茶树干燥型大冻害气象因素的logistic回归分析

Logistic Regression Analysis of Meteorological Factors of Dry Heavy Freezing Damage of Tea Plant

茶树大冻害是制约北方茶叶发展的一个重要因素。为解决这一问题,从山东主茶区的日照、青岛、临沂3个地区42年、共计126年的气象资料中,选择可能导致茶树干燥型大冻害形成的30个气象因子作为自变量,以干燥型大冻害作为因变量,进行单因素logistic回归分析。根据单因素分析的结果,从30个气象因子中筛选出10个有统计学意义的气象因子进行多因素logistic回归分析,建立茶树干燥型大冻害发生的logistic回归模型,并对模型进行评价。结果显示,1月平均气温、上年8月平均气温、上年11月空气相对湿度3个气象因子,是茶树干燥型大冻害发生的主要决定因素。研究结论为山东以及其他北方茶区有效预防茶树干燥型大冻害提供了重要依据。

Heavy freezing injury is an important factor restricting the development of tea industry in northernChina. To solve this problem, the author studied the meteorological data of 42 years in Qingdao, Rizhao, Linyi（which are major tea producing areas in Shandong Province）, and chose 30 meteorological factors as dependentvariable, chose the dry heavy freezing injury as independent variable and made single factor logistic regressionanalysis. According to the result of single factor logistic regression analysis, the author chose 10 meteorologicalfactors with statistical significance to make multivariate logistic regression analysis, established the logisticmodel of tea plant dry heavy freezing injury. By evaluating the model, the author found out that the average airtemperature in January, the average air temperature in last August and the relative air humidity in lastNovember were the main determinants that caused the tea plant dry heavy freezing injury. Conclusions wouldprovide a basis for effectively preventing dry heavy freeze injury of tea plant in Shandong and other northerntea areas.