Study On Suitable Sowing Date Of Maize Based On Crop Growth Models

Maize sowing date is one of the main factors affecting yield of maize.Different stages of growth and development of maize can be directly influenced by different sowing time.This study takes advantage of crop growth model to explore suitable sowing periods of spring maize in Guiyang area,which makes the environment be as good as possible for maize growth in stages,so as to effectively improve the yield level.The idea of this study is to get the historical production situation in the three decades by simulating the maize production using two crop growth models,DSSAT and Hybrid-Maize,respectively,and to get the distribution of the highest yields in three patterns of weather: rainfall,accumulated temperature and solar radiation,so the optimal sowing dates of the years of different weather patterns are determined.?.The calibration results of the models are:In the simulation of phenological periods and grain yield,the average RAE(Relative Absolute Error)and RRMSE(Relative Root Mean Square Error)of DSSAT and Hybrid-Maize are controlled within 20%,so the two models can be applied to simulating grain yield in this study.?.The following main conclusions are obtained by simulating the sowing date experiment with the calibrated models:1.The suitable sowing dates obtained through DSSAT simulation:For drought-prone years,years with normal accumulated temperature,or years with normal solar radiation,it is suitable to sow late,in late mid-April;for years inclined to flood or years with weaker solar radiation,suitable in mid-April;for other years,in early April,and even can be advanced to around March 20 for years with stronger solar radiation.2.The suitable sowing dates obtained through Hybrid-Maize simulation:For years with weaker solar radiation,it is suitable to sow in mid-April;for years inclined to drought or years with stronger solar radiation,suitable sowing time can be around March 20 and also can be in late mid-April for years with stronger solar radiation;for other years,in late mid-April.In addition,the suitable sowing dates for the best average annual yield are around mid-April.?.Holt exponential smoothing method is used to perform daily weather forecasting and combined with crop model to predict the appropriate sowing date for spring maize in the next year.The result is: comparing the yields obtained under the predicted meteorological data and the actual meteorological data,it is found that the agreement degree of the sowing dates corresponding to the high yields respectively is over 70%.In addition,the RAE mean value and the RRMSE mean value of the predicted yield itself are controlled within 20%,and the distribution of RAE and RRMSE values are also concentrated in 0-19%.It shows that this forecasting method has good practical application value.