This publication presents a GIS-based approach for revising the definitions of global maize production environments, called “mega-environments” (MEs), used by CIMMYT and its partners. A cluster analysis was performed on climate data, representing a four-month growing season, for key maize producing locations. Assuming rainfed production, the onset of the growing season was determined based on the month when the ratio of precipitation over potential evapotranspiration exceeds 0.5. Diagnostic criteria for mapping MEs were based on cluster analysis results and expert knowledge. The resulting maps can be used to select appropriate target environments for maize germplasm and trials, as well as in priority setting and site selection for global maize breeding programs.

Modern industrial agriculture is not sustainable because of its heavy reliance on petroleum, a non-renewable source of the energy used in farming, and because of pollution caused by petroleum products such as fertilizers and pesticides. A systems analysis of farming suggests that agriculture will be more sustainable when services of nature, such as nutrient recycling by soil micro-organisms and natural controls of insects, replace the services now provided by energy from petroleum. Examples are drawn from the Southeastern USA, but lessons learned can be applied worldwide.

The projected increase in world population levels and the subsequent rise in food demand represents a huge challenge for agricultural production systems worldwide. This publication examines the opportunities and challenges raised by the use of plant genetic resources and highlights the contribution that data from multi-environment yield trials can provide for the definition of adaptation strategies and yield stability targets in plant breeding programmes. It contains a case study about a durum wheat crop programme in Algeria, and also includes a CD-ROM with data from IRRISTAT, a programme developed by the International Rice Research Institute (IRRI).