Simulated rice yields as affected by interannual climate variability and possible climate change in Java

Abstract

About 60% of the rice produced in Indonesia is grown in the fertile soils of the island of Java. Introduction of the high-yielding rice varieties and improvement of cultural technique have increased rice production, and self-sufficiency was attained in 1984. However, increasing population and decreasing land for rice cultivation could threaten the food supply in the country. Rice production is also threatened by interannual climate variability and possible climate change. To provide policy- makers and planners with information to formulate a strategy to cope with interannual climate vari- ability and the possible climate change, rice yields of 2 production areas on Java were simulated using the DSSAT (Decision Support System for Agrotechnology Transfer) rice growth simulation model. The crop model predicted lower rice yields for different management options, compared with experiment plots, but predicted yields similar to or slightly higher than the farmers' yield. In general, the predic- tions relate quite well. The GISS, GFDL, and UKMO climate models predicted higher rainfall, solar radiation, and temperature in both locations. In the higher rainfall and lower temperature of the West Java site, the climate change scenarios reduced rice grain yield in both the first and second crops. Dur- ing normal years in the relatively warmer and dryer climate of the East Java site, there was no signifi- cant yield reduction due to climate change, except under the UKMO scenario in the second crop. Because high temperature and CO2 concentration favor rice growth, development of more heat-toler- ant varieties probably can compensate for the yield losses due to climate change in the future. Except for the GISS and GFDL climate scenarios in the first crop and the baseline climate scenario in the sec- ond crop in the West Java site, higher yield losses were predicted because of interannual climate vari- ability. Since the dry spell threat is more imminent and frequent, to improve preparedness a short-term climate prediction for the tropical region is urgently needed.