The issue of spatial scale in hydro-economic modeling of global and national food and water systems to address environmental and hunger policy questions

There are numerous models available that aim to address food and water policy at different spatial scales. The question to be asked of these models is "What is the importance of spatial scale on hydro-economic modeling used to address environmental and hunger policy questions?"; The International Food Policy Research Institute's (IFPRI) IMPACT-Water model was evaluated at 2 different spatial scales (69 basins vs. 281 basins) and the results from each version were compared to evaluate the importance of spatial scale on environment and hunger policies. Most indicators and results such as those related to hunger require comparison at the local/regional scale. In order to provide a detailed analysis comparing the results between the two different spatial scales of IMPACT-Water, three case studies at the regional scale were chosen to represent different hydro-climates and economic heterogeneity: Central Asia, Europe, Southern Sub-Saharan Africa. Results from this analysis imply that spatial scale does have an impact on model results used to inform environment and hunger policy. Impacts are stronger in regions of economic and hydro-climate heterogeneity.; To evaluate yet another, more detailed spatial scale issue, the second part of the research focuses on evaluating the importance of spatial scale and management on river basin modeling for global food production. Four case studies were evaluated (Missouri River Basin, Senegal River Basin, Yellow River Basin, Volta River Basin) in addition to performing the analysis at 2 different global river basin representations; one with 69 basins and another with 126 basins. In general, one risks the possibility of overestimating available water in basin representations where rivers are in parallel. If this occurs in areas where irrigated agriculture is a significant contributor to the global irrigated production, one may be greatly overestimating the potential of global irrigated agriculture. There is little to no impact on basin representation where the main river is in series. While spatial representation may not be an issue, modeling the correct management may be. Global modelers must use caution in aggregating basin representations. One must determine the layout of each basin representation and the corresponding level of management to aid them in completing a useful and representative analysis.