采用判别分析法,根据浙江省富阳市1984～1995年12年的气象资料,经6个预报因子建立蘑菇害螨发生程度预测模型。经比较,筛选出上年7～8月份温湿系数,当年7～8月总雨日及温度指数3个因子,其组成的判别方程历史拟合率达917%。经两年的预测验证,结果与实际相吻合。 According to the meteorological data from 1984 to 1995 in Fuyang City, Zhejiang Province, discriminant analysis was used to establish a prediction model of the degree of occurrence of pest mites through six predictors. After comparison, the factors of temperature and humidity from July to August in the previous year were screened, and the total rainy days and temperature index from July to August of that year were three factors. The historical fitting rate of the discriminant equation of the composition was 917%. After two years of predictions, the results are in line with the actual situation.