Agroforestry and local knowledge
Contrary to conventional perceptions, where appropriate land and tree tenure have prevailed, increasing human population density in rural areas has often been associated with increasing tree cover and the development of other measures associated with sustaining farm productivity, says Dr Fergus Sinclair, University of Wales, Bangor (UK). Mary Tiffen's work on environmental recovery in Machakos, Kenya is probably the best known example, but there are several other well documented examples in Africa, such as where increasing population density and hence labour availability have led to sustainable agricultural intensification, rather than degradation, around Kano in northern Nigeria.
These developments have generally involved a large number of locally available tree species, encouraged by farmers in various niches within their farms - that is, the development of mosaics of complex and biodiverse agroforestry practices. In many cases, a sophisticated local understanding of how trees interact with crops and affect soil has been generated by farmers and is largely complementary to what is known scientifically. DFID together with the Ford Foundation, the Rockefeller Foundation and Winrock International have funded research over the past five years in Tanzania (dry rangeland), Kenya (fruit trees on farms), Sri Lanka (forest gardens), Thailand (miang tea in thinned, evergreen forest) and Nepal (fodder trees on hill farms) that has led to the development of a general methodology for acquiring local ecological knowledge from farmers for use in agroforestry research and extension.
In Nepal, subsistence farmers have detailed knowledge of a wide range of native trees which are important as livestock fodder during the dry season. They classify a number of tree species in their locality into subtypes, that are not botanically recognised, according to leaf morphology and nutritional value. Furthermore, the farmers have several classification systems related to the nutritive value of fodder, its ability to satisfy animal appetite and the effects of trees on associated crops and soil. Some of these classifications by farmers have been found to correspond to biologically pertinent attributes of fodder such as protein content and overall dry matter digestibility.
Research workers from NRI, University of Wales, Bangor and CSIRO derived a fuzzy mathematical model of this detailed local knowledge, which converts the qualitative data into quantitative terms to explore the potential for combining farmer knowledge about a wide range of tree species with quantitative scientific data and understanding. Predicted values for supplemented fodder intake and milk yield with three tree fodder species, consistent both with farmer practice and scientific data, were obtained from the model.
Appreciation of the sophistication of local knowledge has had a profound effect on research and development, and knowledge-based systems that combine knowledge from farmers and scientists have been recognised as a potentially powerful tool for interpreting decision making by farmers in complex situations. It is anticipated that this type of approach could be applied more widely to make more complete use of local qualitative knowledge gathered in participatory rural development projects.
