Evaluation of regional climate scenario data for impact assessment of climate change on rice productivity in Korea

Abstract

Spatial evaluation of the uncertainty associated with climate data would allow reliable interpretation of simulation results for regional crop yield using gridded climate data as input to a crop growth model. The objective of this study was to examine the spatial uncertainty of regional climate model data through determining optimal seeding date with the ORYZA2000 model for assessment of climate change impact on rice productivity in Korea. The optimal seeding date was determined at each grid point using regional climate model outputs under the RCP 8.5 scenarios. In major rice production areas such as inland plain regions, where temperatures of regional climate data were relatively accurate, the optimal seeding date determined using those gridded data were reasonable. However, areas with complex terrains including areas near bodies of water, e.g. coastal areas, riverbasins, lakes, and mountainous areas, had a relatively large uncertainty of the optimal seeding date determined using the regional climate data. These results indicated that the uncertainty of regional climate data at a high spatial resolution of 12.5 km should be taken into account in the regional impact assessment based on crop growth simulations in Korea. In addition, further studies would be merited to assess the impact of climate change on rice yield at an ultra-high spatial resolution of 1 km in Korea. Crop yields were projected to decrease after the 2020s when crop yield simulations from inland plain areas were considered, which suggested that adaptation strategies should be established and implemented in the near future.