Simulating the impact of climate change on rice production in Asia and evaluating options for adaptation.

The likely effects of climate change caused by increasing atmospheric carbon dioxide levels on rice production in Asia were evaluated using two rice crop simulation models, ORYZA1 and SIMRIW, running under fixed-change' climate scenarios and scenarios predicted for a doubled-CO2 (2xCO2) atmosphere by the General Fluid Dynamics Laboratory (GFDL), the Goddard Institute of Space Studies (GISS) and the United Kingdom Meteorological Office (UKMO) General Circulation Models. In general, an increase in CO2 level was found to increase yields while increases in temperature reduced yields. Overall rice production in the region was predicted by the ORYZA1 model to change by 6.5, -4.4 and -5.6% under the GFDL, GISS and UKMO 2xCO2 scenarios, respectively, while the corresponding changes predicted by the SIMRIW model were 4.2, -10.4 and -12.8%. The average of these estimates would suggest that rice production in the Asian region may decline by -3.8% under the climate of the next century. Declines in yield were predicted under the GISS and UKMO scenarios for Thailand, Bangladesh, southern China and western India, while increases were predicted for Indonesia, Malaysia, and Taiwan and parts of India and China. Modification of sowing dates at high latitudes, where warmer temperatures allowed a longer growing season, permitted a possible transition from single-cropping to double-cropping at some locations, an adaption that could potentially have a large positive impact on national rice production in some countries. Planting date could also be adjusted to avoid high temperatures at the time of flowering which can cause severe spikelet sterility in some varieties, although a delay in planting in some cases may prevent a second crop from being obtained because of high temperatures later in the season. Selection for varieties with a higher tolerance of spikelet fertility to temperature was shown to be capable of restoring yield levels to those predicted for current climates. The use of longer-maturing varieties to take advantage of longer growing seasons at higher latitudes may instead result in lower yields, due to the grain formation and ripening periods being pushed to less favorable conditions later in the season. A better strategy might be to select for shorter-maturing varieties to allow a second crop to be grown in these regions