We investigate the effects of different climate scenario resolutions on estimates of the impacts of future climate change on agriculture in the United States. Climate scenarios were developed using both a coarse resolution, global scale general circulation model and a spatially more refined regional climate model, nested within the coarse model. The scenarios are similar on a very broad regional scale, but show important differences on a subregional scale. In most areas the fine scale scenario produces a more severe climate change. Simulated changes in crop yields (e.g., cotton, soybean, corn, wheat) were constructed under both the coarse and fine scale scenarios for the conterminous United States. The results demonstrate that the spatial scale of climate scenarios affects the estimates of regional changes in crop yields on several levels of spatial aggregation and the economic impact on the agricultural sector as a whole. For the elevated CO2 case, national economic welfare increased under the coarse scale climate scenario, but remained virtually unchanged under the fine scale scenario. With adaptations, both scenarios showed substantial increases, but these were still considerably larger for the coarse scale scenario. Regional indicators of economic activity were of opposite sign in some regions, based on the scenario scale for both cases. Such differences in economic magnitudes or signs become important in public policy debates concerning climate change. Hence refinement of spatial scale of scenarios should be carefully considered in future regional integrated assessments.
