Stability Evaluation Of Different Accumulated Temperature Model

Spring maize is one of the main crops in northeast China, and its yield prediction and growth period prediction are taken seriously by the scientific research and business. Accumulated temperature as heat index reflecting crop growth and heat resources, is very important in the composition of crop varieties, planting system, introduction, breeding, growth period prediction and prevention and control of plant diseases and insect pests. But in actual production, the environmental conditions often cannot meet the needs of crops. Due to the comprehensive effect of the meteorological factors such as the temperature, illumination and precipitation, accumulated temperature is different in crop varieties, planting area, planting season and year. The stability of accumulated temperature should be fully considered in crop varieties, regions and years., when accumulated temperature method is used to analyze the crop growth rate. Different models and different calculation method have different influence to the stability of the accumulated temperature.Based on growth data and ground observation data in the same period of spring maize varieties called ‘danyu 13’ ‘longdan 13’ ‘dongnong 248’ and ‘sidan 19’ in northeast China, three kinds of nonlinear accumulated temperature model respectively put forward by Shen Guoquan, Gao Liangzhi and Yin Xinyou are fitted to obtain the model parameters. Precision and accuracy are got by the comparision of the simulation results on three kinds of accumulated temperature model, and the contrast and stability evaluation of the effective accumulated temperature calculated by three kinds of nonlinear model and commonly used method are made.RMSE is the important standard of analyzing the fitting equation with different varieties, different sites, different growth period. It is found that the fitting results are different in the different growth stage even if on the same site in the fitting of the same variety. There are differences on the same accumulated temperature model in the process of fitting the same growth stages because of different varieties. It is similar for the superiority and quality of fitting the same accumulated temperature model with the same variety in different sites.On the analysis of variation coefficient and relatively extreme value of the effective accumulated temperature calculated through four methods, stability of three kinds of nonlinear accumulated temperature model is superior to the commonly used method in the whole growth period. In emergence to jointing stage, four methods are consistent on the stability of the “Danyu 13” varieties, however, commonly used method is slightly inferior to the accumulation of three kinds of nonlinear model of the accumulated temperature in “Longdan 13”, “Dongnong 248”and “sidan 19”three varieties. The worst stage of stability is the time from jointing to heading stage, no matter what breed. In the heading to mature stage, the stability of the four methods is similar. The results of the study can provide theoretical basis and technical support for the northeast spring maize development and yield prediction work.