Use of Spatial Analysis for Global Characterization of Wheat-Based Production Systems

Authors: HODSON, D - CIMMYT INT MEXICO; White, Jeffrey

Submitted to: Journal of Agricultural Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/13/2006
Publication Date: 2/13/2007
Citation: Hodson, D.P., White, J.W. 2007. Use of Spatial Analysis for Global Characterization of Wheat-Based Production Systems. Journal of Agricultural Science 145: 115-125.

Interpretive Summary: To meet expected global demand for wheat, the International Center for Maize and Wheat Improvement (CIMMYT) increasingly uses software tools such as geographic information systems (GIS) to guide priority setting and provide a strong geographic perspective to specific research activities. This paper reviews potential applications of GIS and related technologies to global wheat research, drawing from experiences at CIMMYT. Wheat crops occupy 200 million hectares globally, ranging from sea level to over 3500 m elevation and from the equator to above 60°N latitude. Using information of climate, soils and crop management, CIMMYT used GIS to classify regions into twelve wheat megaenvironments using quantitative criteria based largely on temperature and rainfall regimes. The new classification is valuable for targeting varieties and agronomic practices to appropriate areas globally as well as to examine potential impacts of global change. For example, in high yield potential, irrigated wheat production areas of south Asia, global warming might result in an estimated additional 200 million people residing in areas where wheat production experiences heat stress. GIS also permits exploring scenarios for potential disease epidemics, such as for the stem rust race Ug99, where analyses suggest the race may spread via prevailing winds from East Africa to the Middle East and eventually to Asia. Other examples are discussed, noting the value of moving from static descriptions to ones that consider year to year variation in climate as well as global change. Overall, GIS is proving invaluable for providing a stronger, more quantitative geographic perspective to wheat research.

Technical Abstract: Research centres of the Consultative Group for International Agricultural Research (CGIAR) have made major contributions to agricultural development, but there is concern that future impact will only come with much more targeted and knowledge-intensive interventions. Increasingly, application of modern tools and technologies are crucial elements in order to support and enhance the effectiveness of international agricultural research. Bread and durum wheats (Triticum aestivum and T. durum) occupy an estimated 200 million hectares globally, are grown from sea-level to over 3500 masl, and from the equator to above 60 °N latitude in Canada, Europe, and Asia. For organizations like CIMMYT, which seek to improve wheat production in the developing world, understanding the geographic context of wheat production is crucial for priority setting, promoting collaboration, and targeting germplasm or management practices to specific environments. Increasingly important is forecasting how the environments, and their associated biotic and abiotic stress patterns, shift with changing climate patterns. There is also a growing need to classify production environments by combining biophysical criteria with socioeconomic factors. Geospatial technologies, especially geographic information systems (GIS), are playing a role in each of these areas, and spatial analysis provides unique insights. Use of GIS to characterize wheat production environments is described, drawing from examples at CIMMYT. Since the 1980s, the CIMMYT wheat programme has classified production regions into mega-environments (MEs) based on climatic, edaphic, and biotic constraints. Advances in spatially disaggregated datasets and GIS tools allow MEs to be characterized and mapped in a much more quantitative manner. Parallel advances are improving characterizations of the actual (versus potential) distribution of major crops, including wheat. The combination of improved crop distribution data and key biophysical data at high spatial resolutions also permits exploring scenarios for disease epidemics, as illustrated for the stem rust race Ug99. Availability of spatial data describing future climate conditions may provide insights into potential changes in wheat production environments in the coming decades. There is a pressing need to advance beyond static definitions of environments and incorporate temporal aspects to define locations or regions in terms of probability or frequency of occurrence of different environment types. Increased availability of near real-time daily weather data derived from remote sensing should further improve characterization of environments, as well as permit regional-scale modelling of dynamic processes such as disease progression or crop water status.