利用气象因子构建东港市地区水稻二化螟发生等级的预测模型

The Prediction Model of Rice Chilosuppressalis Occurrence Grade in Donggang City was Established by Meteorological Factors

为建立辽宁省东港市地区水稻二化螟发生等级的预测模型,以支持该地区水稻二化螟的科学精准防治。收集整理2012年至2021年东港市地区的历史气象资料以及水稻二化螟的历年虫情等级,分析水稻二化螟在田间的发生动态,并利用DPS 18.1软件中的逐步回归分析法和通径分析法构建并筛选有效的预测模型。水稻二化螟成虫历年在田间的发生高峰期主要集中在6月中上旬(6月1日~6月20日)和8月中上旬(7月31日~8月19日)。利用逐步回归分析法筛选出6项关键气象因子:6月份降水量,8月份降水量,6月份平均湿度,8月份平均湿度,4月份平均温度以及8月份平均温度,并由此构建了该地区水稻二化螟发生等级的预测模型:Y=5.1287+0.1468X_(2)-0.6558X_(6)+0.9123X_(10)+0.4521X_(12)-0.9530X_(16)-0.8540X_(18)(F=419.54,P=0.0203),经过对2012年至2019年相关数据的回测以及对2020年与2021年水稻二化螟发生等级的预测,显示该预测模型的拟合度与预测精准度均较好;通过通径分析,可知6月份降水量与8月份降水量是影响水稻二化螟发生等级的最重要的气象因子。本研究通过逐步回归分析法所构建的水稻二化螟发生等级预测模型能够为辽宁省东港市地区科学防治水稻二化螟提供有效的支持。

To establish a prediction model for the occurrence grade of rice Chilosuppressalis in Donggang City,Liaoning Province,and to support the scientific and accurate control of rice Chilosuppressalis in this area.The historical meteorological data of Donggang city from 2012 to 2021,as well as the historical epidemic grade of rice Chilosuppressalis,were collected to analyze the occurrence dynamics of rice Chilosuppressalis in the field,and the effective prediction model was established and screened by stepwise regression analysis and path analysis in DPS 18.1 software.Tworice Chilosuppressalis adult calendar year during the outbreak of the field are mainly concentrated in early 6 months(June 1-June 20)and early mid-august(July 31-August 19).Six key meteorological factors were screened out by stepwise regression analysis:precipitation in June,precipitation in August,average humidity in June,average humidity in August,average temperature in April and average temperature in August,and the prediction model of occurrence grade of rice Chilosuppressalis in this region was established:Y=5.1287+0.1468X_(2)-0.6558X_(6)+0.9123X_(10)+0.4521X_(12)-0.9530X_(16)-0.8540X_(18)(F=419.54,P=0.0203),Based on the data from 2012 to 2019 and the prediction of the occurrence grade of rice Chilosuppressalis in 2020 and 2021,the fitting degree and accuracy of the prediction model are good.Through path analysis,precipitation in June and August were the most important meteorological factors affecting the occurrence grade of Chilosuppressalis.The prediction model of rice Chilosuppressalis occurrence grade established by stepwise regression analysis can provide effective support for scientific control of rice Chilosuppressalis occurrence grade in Donggang City,Liaoning Province under the background of rice Chilosuppressalis occurrence level becoming more and more serious.