Study On Modeling Effects Of Genotype And Sowing Date On Rice Prductivity

It is of great significance to increase rice yield to ensure our national food security and promote sustainable development of agriculture in the future. In this study, RiceGrow, a rice simulation model, was used to explore the effects of different genotypes and sowing dates on rice production and then increase the rice yields.RiceGrow model was first tested and evaluated using different eco-sites field experiment data, involving different rice cultivars, different years, different sowing dates. Then with the well evaluated model, RiceGrow, this study aimed to investigate the effects of different phenological parameters in RiceGrow on rice growth days and yield through designing different phenological parameters changes at five eco-sites (Wuchang, Xuzhou, Xinghua, Xinyang and Hefei, China) sowing one-season rice cultivar. Additionally, using long-term weather data of Hangzhou, Yizheng and Xuzhou from 1981 to 2010, this study explored different sowing date's effect on rice production.RiceGrow was calibrated and validated by the field experiments involving different varieties, different years and different treatments at Wuchang, Xuzhou, Xinghua, Xinyang and Hefei. The RMSE (Root Mean Square Error) between the simulated values and observed values of Heading, Maturity and grain yield were:2.6 d,3.1 d and 869.7 kg ha-1, NRMSE (Normalized Root Mean Square Error) were:2.3%,2.1% and 11.5%. Simultaneously, RiceGrow was compared with a tropical rice model ORYZA2000 which was widely used in other regions. The results showed that RiceGrow model could simulate well in the yield change of different sowing dates in Hangzhou, Yizheng and Xuzhou, RMSE of growth days, aboveground biomass and yield were 4.6 d,1539.2 kg ha-1 and 792.2 kg ha-1, respectively; NRMSE were 3.7%,17.4% and 10.2%, respectively. There were both good agreement between trends in measured and simulated rice growth days, aboveground biomass and yields of different sowing dates simulated by RiceGrow and ORYZA2000. In short, RiceGrow showed a quite good performance both in predicting rice growth days and yield in different agro-ecological conditions and sowing dates. Through designing different genetic parameters related to rice phenology, this study analyzed effects of phenological parameters on rice productivity. Simulation experiments were conducted to quantify and analyze the impacts of single parameter and the combination of IE (Intrinsic earliness), PS (Photoperiod sensitivity), TS (Temperature sensitivity) and BFF (Basic filling factor) by setting the value range and step among the four different parameters related to rice phenology in RiceGrow. The results showed that increasing IE and BFF could remarkably decrease rice growth days, but increasing TS and PS had a contrary trend; the yield increased with the PS and TS increasing in a certain extent, but decreased with the BFF increasing, without remarkably change with the IE changing. And the combination of IE, PS and TS would have great significantly differences in rice growth days and yields. The results revealed that with the growth days increasing, the rice yields would increase simultaneously at all five eco-sites; up to a certain length, the yield did not chang significantly or declined first and then increased. And there existed well comibination of different phenological parameters to obtain high yields with short growth duration compared with check varities in each eco-site in different year types, and it revolved IE, PS and TS ranging from 0.5 to 0.8,0.020 to 0.036 and 6.6 to 9.0, respectively. With yield increasing more than 5% contrasting with check cultivars, there would be different changes in different phenological parameters in different year types. Compared with the cultivars sown before in each site, the predicted maturity days shortened average 6.0 d, whereas the yields of high-yielding cultivars in each site ranged from 6701 to 10262 kg ha-1, increased by 17.5%.With a long-term historical weather data from 1981 to 2010 in Hangzhou, Yizheng and Xuzhou, this study also explored different sowing date's effects on rice productivity. The simulation of sowing date experiments showed that different rice cultivars (Liangyoupeijiu, Wuxiangjing 14 and Shanyou 63) could be sown in a wide sowing range, and with the temperature rising, there would be a wider sowing window in the future. The simulation results also provided the optimum sowing dates with highest yield of different rice cultivars in different regions mostly ranging from April 19 to May 3 in Hangzhou, April 10 to May 10 in Yizheng, and April 24 to May 14 in Xuzhou. The highest yield obtained sowing on optimum dates in Hangzhou, Yizheng and Xuzhou, varied from 8173 to 11656,8912 to 11379, and 8283 to 10816 kg ha-1, respectively. And the suitable sowing date with a high probability resulting high yields of rice were between April 24 and May 7 in Hangzhou, April 17 and May 2 in Yizheng, and April 18 and May 21 in Xuzhou. Compared with the yields of traditional sowing dates in different eco-sites, yields of different rice types sowing on optimum dates increased average by 13.7%. This study might provide a method using crop model to determine the suitable sowing dates of a new cultivar in different regions.