Rice Yield Simulation In Nanjing Under Climate And Diffuse Radiation Change Scenarios

Climate changes in future will impact grain yield. In this study, we chose rice growing in Nanjing located in the plains of Yangtze River delta. Based on modified and verified rice model ORYZA2000, we simulated the rice yield from 2016 to 2100 responding to the changes in temperature, CO₂ concentration, and diffuse radiation fraction which caused by atmospheric aerosol optical depth （AOD） under RCP2.6, RCP4.5, RCP6.0, RCP8.5, the four scenarios in the Intergovernmental Panel on Climate Change （IPCC）AR5.The main research results are:（1） In ORYZA2000 model, the diffuse radiation fraction in solar radiation is an important factor which influences crop yield. Yield and total aboveground dry matter are sensitive to diffuse radiation fraction changes. When the change in diffuse radiation fraction reaches 10%, the sensitivity coefficient of yield to diffuse radiation fraction is 1000. And the sensitivity coefficient decreased with the diffuse radiation fraction increased. Under the current radiation levels, when diffuse radiation fraction increased by about 30%, rice yield and total aboveground dry matter reached the maximum. （2） Comparing the simulated diffuse radiation fraction with the observed value, we found the mean errors at three representative moments chosen by model （around 12:30-13:00,14:30-15:00.16:30-17:00 respectively） are 9.9%,13.1% and 16.7%. The root mean square errors in the three times are 0.12,0.14, and 0.16, respectively. And, the relationship between simulation errors （the difference between simulated value and observed value） and atmospheric transmission coefficient is quadratic in the three times. Therefore, we use the relationships to improve the radiation module. We found that the root mean square errors decreased 0.01,0.02, and 0.02 at three moments, respectively. Based on the modified model, the deviation between simulated and observed diffuse radiation fraction still exists. But, the results can provide reference and basis for improving ORYZA2000 model.（3） In CMIP5 model, the simulated changes in air temperature and CO₂ concentration in the future were consistent with the settings in RCP2.6, RCP4.5, RCP6.0 and RCP8.5. The diffuse radiation fraction in RCP6.0 increased during 2031-2060. In 2051-2060 it increased by 9.72%, compared with the value in 2015. Then, it began to fall. In RCP2.6, RCP4.5, and RCP8.5, diffuse radiation fraction value changes are similar from 2016 to 2100. By the end of this century, responding to the increase in temperature, rice yield decreased by 5.46%,16.1%,24.23%, and 37.63% in four scenarios, compared with it in 2015. The yield increased by2.97%, 14.7%,23.09%, and 32.61%, respectively with increases in CO₂ concentration in the four scenarios. The yield decreased by7.03%,7.14%,6.05%, and 7.03% with decreases in diffuse radiation fraction in the four scenarios by the end of this century.（4） When three factors change at the same time, the rice yield decline percentage in four scenarios were 10.04%,11.06%,11.28% and 10.04% respectively, compared with the yield in 2015. And, the yield declination in RCP6.0 was similar to that in RCP4.5. Fertilization effect of CO₂ concentration can not offset the decline in yield caused by rising temperatures and decreasing diffuse radiation fraction. The effects of temperature, CO₂ concentration, and the diffuse radiation fraction on rice yield do not equal the overlying simply the effects of the three single factors. Therefore, the most important factor impacting rice yield is temperature in the future. In addition, the diffuse radiation fraction is one of important factors, which cannot be ignored.