Ethnopedology for solving problems of soil management and sustainable agriculture implementation in West Africa Savannah regions

Recently many scientific studies have shown the relevance of ethnopedological informations in the different agro-ecological and socio-cultural context of Africa. However, information did not sufficiently serve in solving the problems of land degradation and famine affecting seriously Sub-Saharan nations. Strategies of best and wide use of indigenous knowledge are to be refined. The main objective of the present article is to analyze the ethnopedological informations and to discuss the possibilities of their integration to databases development. The structure and the quality of ethnopedological informations allow their integration in large and dynamic databases for better soil and environmental management and sustainable agriculture. Such databases must also include information from conventional academic studies (as biophysical, socio-economic and market studies).


INTRODUCTION
It is now widely recognized that traditional knowledge has been used during centuries for the rational management of natural resources, biodiversity and agro-ecosystems (Warren, 1992a;Barrera-Bassols and Zinck, 2003).Recently, many scientific studies, especially from the 1990s, have shown the relevance of local knowledge related to soils and land resources.Indigenous soil knowledge is very complex and include, among others, soil classification, farming skills and agronomic behavior of soils.This knowledge, in literature, has multiple qualifiers (traditional, local, native, indigenous ...) which are interchanged, but does not have the same etymological meaning.It refers, in all cases to ethnopedology.We must remember that the ethnopedology, compared with ethnobotany and ethnozoology, is newly structuring (Barrera-Bassols, 2003), from the combination of natural and social sciences (soil science, geopedological inventory, social anthropology, rural geography, agronomy).Since 1989, the average scientific studies in ethnopedology is 33 per year (Barrera-Bassols and Zinck, 2003), resulting in a rapid growth of the discipline.In West Africa, particularly in the savannah regions, old and recent studies have been conducted in different agro-ecological and sociocultural situations (Dabin, 1951;Warren, 1992b;Diallo and Keita, 1995;Diallo et al, 1998;Mikkelsen and Langohr, 2004;Yacubu et al., 2014) The main objective of this review article is to analyze the ethnopedological informations and to discuss the possibilities of their integration to databases for better soil and environmental management and sustainable agriculture.

Difficulties
Generally, scientists working on indigenous knowledge (IK) have often noted the difficulties of access to this knowledge (Gadgil et al., 1993;Diallo et al., 1998).These difficulties appear to be related to the complex nature of IK where concepts and practices are interconnected with local beliefs and religious practices.In some cases, access to such knowledge is hampered by the inadequacy of the methods used by modern science (Diallo et al., 1998).The difficulties do not always permit the identification of relevant information from local knowledge, those utilizable in dynamic databases for solving current development problems.

Research process for relevant information
For better access to indigenous knowledge, each academic discipline of modern science must develop appropriate methods.
A method applicable in ethnopedology was proposed in Mali (Diallo et al., 1998).It should include surveys of local perceptions of land, conventional field soil characterization, and soil samples analysis in laboratory.Previously, the physical context of the study must be defined relative to a known agroecological zone framework.People whose knowledge is sought by the study should be clearly identified in relation to linguistic and socio-economic frameworks well- Diallo and Diallo 813 defined.

Surveys of local perceptions of lands
They must start with a consultation with representatives of the community, for example a village or group of villages.During a meeting, the researcher must be clear about the objectives of the study and create conditions for open collaboration.Subsequently, well-targeted surveys should cover a limited number of people (5 for example) but well recognized in the community as people particularly well informed on the soils and land resources.
Questions should afford a list of soil types and characteristics that farmers attribute to them.This list, after the first few tentative inquiries, should be improved later.

Conventional field soil characterization
This characterization must be conducted with farmers.All soil types listed above must be identified in situ, using the criteria of the farmer and those of conventional pedology.
The opening of graves should be performed at locations determined from the indications of farmers.The description of the environment of soil profile should particularly emphasize on topography, water regime, vegetation and land use practices.In the description of soil profile, attention must be grant to the characteristics which allow judging the abilities of agricultural land and anticipating problems arising from their cultivation.However, observations related to soil processes should not be neglected.Soil samples should be collected from representative profiles of soil types for laboratory analysis.

Soil analyses in the laboratory
Each soil type identified in a local system can be attached, without much difficulty to a well known scientific classification.The analyses will focus on some detail properties (particle size distribution, carbon, nitrogen, and pH, etc).Further analysis can be viewed in terms of financial and material resources and information already accumulated on the soil type.

Summary of collected data
The information provided by farmers and those obtained using the soil science methods (in situ and in laboratory) should be considered as a whole.Thus, it becomes possible to establish a diagnosis, define the average characteristics of each soil type of the studied system

ETHNOPEDOLOGICAL INFORMATION QUALITY
The quality of information from ethnopedology is an important factor when it is question to introduce this information in complex databases.Studies conducted by academic institutions, particularly in Africa, permit today a pertinent analysis of soil information quality from ethnopedology.

Structure of soil classification in ethnopedology
The soil classifications studied in Mali (Dabin, 1951;Diallo and Keita, 1995;Diallo et al., 1998) are mainly from Bambara and Malinke ethnic groups, which are in same dialectal entity (Manding dialect).Manding dialect is a spoken language in many countries of West Africa (Burkina Faso, Gambia, Guinea, Ivory Coast, Mali, Senegal, Sierra Leone).The criteria used to distinguish soils are the topographic position, the texture and gravel content and the color of the surface horizon.
Texture and gravel content criteria occur unambiguously in all peasant soil classifications and at higher level.It permits to distinguish three main types of soil: Gravelly soil (Bèlè dugukolo), sandy soil (Cincin dugukolo) and clay soil (Bogo or Boi).
Color, as the texture is a criterion always used, but at a secondary level.So, in textural categories, distinctions are still made according to color: black (fing), red (blé ou oulé), and white (diè).
In some indigenous soil classification, soil type can be set from a particularly well displayed behavior.This is the case of Dakissè dugukolo, a soil type of a local classification in Kangaba region, Upper Niger basin in Mali.It is a subtype of Lè dugukolo, developed on an alluvial terrace of Niger River.Here, the term dakissè refers to the granular structure of the soil.In the local language, dakissè means the grain of Hibiscus sabdariffa. 1) is not always observed in all ethnopedology system, and classification can be limited to a repertory of soil types (Table 2).However each soil type has its attributes, enough clear without any confusion possibility.

Hierarchical classification (Table
The use, texture and color as criteria in indigenous soil classification has been noted by many other researchers working in West Africa (Warren, 1992b;Mikkelsen and Langohr, 2004;Yacubu et al., 2014) and other parts of the World (Barrera-Bassols and Zinck, 2003).

Ethnopedological information related to soil qualities
In different agro-ecological situations, farmers know clearly the capabilities of every soil type of their classification system and also, the sensitivity of theses soils to process as fertility decline under cultivation, runoff and erosion, etc.This observation appears in studies carried in Bagoe watershed (Sikasso region, Mali) as shown in Tables 3 and 4, according Kassogué (2013).

Ethnopedological information and development of data base for soil management and sustainable agriculture
Ethnopedological information as examples describe above, are utilizable in data base for soil and  environmental management and the implementation of sustainable agriculture.In this order, attention can be paid to local classification system, local people perceptions of each soil type (it is fundamental properties, abilities and sensitivity to major process, as the rate of fertility decline, proliferation of weeds, runoff and erosion).However, data from conventional academic studies are very useful and irreplaceable (Figure 1).But it is necessary to organize their acquisition and their use with maximum attention.They must include data related to the biophysical and climate context, socio-economic characteristics, and market information both at global and local levels.

CONCLUSION
The structure and the quality of ethnopedological information allow their integration in large and dynamic databases for better soil and environmental management and sustainable agriculture.Such databases must also include information from conventional academic studies (as biophysical, socio-economic and market studies).

Table 2 .
Indigenous non hierarchical soil classification (case of Djitoumou, agro-ecological zone, Mali) in accordance with

Table 4 .
Farmers appreciation of soils sensitity to runoff and erosion in Bagoé basin (Sikasso region, Mali) in accordance with Kassogué (2013) Some categories of information needed for local date base for soil management and sustainable agriculture (adapted fromDiallo, 2004).