Contents - Previous - Next

Development of a framework for holistic land characterization and development at different scales


In the past, national institutions and also the various FAO Services that dealt with individual aspects of natural resources and their use, tended to go their own way, both conceptually and in the execution of field programmes.

Soil specialists gave much weight to soil profile characterization and classification, and their mapping work centred on these classification units. This occurred even though the landscape-ecological context and the concept of land units were obvious and easy guides in the delineation of soil units, associations and complexes. There were instances in the early FAO field programmes where soil mapping specialists who wanted to use geomorphological criteria as a framework for mapping legends were actually dissuaded from doing so by their superiors.

Hydrologists had methodologies of their own, concentrating on the lateral dynamics of water resources, and rarely using landscape units as unifying criteria. Irrigation and drainage specialists tended to identify spatial units on the basis of rigid grid observations on meso-topography and soil-water properties. Civil engineers engaged in road building and other construction work had their own sampling techniques at fixed distances, disregarding any information from other disciplines that might be available.

Soil fertility and fertilizer promotion specialists tended to forego soil classification criteria and soil mapping units as starting points, preferring random or grid experimental and demonstration plots (one of the reasons being that soil classification criteria centred on the subsurface layer rather than the topsoil).

Vegetation and forestry specialists, too, paid scant attention to land units, and the same held true for specialists on agroclimatic conditions - even though the latter should include near-surface and soil climatic aspects.

Such monodisciplinary activities often resulted in a tangle of boundaries of land management units when maps of field information on the various natural resources were combined. As a consequence, land use planning and its execution had no clear-cut natural terrain units as a starting point.

Wim Sombroek Director, Land and Water Development Division, FAO, Rome

FIGURE 1. Landscape-related nutrient dynamics in mixed farming systems

1 Local nutrient mining

3 Local nutrient enrichment

5 Anthropic nutrient export

7 Possible nutrient fatigue

2 Nutrient harvesting

4 Natural nutrient import

6 Anthropic nutrient import

8 Integrated nutrient husbandry

It should be mentioned that physical geographers have been trying all along to advocate a landscape and catchment area approach. This resulted in the "catena" concept of Milne, the "land system" approach of Australia, the "geo-pedo-morpho-hydrological" and "landscape ecological" concepts of Troll, Tricart and others (see Vink, 1986). These concepts were, however, not easily adopted by civil and hydraulic engineers, soil scientists, agronomists and foresters respectively, because of different educational backgrounds, professional experience, and location in different, sometimes competing, institutions and ministries. One of the practical reluctances in accepting a landscape-ecological approach consisted of the fact that geomorphologists among themselves were often in disagreement on a unified approach; they developed a number of very fancy schemes on geomorphodynamics but not one single classification scheme on landforms that could be easily understood and used by non-specialists on the subject.


With the advance of aerial photography, satellite remote sensing and Geographic Information Systems as tools, the advantages of a holistic approach to land characterization and land management have become obvious. Time has come for international organizations such as FAO, as well as national institutions, to start cooperating more effectively, on the basis of unifying concepts. The Agro-Ecological Zoning system of FAO has achieved a degree of integration at scales between continental (1:5 million) and about 1:50000 (district or province-level), but hydrological and vegetational aspects have not yet been harnessed in the system.

The degree of holism hinges on the definition of "Land". I advance the following one (Sombroek, Brinkman and Gommes, 1993), now being incorporated in the draft texts for an Intergovernmental Framework Convention for Control of Desertification and Drought ("desertification" understood to be "land degradation in dryland areas"):

"Land is a delineable portion of the earth's terrestrial surface, encompassing all attributes of the biosphere immediately above or below this surface, including those of the near-surface climate, the soil and terrain forms, the surface hydrology (including shallow lakes, rivers, marshes and swamps), the near-surface sedimentary layers and associated groundwater and geohydrological reserve, the plant and animal populations, the human settlement pattern and the physical results of past and present human activity (terracing, water storage or drainage structures, roads, etc.)."

In this holistic approach, a unit of land has both a vertical aspect - from atmospheric climate down to groundwater resources, and a horizontal aspect - an identifiable repetitive sequence of soil, terrain, hydrological and vegetation or land use elements ("landscape", "land unit" or "terroir" units). Mineral resources and deeper geohydrological resources (confined aquifers) would, however, be excluded from land attributes.

One should therefore integrate all compartments vertically; from groundwater-related qualities, through qualities of soil profile, soil surface, slope position and vegetative cover, to overhead climatic qualities. The qualities of the vegetative cover came to the fore at the recent involvement in agro-ecological and socio-economic zoning of Amazon countries.

TABLE 1. Land qualities (amplified from Sombroek 1994)


• Atmospheric moisture supply: rainfall, evaporation, dew formation.
• Atmospheric energy for photosynthesis: temperature, daylength, sunshine conditions.
• Atmospheric conditions for crop ripening, harvesting and land preparation: dry-spell occurrence.
• Liability to atmospheric calamities: hazard of tornadoes, hailstorms, etc.


• Value of the standing vegetation as "crop" (e.g. timber).
• Value of the standing vegetation as germ plasm (biodiversity value).
• Value of the standing vegetation as protection against soil degradation.
• Value of the standing vegetation as protection for crops and cattle against adverse atmospheric influences.
• Hindrance of vegetation at introduction of crops and pastures: the land "development" costs.


• Surface receptivity as seedbed: the tilth condition.
• Surface treadibility: the bearing capacity for cattle, machinery, etc.
• Surface limitations for the use of implements (stoniness, stickiness, etc.): the arability.
• Spatial regularity of soil and terrain pattern: the degree of freedom at determining the size and shape of fields with a capacity for uniform management.
• Surface liability to deformation: the occurrence or hazard of wind and water erosion.
• Accessibility of the land: the degree of remoteness from means of transport.
• Surface water storage capacity of the terrain: the presence or potential of local "waterholes", on-farm reservoirs, bunds, fish ponds, etc.
• Surface propensity to yield runoff water (for local water harvesting or downstream water supply).
• Accumulation position of the land: degree of fertility renewal and/or crop damaging by overflow or overblow.


• Physical soil fertility: the net moisture storage capacity in the rootable zone.
• Physical soil toxicity: the presence or hazard of waterlogging in the rootable zone (i.e. the absence of oxygen or the excess of CO2).
• Chemical soil fertility: the availability of plant nutrients.
• Chemical soil toxicity: salinity or salinization hazard; excess of exchangeable aluminium.
• Biological soil fertility: the N-fixation capacity of the soil biomass; the microbial capacity for the transformation of fresh soil organic matter into readily available plant nutrients.
• Biological soil toxicity: the presence or hazard of soil-borne pests and diseases.


Groundwater level and quality in relation to (irrigated) land use.
• Substratum potential for water storage (local use) and conductance (downstream use).
• Presence of unconfined freshwater aquifers.
• Substratum (and soil profile) suitability for foundation works (buildings, roads, canals, etc.).
• Substratum (and soil profile) as source of construction materials.

One should also integrate all aspects horizontally at the landscape level. This is the land unit approach of physical geographers, which takes into account the typical, micro-geographically repetitive elements of terrain, top or plateau, scarp or upper slope, main slope, lower slope or springline, bottomland or flood plain; with their mutual influence whether natural or under current land use. This influence can be in the sense of internal hydrology (for instance, rainfall moving into the soil of the plateaux and surfacing at the springline, including the lateral movement of chemical substances such as salts and silica), or the surface transport of soil material through erosion from upper slopes and accumulating in the bottomland or flood plain (see also Falkenmark text). Either of these processes can be detrimental or positive at the receiving end, depending on the rate of transport and the prevailing climatic conditions. The lateral influence also relates to chemical soil fertility. Nutrients may be transformed downslope by natural processes, or from outlying land to arable fields near homesteads in traditional farming systems.

It is true that the water component of the above holistic definition of "land" has some special attributes: bulkiness, mobility, transience or evasiveness and, at the same, time is site-specific in its management. Most of the water attributes can be harnessed in a land-centred GIS system. In practice, however, one may want to continue to use "water" and "land" as of equal weight (as in the AGL Division of FAO); then the "land" definition should be less holistic and be restricted to soil and topographic features (land sensu strictu; ss)

Whether used sensu latu or sensu strictu, a key unifying element of land characterization is the landscape unit. Only in rare cases, when the topography is homogeneous over large distances - the Great Plains of the USA or Russia; the Cerrado lands around Brasilia - may one disregard the landscape or land unit concept. In most actual field situations, the land facet analysis of a land unit is of paramount importance for sustainability assessments. This is also the main reason behind the "soil-and-terrain" (SOTER) approach of ISSS/ISRIC (Oldeman, this workshop), now also adopted by FAO for inventories at 1:1 M and more detailed scales.

The various attributes of units of land can be translated into land qualities (see Table 1), as a starting point for multidisciplinary land evaluation and subsequent land use planning. Implementation of the latter should take into account the current land use practices and patterns and the overall socioeconomic conditions, as separate layers in a GIS system, and should employ land use negotiation methods to become acceptable to all stakeholders (Röling, 1994).

For the purpose of our discussions on concepts, frameworks, and strategies in a holistic approach, one may usefully make distinctions among local, i.e. village-level landscapes; subcatchment areas; whole river basin, major lake or coastal catchments within a climatic zone; and international river basins that traverse different climatic zones and major geological structures. In this scaling sequence the first two units can best be illustrated with transverse cross sections and block diagrams, while the others are discussed more easily in a longitudinal context.

In the following, I will use some examples from my own field experience in Africa and Latin America, but all participants will have their own examples as well. I have no readymade proposals on the different approaches and the set of parameters to be considered with each of the above geographic entities and scales.

Local, village-level landscapes

• Scope of intervention: physical improvement of land, water and plant nutrient conservation and management on a village scale, with full participation of the local community ("gestion de terroir").

• Scales of inventory and evaluation: 1:5 000 to 1:10 000

• Transverse links between land attributes and land use pattern

• Mainly transverse drainage influences

• Prime building block of soil-and-terrain units in the SOTER approach

• Examples:

- Integrated water-soil-nutrient management in small foothill-and-valley complexes within the Himalayan range: forthcoming NET/FAD project
- Plant nutrient husbandry in semi-arid to sub-humid lands of Africa: Hausaland.
- Rainwater harvesting in microcatchments of desert areas: Negev, see Bruins, Evenari and Nessler, 1986

Subcatchment areas

• Scope of intervention: district-level multipurpose land use planning for (re)settlement schemes

• Scales of inventory and evaluation: 1:50 000 to 1:100 000

• Links between land attributes and between land use patterns mainly transverse

• Dendritic drainage influences

• Several soil-and-terrain units, often with the same land facets or components but in different percentages

• Examples:

- The integrated rural development planning of the Keita district in the Sahelian zone of the Niger Republic: ITA/FAO's glossy publication, and Sombroek's notes of January 1994 (Appendix 1)
- The Fouta-Djallon integrated development project of Guinea: UNDP/FAO project starting

Coastal zone series of catchments,(within same ecological zone)

• Scope of intervention: integrated coastal zone management (upland forestation and agricultural land use; irrigation and drainage of the plains; protection of marshes and swamps; fisheries and aquaculture; coastal defence works; water supply to towns; harbour development; tourism)

• Scale of inventory and evaluation: ± 1:250 000

• Links between land use pattern and freshwater - brackish water - mainly longitudinal seawater conditions

• Parallel drainage and two-way flooding pattern

• Limited number of soil-and-terrain units, but clearly distinguished in composition

• Examples:

- The lowlands of Southern Sumatra and Southern Kalimantan: Indonesian transmigration projects
- The Laguna Merin - Laguna dos Patos areas of coastal Uruguay and Southern Brazil: UNDP/FAO Laguna Merin Irrigation and Drainage Development project (Sombroek, Averbeck and Duran, 1969)

River basin catchment areas,(mainly within same ecological zone)

• Scope of intervention: catchment-level assessment of water resources and their potential use, especially downstream (conservation or reforestation projects in upstream parts, irrigation and drainage projects in lower parts; some fishery development); identification and protection of high biodiversity values; national parks and indigenous delineation of reserves

• Scale of inventory and evaluation: 1: 100 000 to 1: 250 000

• Links between land and water attributes and land use patterns are mainly longitudinal

• Converging drainage and river flow pattern

• Many soil-and-terrain units, only partly with the same land facets in different percentages

• Examples:

- The Rima-Sokoto Valley project in Northwestern Nigeria: UNDP/FAO, 1966; Sombroek and Zonneveld, 1971
- The Lower Mekong River Development: many Committee reports
- The Amazon region: agro-ecological and socio-economic zoning proposals of FAO for Brazilian part and the region at large
- Tajo River basin, Spain: Martinez Beltrán, 1993

Major lake catchment areas,(within same ecological zone)

• Scope of intervention: as under river basin areas, but with special attention to influence of midstream water use on the sustainability of fishery in the lake; aquifer management; polluting influences, including salinization

• Scale of inventory and evaluation: 1: 500 000 to 1: 1 000 000

• Patterns of land attributes and land use in circular bands around the lake

• Centripetal drainage and flooding pattern

• Many soil-and-terrain units, often without any land facet links

• Examples:

- The Lake Chad basin: UNDP/Unesco Synthèse Hydrologique du Basin du Lac Chad (1969) and FAO IAP/WASAD Sub-Regional Action Plan for the Basin (1993)
- The Aral Sea region (Uzbekistan, Kazakstan, Turkmenia): numerous reports
- The Lake Victoria catchment area: FAO/TCP project
- The Lake Tanganyika catchment area: GEF/OPS project

International river basins, (traversing several distinct ecological and geological zones)

• Scope of intervention: modelling of the conservation and development of upstream water resources for downstream water use, taking into account existing international agreements on water apportioning

• Scale of inventory and evaluation: 1: 1 000 000 to 1: 5 000 000

• Very complicated links between land and water attributes and land use patterns; GIS subsystems required

• Mainly longitudinal river discharge elements, including interruption structures (dams; major diversions)

• A multitude of very diverse soil-and-terrain units; appropriate SOTER "shell" approach, with climatic and geohydrologic conditions as essential relational data bases

• Examples:

- The Nile basin: modelling studies of USAID/FAO, ongoing (Appelgren, 1994)
- The Euphrates-Tigris basin: pending regional peace agreements
- The Indus-Ganges system of the Indian subcontinent and the Himalayan countries:
- the Agro-Ecological Zoning studies of Bangladesh, under present conditions and under a scenario of global climate change (UNEP/FAO/EC/IIASA project, starting)

Note: There is merit in treating major deltas as separate entities because of their specific characteristics (Nile Delta, Mekong Delta, Orinoco Delta, etc.)


At each level of scaling or geographic aggregation, as outlined above, a multidisciplinary and holistic approach is warranted and feasible. However, the weight between the different disciplines, the approach and the tools to be used are different, as a consequence of the different main purposes of the studies and the interventions envisaged. The smaller the scale, i.e. the larger the area under consideration, the more the longitudinal and water-flow related aspects come to the fore.


Appelgren, B.G. 1994. River basin management as a strategy for implementing national water resources policy. In: Management of rivers for the future. MARDI, Kuala Lumpur, Malaysia.

Bruins, H.J., M. Evenari and U. Nessler. 1986. Rainwater-harvesting agriculture for food production in arid zones: the challenge of the African famine. Applied Geography 6, 13-32. Butterworth, London.

Martínez Beltrán, J. 1993. Soil survey and land evaluation for planning, design and management of irrigation districts. CIAM (Centre for International Advanced Mediterranean Studies). Zaragoza, Spain.

Röling, N. 1994. Platform for decision-making about ecosystems. In: The future of the land: Mobilising and integrating knowledge for land use options (proceedings of a Conference of Wageningen Agricultural University, August 1993). John Wiley, London.

Sombroek, W.G. 1994. The work of FAO's Land and Water Division in sustainable land use, with notes on soil resilience and land use mapping criteria. In: Greenland, D. and I. Szabolcs, 1994. Soil resilience and sustainable land use (proceedings of a symposium in Budapest, October 1992). CABI, Wallingford, United Kingdom.

Sombroek, W.G., R. Brinkman and R. Gommes. 1993. Land degradation in arid, semi-arid and dry-subhumid Areas. I. Definitions, concepts and databases. FAO internal paper.

Sombroek, W.G. and I.S. Zonneveld. 1971. Ancient dune fields and fluviatile deposits in the Rima-Sokoto River Basin (Northwestern Nigeria); geomorphologic phenomena in relation to Quaternary changes in climate at the southern edge of the Sahara. Soil Survey Papers 5, STIBOKA, Wageningen.

Sombroek, W.G., H. Averbeck and A. Duran. 1969. Soil studies in the Merim Lagoon Basin, Uruguay/S. Brazil, (report, maps and cross sections). CLM/PNUD/FAO Report, Treinta y Tres, Uruguay.

Vink, A.P.A. 1986. Soil survey and landscape-ecological survey. In: Annual Report 1985. ISRIC, Wageningen.

Appendix 1: The integrated rural development project of Keita (Niger): its sustainability and replicability (Some technical notes)


The Keita area of 5000 km2 is located in the centre of a specific wedge-shaped geologic zone, which has its lower tip about 30 km south of Sokoto in Nigeria, its western edge going through Birnin'konni to 100 km north of Tahoua, and its eastern edge running along Bouza to 40 km south of Abalak.

Most of the area east and west of this wedge has acidic gritty, sandy and clayey deposits of continental origin, vast plains of ancient aeolic and fluviatile deposits, and scattered low laterite plateaux. The Sokoto-Keita wedge, however, has mainly marine strata, a very thick and extensive lateritic crust in many places, a broken topography in between, extensive impermeable colluvia, sub-recent dune formations of permeable fine sands, and fertile sandy to loamy alluvia in the lower parts of the valleys. This combination of landscape features - for details see the following section - has resulted in a relatively dense human population, of predominant agricultural orientation. It has doubled to 160 000 persons in the Keita area itself in the period just before the great drought of 1965, causing a relatively high degree of degradation of land and water resources and an acute need for additional arable land.

The Keita project, started in 1984 and to continue to June 1996, aims at alleviating this predicament by a combination of technical and socio-economic interventions, in close collaboration with local authorities and very intensive village-level people's participation. The physical interventions, varying from small windbreaks to dam construction (for details see below) are being carried out in a very pragmatic way - learning by experience - and have by now resulted in a recuperation and improvement of the physico-biotic conditions in about 40-50% of the main, eastern part (3400 km2) of the Keita sub-prefecture. Many of the simple technical works have been carried out through WFP-arranged food-for-work participation by the local population, especially women (about 20 000 of the able men are leaving the area for the dry season to work in more southern countries: "exodants").

As shown strikingly during the reporter's field visit in July 1992, when a two-hour torrential rainfall of 130 mm (i.e. one third of the average yearly rainfall) caused substantial flood damage and loss of precious water, the package of interventions has to be extended systematically to all vulnerable landscape elements of the Keita area. This will require outside financial and technical support through the current tripartite arrangement NER/ITA/FAO, as well as WFP assistance for another five to ten years. Only then can one expect that the still-growing population can take full itself of adequate and further improved land management on a sustained basis.

Wim Sombroek, Director, Land and Water Development Division,FAO, Rome

Land improvements and agricultural in Keita


Restoration of plateau based on subsoiled cultivated or planted strips alternating with bare strips for water harvesting


Rehabilitation of steep slopes with infiltration trenches


Development of village reservoirs and livestock water points above weirs and dams constructed to contra peak flows of seasonal streams


Rehabilitation of bottom slopes based on subsoiled cultivated or planted strips alternating with bare strips for water harvesting


Development of dry-season farming on lowlands following Runoff


Development of water ponds in the valley bottom allowing irrigation of cropland


Construction of dams and weirs in the streams using local materials and gabions allowing some groundwater recharge and flooding of lowlands for cultivation

It should be kept in mind that the successes of the Keita project thus far are the result not only of relatively high financial inputs, but for a good part are also due to the exceptionally good cooperation between the foreign technical advisers and their local counterparts, the large degree of delegation of executive action to the project team in the field, the pragmatism of the approach per landscape element, and the village-level co-decision making by the population itself.



To be able to judge to what degree the Keita experience can be replicated in other parts of the country and elsewhere, one should take advantage of the following:

• more systematic information on the economics of the Keita intervention, to be available in easily accessible form and making use of the existing data bank at the project. A start has been made through the engagement of APOs and students, and is likely to be strengthened by the proposed cooperation of PEICRE (a combination of Italian universities and scientific institutions). Strengthening of the agro-economic and hydro-engineering economic component of the project team itself will be useful, too. In this context the recommendation should be noted for an APO to monitor hydrological changes as a result of the interventions;

• basic data on the physico-biotic and the socio-economic conditions of representative sites of the other physiographic regions subject to desertification, in comparison with the Keita situation. To substantiate the need for a comparative study, it is necessary to describe the Keita situation in some detail, concentrating on the physical conditions and interventions (cf. Carucci 1990; Laurent 1991; twin maps at 1:10 000 scale on physiographic units and intervention proposals [1985-1992]).

Specific physico-biotic conditions of Keita

Rainfall is restricted to the period June-September (hivernage) and occurs rather irregularly over the years. The present average is about 350 mm (500 mm before 1966) often as heavy showers (50-60 mm) with strong erosive power. The dry season (contre-saison) is very hot and has a strong dry wind (harmattan) during its central months.

Physical Features

The following landscape elements (facies) can be discerned:

- plateaux of the clayey to gritty Adat Doutchi formation [Kalambaina]1 with a veneer of acid loamy soil of low permeability (sealed surfaces);

1 Between square brackets the equivalent naming in the adjoining Rima-Sokoto area of Nigeria (Sombroek and Zonneveld, 1991).

- formations dunaires and associated fine sand layers, locally on the central plateau parts [Illela coversands], very permeable;
- versants de plateau: rocky upper scarps;
- collines rocailleuses: hills below the level of the plateaux, often isolated, covered with lateritic and clayey materials of an older geologic formation [range, Dukamaie] of marine clays and marls; low permeability;
- glacis: colluvia of local geologic materials, compact loamy soils of low permeability;
- formations dunaires stabilisées et versants ensablés of very permeable fine sands [Illela coversands]. Locally, dunes that turned active after the drought of 1965;
- vallées, nearly flat river terraces of alluvial sandy to loamy material; fertile soils of high permeability, locally with extra seasonal water supply by the floods.

The drainage system is characterized by ephemeral streams called "koris", with 20-30 periods of flooding, lasting from a few minutes to some days.

Vegetation and traditional land use

The plateaux, versants and collines are traditionally used for silvipastoral purposes, but their grassy and shrubby cover is sparse. The formations dunaires and the vallées are the traditional areas of rainy season arable cropping, mainly of millets, with fair yields only (300 kg/ha). The valleys, which have a relatively dense arboreal presence (Acacia albida, and others), are in part also used for dry season cropping (onions as an important cash crop) wherever extra flood water happens to have accrued or the ground water level is shallow.

Degradation aspects

Due to the strong population increase, overgrazing and the diminished rainfall since 1965, the vegetation of the plateaux, upper versants, collines and glacis has severely degraded, resulting in accelerated run-off and impoverished grazing grounds. Some of the formations dunaires have become active during the drought years, resulting in loss of arable land. The valley lands have been subject to increased gully erosion and bankfalls along the ephemeral rivers, and their ground water level has gone down, causing shortage of drinking water for man and animal.

Project interventions

The objectives of the project were: to arrest the loss of fertile soil, to control the run-off from the upper areas, to stabilize the moving dune areas, to restore ground water levels in the valleys, to improve their microclimates, to increase the total area of arable land by using parts of the plateau and glacis areas for arable cropping rather than for grazing, and to secure alternative animal feed or fodder. To reach these objectives, the following technical works have been or are being executed:

- fixation des dunes vives through dead hedges of millet stalks (10 m spacing) and tree replanting in between;
- brise-vents on some of the valley lands (50 x 100 m spacing), to create a dry season improved microclimate - reportedly resulting in 30% higher yields of the crops in between;
- banquettes sur les glacis (80 m length, 25 m spacing) to arrest their run-off and to make the non-stony loamy soil receptive for rainwater infiltration through yearly ox ploughing ("unites de culture attelée"); bunds of the banquettes protected from degradation by trampling and soil structure collapse through a tight cover of laterite blocks - the arable land so gained is reported to give yields of 530 kg/ha, with successful tree planting immediately above the bunds;
- banquettes sur les plateaux, type sylvo-pastoral (100 m long, 15 m wide, with a water harvesting area or "impluvium" of 30 m width in front) - resulting in a triple amount of water inside the banquettes for use by trees and fodder. Applied on those plateau parts where soils are shallow and very stony.
- banquettes sur les plateaux, type sylvo-agricole (100 m long, 15 m wide, impluvium area of 30 m). Applied to those plateau parts where soils are somewhat deeper, but requiring a yearly subsoiling of 50 cm with heavy machinery to break up sealing and compaction - reported to give crop yields inside the banquettes of 650 kg/ha.

For both types of banquettes the bunds need to be protected by a cover of laterite blocks to prevent their degradation:

- tranchées de reboisement (3 m length, 50 cm width, 60 cm depth, density about 700 per ha), dug concentrically on the collines rocailleuses, to generate fodder and fuel and for run-off and water erosion control; using local and some exotic (Prosopis) tree species. This most remarkable feature of the Keita interventions is done nearly exclusively by women's hand labour through a WFP food-for-work arrangement;
- seuilles d'épandage in the smaller koris to control their water flow and bring water to the adjoining land or refill ground water levels; constructed with gabions of laterite blocks;
- barrages d'écrêtage avec déversoirs in the upper subcatchments to control water flow downstream, to create cultivable land behind the dams, to provide water for man and animal in the dry season and to develop fisheries. Constructed partly with gabions of laterite blocks, partly with soil material and a protective sheet of laterite blocks.

In addition to these structural works, the project carries out a number of complementary activities in the socio-economic field: pistes de liaison entre les villages, puits villageois et maraîchers, écoles, dispensaires, magasins, abris pour les moulins, alphabétisation, nutrition, santé, systèmes créditaires, all in response to the wishes of the local population and with their effective participation.

Conclusion on Keita

The above summary of the physical and socio-economic conditions and interventions may make it evident that there are a number of specific favourable aspects in Keita:

• pronounced topographic difference and areas with impermeable substrata favouring dam construction;

• areas with impermeable surface conditions favouring water harvesting;

• abundance of lateritic materials suitable for bund protection, for preparation of gabions as basic construction blocks, and for road building;

• a substantial percentage of traditional arable land as backbone for sustainability and development, including flat lands that will benefit from water redistribution techniques in the dry season;

• a basically agricultural rather than pastoral orientation of the population, a relatively high density of this population and strong population growth, and a willingness to cooperate in community efforts, whether or not stimulated by food-for-work arrangements; altogether making the restoration of land and water resources and the realisation of extra arable land (including animal feed production) a feasible technical and socio-economic undertaking.


The Keita combination of conditions is unlikely to exist in most other parts of the regions (the Abalak extension of the Keita project included). More modest combinations of interventions will be required, but their precise nature is still to be established. It is proposed that an inventory be carried out, through a special, short project, of the physico-biotic and socio-economics of four to six test areas, constituting representative subcatchments of major sub-Sahelian regions of the country. In such test areas the various landscape elements would be quantified as to their relative percentage, their physical characteristics (agro-geo-hydro-morpho-pedologic features), their actual land use, their degree of degradation and their supply of construction materials, especially laterite blocks. This should be accompanied by a review of the socio-economic conditions of these regions: population size and pressure; agricultural or pastoral orientation; amount of spare labour force; willingness to cooperate with external interventions, etc.

The results of these inventories are then to be compared with the economics of the various interventions at Keita, with the purpose to establish which particular combinations of these interventions are likely to be appropriate for each of the regions.


Carucci, R. 1990. Aperçu sur l'approche territoriale et méthodologies d'interventions dans la lutte contre la désertification de l'arrondissement de Keita. In: Les actes du séminaire national sur l'aménagement des sots, la conservation de l'eau et la fertilisation à Taoua. A. Berrada. February 1989. INRAN, Niamey. pp. 15-28.

Laurent, J.F. 1991. Aménagement des sous-bassins versants de la vallée de Keita, rapport de stage. Université Laval, Québec.

Contents - Previous - Next