气象因素与杨干象虫口密度的多元线性回归和判别分析  

Multiple Linear Regression and Discriminant Analysis of Meteorological Factors and Pest Population Density of Cryptorrhynchus lapathi

作　　者：张福丽[1] 景天忠[2] 王志英[2] 谢淑萍[3] 张晖 

机构地区：[1]周口师范学院生命科学系植物遗传与分子育种重点实验室,河南周口466001 [2]东北林业大学,黑龙江哈尔滨150040 [3]海林林业局,黑龙江海林157100 [4]黑龙江省林业厅,黑龙江哈尔滨150001

出　　处：《安徽农业科学》2011年第15期9000-9001,共2页Journal of Anhui Agricultural Sciences

基　　金：国家林业公益性行业科研专项(200804023)资助

摘　　要：[目的]研究气象因素对杨干象发生的影响。[方法]利用多元线性回归和判别分析的方法分析了1～12月份月极低温度、月极高温度、月平均温度、月降水量等对杨干象发生的影响。[结果]多元线性回归后,建立了回归模型:y=0.781+0.021x16+0.004x26+0.058x20+0.015x23+0.055x34,模型的预测精度达到了利用模型进行预测的基本要求。通过对进入多元线性回归模型中的自变量和因变量的关系进行判别分析,进一步确定了影响杨干象虫口密度的重要气象因子,分别为4月极低温度(x34)、2月极高温度(x23)、1月平均温度(x16)、3月平均温度(x26)和1月降水量(x20)。[结论]4月极低温度(x34)、2月极高温度(x23)、1月平均温度(x16)、3月平均温度(x26)和1月降水量(x20)是影响杨干象发生的重要气象因子。[Objective] The aim was to study the influence of the meteorological factors on population density of Cryptorrhynchus lapathi.[Method] The influence of the lowest temperature,the highest temperature,the mean temperature and the precipitation from January to December on population density of Cryptorrhynchus lapathi was studied by multiple linear regression and discriminant analysis.[Result] Regression model was established,which was y=0.781＋0.021x16＋0.004x26＋0.058x20＋0.015x23＋0.055x34.Prediction accuracy of the model achieved to the basic requirements.The relationship of independent variables of multiple linear regression model and dependent variable was analyzed by discriminant analysis.The important meteorological factors impacting the population density of Cryptorhynchus lapathi were further determined.They were the lowest temperature of April（x34）,the highest temperature of February（x23）,the mean temperature of January（x16）,the mean temperature of March（x26） and the precipitation of January（x20）.[Conclusion] The important meteorological factors affecting Cryptorhynchus lapathi population density were the lowest temperature of April（x34）,the highest temperature of February（x23）,the mean temperature of January（x16）,the mean temperature of March（x26） and the precipitation of January（x20）.

关 键 词：气象因素 杨干象 虫口密度 多元线性回归 判别分析 

分 类 号：S763.38[农业科学—森林保护学]