玉米倒伏胁迫影响因子的空间回归分析

为指导玉米新品种的推广，采用回归模型分析玉米主产区倒伏胁迫空间分布成因。该文用多元逐步线性回归法筛选黄淮海夏播玉米区的倒伏胁迫的决定因素，比较普通最小二乘法线性回归模型和地理加权回归模型的结果，以确定倒伏胁迫及其决定因素是否存在空间非平稳性和空间依赖性。结果表明：在探索倒伏的空间异质性时，地理加权回归模型显著优于普通最小二乘法线性回归模型；日降水量是玉米倒伏胁迫的主要环境成因，且倒伏程度随日降水量增加而加重；土壤含氮量、留苗密度和日平均风速与倒伏的关系随空间位置而发生正负向变化，因地制宜的分析倒伏成因才能客观有效的指导农民种植生产。

Spatial regression analysis on influence factors of maize lodging stress

A regression model was applied to analyze the reason causing the spatial distribution of maize lodging in the main maize-growing areas, in order to guide the promotion of new maize varieties. Multivariate stepwise regression method was adopted in this study to select the decisive factors of maize lodging in the Huang-Huai-Hai summer maize area. The aim was to figure out whether there was spatial nonstationarity and spatial dependence between the lodging stress and its relative determinants by comparing with the analysis results from the ordinary least squares linear regression model and geographically weighted regression model. The results demonstrated that geographically weighted regression model was better than the ordinary least squares linear regression model when analyzing the special heterogeneity of maize lodging; Daily precipitation was the main environmental determinant of maize lodging stress and it had a positive influence on maize lodging stress; The relationships of soil nitrogen content, planting density and daily average wind speed changed with space position positively and negatively; Therefore analyzing the causes of maize lodging stress according to the local conditions is necessary for providing an objective and effective guidance of plant production to farmers.