Contrasting approaches to integrating indigenous knowledge about soils and scientific soil survey in East Africa and Bangladesh

Methodologies for collection and integration of scientific and indigenous soils knowledge are discussed in relation to two interdisciplinary projects involving soil scientists, other natural scientists and anthropologists. In Uganda and Tanzania, participatory methods paralleled scientific soil survey. Indigenous or `localʹ soil classification was explored by a semistructured, iterative discussion with farmers, resulting in classes that could be related to scientific taxa. However, the relation of farmersʹ cognitive soil maps to scientific soil maps in the Geographical Information Systems (GIS), developed as an integration domain, was problematic. In-depth analysis was only achieved through geo-referencing local knowledge (LK) using global positioning system (GPS). In Bangladesh, ethnographic methods obtained local soils knowledge and its sociocultural context, and accompanied scientific surveys of floodplain soils and agroecosystems. Subsequent data processing included database and GIS tools, but there were problems systematically relating the two knowledge bases. The sequencing of, and iteration between, methods used to access and analyse geo-referenced scientific and local soils knowledge are critical considerations in such research.