentrating culture:
- More insistence on the importance of wedgeshaped reinforcement of the terrace walls on the
slope side to make them more durable;
- Systematic measures for improving the soil on terrace structures, especially on the upper
terraces;
- Green manuring, fodder cultivation, improvement of almonds' resistance to frost, cultivation
of other cash crops (safflower, etc.).
GATEKEEPER SERIES NO. SA27
Figure 1: Principal features of a form of water-concentrating culture on smallholdings of
the Anti Atlas (Kutsch, 1982)
Lay-out and functioning of the run-oft structures
1 Simple structure with stone banks and
additional furrows parallel to the slope
2 Structure of a small valley with
periodic rain-flows with small
walls (see also photograph 3)
3 Structure with stone banks with a
triangular profile (see also photograph 'I
Functioning of the run-off structures in the case of stone banks
Rainfall in situ
Especially wet zone
GATEKEEPER SERIES NO. SA27
2. TUNISIA: REGION OF MEDENINE (SOUTH TUNISIA)
In the Tunisian region of Medinine, where rainfall is 100-200 mm annually, population density
is 15-25 people per square km, and the soils are calcareous covered with several metres of loess,
there is an important traditional technique for floodwater harvesting within the streambeds. These
earthdams or tabias, which are 2-5m high and up to 100m long, are constructed across valley
floors. In 60% of the cases they are equipped with one or more lateral spillways (menfess), and
soils gradually accumulate upslope the tabias (Figure 2). The principal crops are fruit trees
(olives, almonds, etc.), but barley, lentils and beans are also grown.
These tabias, if adequately maintained, allow a certain control of floodwater. But the present
trend is lack of maintenance and slow abandonment of these indigenous techniques because of
strong migration, which is a direct threat to numerous small earth dams recently constructed (with
tractors) on relatively flat areas downstream.
State intervention has not been successful, as technical services introduced and imposed central
spillways, instead of lateral spillways in the tabias, which are costly to construct, prone to damage
and retain too little water (according to the farmers). Some 30% of the tabias are now equipped
with a central spillway (masraj).
In order to avoid or reduce damage to structures in the lowlands, Bonvallot (1986) proposes more
support from the State for the farmers in the highlands; he suggests the suppression of the central
spillways, because they are not appropriate, imposed from outside, and rejected by the farmers.
The State should also provide some labour and tools for maintenance.
GATEKEEPER SERIES NO. SA27
Figure 2: Tabias and jessour from southern Tunisia (Bonvallot, 1986)
GATEKEEPER SERIES NO. SA27
3. SOMALIA: TRADITIONAL SMALL SCALE WATER HARVESTING IN HIRAAN
REGION (CENTRAL RANGELANDS)
Traditional small scale water harvesting (WH) systems for crop production are common over an
extensive part of Hiraan Region in the Central Rangelands of Somalia, with up to half of all
families practicing some WH (Critchley et al, 1991). The majority of those involved are
agropastoralists who supplement their income from livestock by cultivation and simultaneously
benefit their animals with the fodder by-products. The systems used are simple and cost little to
construct.
The average annual rainfall for the region is 150-300 mm. Although most of the Central
Rangelands is sandy, those areas where WH is common are the clay zones, where adequate runoff
occurs, not far from the Shabelle river. Fertility is higher than in the sandy zone, but land is
traditionally rested after a number of years to restore the nutrient status.
Small scale WH techniques are implemented voluntarily by the local agropastoralists because,
they say, harvested runoff improves crop performance and, without it, crop failure may occur.
These techniques have evolved over several generations. In certain areas of Hiraan Region, such
as Bula Burti and Halgan, few farms can be observed without caag or gawan systems. These are
low cost systems. The caag system is found where considerable overland flow, or flow from a
small toog (gully) can be captured behind bunds, whereas the gawan system acts mainly as an in
situ moisture conservation system on the plateaus, though it is modified sometimes to accommo-
date runoff from outside.
The caag system is used where slopes are above 0.5% and there is significant runoff to be
harvested. These are long slope, external catchment systems. Water may be diverted into the plot,
commonly from small toogs or even road drains (Figure 3). "Contour" bunds are formed within
the fields, typically to a height of 50 cm and base width of 150 cm. These bunds commonly extend
across the entire plot (often one hectare or more in size), excess runoff typically being spilled
around one upturned arm of the bund. The other arm extends higher upslope. Overflow then may
be collected in front of a second, lower bund. There are rarely more than two such bunds in a field.
An alternative overflow system occasionally used is the incorporation of a plastic pipe (of approx.
10 cm diameter) within the bund. The layout of the bund is achieved by a combination of eye and
experience, but a contour alignment is clearly the objective. A precise maximum depth of
flooding is not specifically designed for; rather it is said that if excess flooding occurs and water
stands for more than five days the bund may be deliberately breached. In practice flooding depths
can reach 30 cm.
The techniques used on gawan sites are on the borderline between WH and in situ moisture
conservation. These are typically almost flat sites which sometimes receive runoff from outside.
The system itself comprises bunds of about 30 cm height, which divide plots of land into a grid,
with individual basins in the order of 500 m2 and upwards in size (Figure 3). Sometimes the basins
are "blind", that is they have no inlet or outlet. However, in the situation where runoff is expected
from outside the plot, water is spread and controlled by the provision of gaps or breaches in the
bunds, which act as spillways. Sometimes runoff is allowed to back-up and spill around a sidewall
of the basin, in a similar fashion to the caag system. The configuration of two fields is never
identical.
GATEKEEPER SERIES NO. SA27
Sorghum is the usual crop of choice, providing grain and also stover which is a valued by-product
for livestock feed. Cowpeas are also common. Sesame is sometimes grown, and if sorghum
becomes waterlogged, then sesame is sometimes planted later into the residual moisture. Dry
planting is practised by some people, though others wait until after the rains have begun. Two
crops are taken during each year, if both the rains succeed. Yields for WH fields are not
specifically available, though general average yields for this area are given as 415 kg/ha for
sorghum and between 330 and 530 kg/ha for cowpeas.
These WH systems are entirely traditional. They are implemented by the farmers themselves at
their own expense. No details are available about costs or labour input, but these are estimated
to be modest, judging by the costs incurred by the Central Rangelands Development Project
(CRDP) on trial plots using similar techniques. The gawan system holds rainfall effectively in
situ and where runoff is harvested it works best where an overflow is provided for, around side
bunds. The less sophisticated models with gaps in bunds are less efficient at impounding runoff.
Smaller basins could be an improvement and this indeed is one of the modifications being tested
under CRDP.
The caag system is a simple and effective method of impounding and spreading flows of runoff
from small channels. Indeed it is basically the type of system which is being introduced under a
number of projects in sub-Saharan Africa, usually at much greater expense and with little extra
efficiency. Nevertheless it is evident that there are some problems with waterlogging in some
situations and also with bund breakages. The caag system could thus be improved in a number
of ways, principally by the introduction of simple surveying instruments to establish the contour
and to determine vertical intervals for spacing of bunds and positioning of spillways.
CRDP are working toward recommendations for extension in the field of water harvesting.
Without revolutionising the techniques or cropping pattern, there are several useful modifications
which could improve the systems and make them more productive and sustainable. These will
form the basis of an extension programme. There may be potential for the introduction of
alternative crops. Other promising crops being tested under CRDP include bulrush (pearl) millet
and green grams mungg bean). CRDP are developing an agropastoral system for cropping in the
Central Rangelands and extension packages are available for sandy zones. It is anticipated that
similar cropping practices would be viable under water harvesting. The basis of the system is the
incorporation of woody shrubs and trees to help maintain fertility as well as to provide economic
products such as timber, fuelwood and browse. A further benefit would be the stabilisation of the
bunds by the establishment of vegetation.
Within the same district, large scale water spreading schemes are also found, whereby toogs of
50 metres width or more are diverted into flood plains and cultivation carried out. A number of
such schemes were started as local initiatives. While these can be very productive, they are only
appropriate for specific sites and commonly suffer from breaches of the diversion bund itself.
GATEKEEPER SERIES NO. SA27
Figure 3: Traditional small scale water harvesting, Somalia
<~ ~
Traditional Small Scale WH I "Caag" System
Central Rangelands, Somalia
artist's impression
slope
I
earth
bund
example of "Gawan" system
GATEKEEPER SERIES NO. SA27
4. SUDAN:TRADITIONAL TERAS CULTIVATION IN THE EASTERN PLAINS
Teras cultivation is a traditional form of WH used extensively in the clay plains of Eastern Sudan
(Critchley et al, in press). The word teras itself refers to the earthen bund which surrounds three
sides of each cultivated plot, and impounds runoff from the plains. The teras system is of
particular interest because it is one of the few examples of traditional WH spread over a very
extensive area. This profile concentrates on one focal point of teras cultivation: the area around
Kassala.
Several ethnic groups practice teras cultivation. Around Kassala the Hedendwa (basically cattle
herders, many of whom have now settled) are the principal practitioners, though some Rashaida
(tent-living semi-nomads) also grow crops under "teras" systems. In the Butana plains, major
groups using these techniques are the Shukriya and the Lahawin, both mainly herders. Although
there is a considerable number of pure pastoralists within Kassala Province, only 12% of the 1.5
million population are considered true nomads. Population density over Eastern Region is about
7 people/square km, though there is considerable local variation.
Kassala is on the threshold of the semi-desert zone to the north, but it is just within the arid zone,
which has 225 400 mm of annual rainfall. The area used for teras cultivation are clay plains.
Often these plains are alluvial in origin, such as the old Gash Delta near Kassala. Many of the clays
show some degree of cracking and indeed there are some vertisols cultivated under this system.
Typically the plains are open with few trees. The trees which do include Balanites aegyptiaca and
Acacia spp. In some places, especially close to settlements and roads, Prosopis juliflora has
become quite densely naturalised.
While there is a considerable range in the design of the structures under teras cultivation, plots
around Kalahout, west of Kassala town, serve as an example. The visual impression of the
landscape is of a checkerboard, with individual plots interspersed within an uncultivated plain.
The plain acts as the catchment, and the catchment: cultivated area ratio is rarely below 2:1 and
is sometimes considerably greater. The system is essentially an external catchment, long slope
technique of water harvesting (Figure 4).
A typical plot may be up to 3 hectares in size: in this case the bottom bund (straight, but
approximating to the contour) is 300 metres or more in length and the upslope arms or wingwalls,
aligned at right angles to the bottom bund, are between 50 and 100 metres long. In addition to
the side arms which define the extent of the plot, there are usually other parallel bunds every 50
metres or less within the plot. The earthen bunds are 35-40 cm in height, after settling, with a base
width of 1.5 2 metres. Most bunds in the Kalahout area have been created with a disc plough,
but the few bottom bunds which have been built up with a front loader are as high as 75 cm. There
is no deliberate provision of a spillway for the evacuation of excess water. On these slopes of
approximately 0.5% the bottom bund will normally breach (or be breached) before runoff can
back up around the tips of the side bunds. In a description of teras cultivation in Butana from the
1960s, the catchment area is referred to as the sadra and the cultivated area as the hugna (Randall
J.R., 1963). Within the main hugna there was sometimes a smaller bunded plot termed a gataa,
which would collect a lesser depth of runoff and could be planted earlier.
GATEKEEPER SERIES NO. SA27
Figure 4: Traditional teras cultivation in Sudan
54 F ~~~~~'5
.2*
~- v>
Traditional "Teras" Cultivation
Eastern Plains, Sudan
artist's impression
Various Patterns of "Teras" Systems
(Butana Eastern Sudan)
from Lebon 1967
"gataa" subsidiaryy cultivated area)
"hugna"
V(main culti-
vated area)
GATEKEEPER SERIES NO. SA27
The teras system is essentially for sorghum production. One of the most common varieties grown
is feterita, a white sorghum with a pigmented testa. Another common variety in the Kalahout area
is aklamoi, a brown seeded, goose-necked variety. Planting is in holes at one metre by one metre
spacing. Yields of sorghum are said to reach 750 kg/hectare in an average-to-good season, though
they are usually lower. It is quite common to see watermelons planted on the bottom bunds, where
they can benefit from the relatively good supply of runoff.
The teras system is entirely traditional and all operations are either carried out by the cultivator
manually, or by hired tractors which are becoming more common. The only example of direct
assistance from an outside source in the Kassala area was the provision of a front loader for
bunding by the Soil Conservation Department in 1987. Maintenance of the bunds is required
seasonally, but this is a modest task unless breaches have occurred.
The system is very widespread in specific areas. Between Kassala junction and the Atbara River,
there are said to be thousands of such plots. The most significant area for teras, in the country are
the Butana plains, to the west of the Atbara River. There are no training or extension programmes
specific to the teras system.
Teras systems are not exclusive to the Eastern Region of Sudan and variations can be found as far
away as South Darfur, where such bunding is apparently sometimes associated with cultivation
of the flood plains of wadis. However, as the term "teras" refers to the bund itself, not all teras
systems are WH systems; some teras on hillsides are for soil conservation.
Within Eastern Region, water spreading from khors is commonly used for cultivation. Often
spreading occurs naturally when the khor bed fans out into a wadi or onto a plain. Sometimes local
cultivators have built traditional diversions, and even used brushwood barrier to collect wind-
driven sand which thus develop into bunds which help spread runoff.
GATEKEEPER SERIES NO. SA27
5. CAMEROON: MANDARA MOUNTAINS (NORTH CAMEROON)
Twenty different ethnic groups practice WH in the Mandara Mountains of Northern Cameroon,
though the Mafa are the most skilled practitioners (Boulet, J. (1975); Riddell and Campbell
(1986). Rainfall varies from about 700 mm in the plains to 1100 mm on the plateaus. Population
density is some 250 per square km.
Farmers have an intricate system of indigenously developed terracing of steep slopes. For fertility
management, they use household, animal wastes and crop residues. Manure is spread and worked
into the soil, and crops are rotated and intercropped. Some cattle are stalled year round. The two
most important trees growing on the terraces are Acacia albida and Khava senegalensis.
There have been recent attempts to settle mountain people in the lowlands, which have not been
successful. Development efforts so far have not built on, nor have they utilised the existing
indigenous knowledge base. Riddell and Campbell (1986) suggest that external intervention
might be appropriate in a number of areas:
- water development
- weed and pest management
- increasing crop inventories
GATEKEEPER SERIES NO. SA27
6. SUDAN:DJEBEL MARRA HIGHLANDS (WESTERN DARFUR)
In the Djebel Marra Highlands of western Darfur, the Fur tribe practice terracing systems (Miehe,
1986). Rainfall is 600-1000 mm, and the soils are dark humic montane under humid highlands
conditions and ash soils. Some 60% of the massif surface is terraced. The population density is
currently some 20-37 persons per square km, down considerably compared with 50 years ago.
Dung is highly valued, and Acacia albida dominates over wide areas where the terrace soils
contain large proportions of volcanic ashes. Crop yields were 650-800 kg/ha for millet and
sorghum in 1982, a rather bad year.
There is currently considerable migration to the lowlands, inducing a switch from an intensive
agroforestry system to simple extensive bushfallow or shifting cultivation practices. There is a
decline of tree canopy, and irrigation development on existing terraces based on cash crops.
GATEKEEPER SERIES NO. SA27
GATEKEEPER SERIES NO. SA27 33
GATEKEEPER SERIES
1. Pesticide Hazards in the Third World: New Evidence from the Philippines. 1987.
J.A. McCracken and G.R. Conway.
2. Cash Crops, Food Crops and Agricultural Sustainability. 1987. E.B. Barbier.
3. Trees as Savings and Security for the Rural Poor. 1988. R.J.H. Chambers.
4. Cancer Risk and Nitrogen Fertilisers: Evidence from Developing Countries. 1988.
J.N. Pretty and G.R. Conway.
5. The Blue-Baby Syndrome and Nitrogen Fertilisers: A High Risk in the Tropics?
1988. J.N. Pretty and G.R. Conway.
6. Glossary of Selected Terms in Sustainable Agriculture. 1988. J.A. McCracken and
J.N. Pretty.
7. Glossary of Selected Terms in Sustainable Economic Development. 1988.
E.B. Barbier and J.A. McCracken.
8. Internal Resources for Sustainable Agriculture. 1988. C.A. Francis.
9. Wildlife Working for Sustainable Development. 1988. B. Dalal-Clayton.
10. Indigenous Knowledge for Sustainable Agriculture and Rural Development. 1988.
D.M. Warren and K. Cashman.
11. Agriculture as a Global Polluter. 1989. J.N. Pretty and G.R. Conway.
12. Evolution of Agricultural Research and Development Since 1950: Toward an
Integrated Framework. 1989. R.E. Rhoades.
13. Crop-Livestock Interactions for Sustainable Agriculture. 1989. W. Bayer and A.
Waters-Bayer
14. Perspectives in Soil Erosion in Africa: Whose Problem? 1989. M. Fones-Sondell.
15. Sustainability in Agricultural Development Programmes: The Approach of
USAID. 1989. R.O. Blake.
16. Participation by Farmers, Researchers and Extension Workers in Soil Conservation.
1989. S. Fujisaka.
GATEKEEPER SERIES NO. SA 27
17. Development Assistance and the Environment: Translating Intentions into Practice.
1989. M. Wenning.
18. Energy for Livelihoods: Putting People Back into Africa's Woodfuel Crisis. 1989. R.
Mearns and G. Leach.
19. Crop Variety Mixtures in Marginal Environments. 1990. J. Jiggins
20. Displaced Pastoralists and Transferred Wheat Technology in Tanzania. 1990. C. Lane
and J.N. Pretty.
21. Teaching Threatens Sustainable Agriculture. 1990. R.I. Ison.
22. Microenvironments Unobserved. 1990. R. Chambers.
23. Low Input Soil Restoration in Honduras: the Cantarranas Farmer-to-Farmer Extension
Programme. 1990. R. Bunch.
24. Rural Common Property Resources: A Growing Crisis. 1991. N.S. Jodha
25. Participatory Education and Grassroots Development: The Case of Rural Appalachia.
1991. J. Gaventa and H. Lewis
26. Fanner Organisations in Ecuador: Contributions to Farmer First Research and Develop-
ment. 1991. A. Bebbington
27. Indigenous soil and water conservation in Africa. 1991. Chris Reij
28. Tree Products in Agroecosystems: Economic and Policy Issues. 1991. J.E.M. Arnold
29. Designing Integrated Pest Management for Sustainable and Productive Futures. 1991.
Michel P. Pimbert
30. Plants, Genes and People: Improving the Relevance of Plant Breeding. 1991. Angelique
Haugerud and Michael P. Collinson
Copies of these papers are available from the Sustainable Agriculture
Programme, IIED, London (2.50 each inc. p and p).
GATEKEEPER SERIES NO. SA 27
The Sustainable Agriculture Programme
IIED
The Sustainable Agriculture Programme of IIED promotes
and supports the development of socially and environ-
mentally aware agriculture through research, training,
advocacy, networking and information dissemination.
The Programme emphasises close collaboration and con-
sultation with a wide range of institutions in the South.
Collaborative research projects are aimed at identifying
the constraints and potentials of the livelihood strategies
of the Third World poor who are affected by ecological,
economic and social change. These initiatives focus on
indigenous knowledge and resource management; par-
ticipatory planning and development; and agroecology
and low external input sustainable agriculture.
The refinement and application of Participatory Rural
Appraisal methods is an area of special emphasis. The
Programme is a leader in the training of individuals from
government and non-government organizations in the
application of these methods.
The Programme supports the exchange of field experi-
ences and research through a range of formal and informal
publications, including RRA Notes, aimed at practitioners
of Rapid and Participatory Rural Appraisal, and the Gate-
keeper Series, briefing papers aimed at policy makers. It
receives funding from the Swedish International Develop-
ment Authority, the Ford Foundation, and other diverse
sources.
International Institute for
Environment and Development
3 Endsleigh Street,
London WC1H ODD, UK
Telephone: 071-388 2117
Fax: 071-388 2826
Telex: 261681 EASCAN G
|
