Department of Agricultural, Food, and Resource Economics Feed the Future Innovation Lab for Food Security Policy

Todd Benson, Athur Mabiso, and Flora Nankhuni. 2016. Detailed Crop Suitability Maps and an Agricultural Zonation Scheme for Malawi: Spatial Information for Agricultural Planning Purposes. Feed the Future Innovation Lab for Food Security Policy Research Paper 17. East Lansing: Michigan State University

EXECUTIVE SUMMARY
Although relatively small, Malawi is a country with significant agro-ecological diversity reflecting the diverse landforms associated with the Great Rift Valley that runs the length of the country. The effects of those landforms on soil formation processes, local climates, the distribution of water resources, and vegetation patterns results in a variety of quite complex farming systems across Malawi that farmers have developed to exploit distinct combinations of local agricultural resources. One important implication of this variety of farming systems is that the comparative advantage of different areas of Malawi for the production of different crops, livestock, and other agricultural products, whether for commercial purposes or for household food security, will differ significantly from place to place. In consequence, any agriculture plans that assume that a single technical approach to improving agricultural productivity and increasing the economic returns that households receive from their agricultural livelihoods will fail in many places. In designing agricultural development policies for the country, the geographic diversity of Malawian agriculture needs to be considered.

This document describes a two-level agricultural zonation scheme to guide agricultural planning in Malawi. This scheme combines broad agricultural development domains – based upon a district-level analysis of agro-ecological potential; physical access to market; and population density – with an extensive set of detailed, more locally relevant crop suitability maps to determine where agricultural development investments might best be located within a relevant development domain.