安徽省花生产量与气象因素的关联度分析及预测模型研究

为分析气象因子对安徽花生产量的影响,构建适用于本地的花生单位产量预测模型,以期为探讨花生经济效益、应对气象灾害风险管理提供参考。对安徽各市2000—2017年花生气象产量和气象因子进行灰色关联度分析,筛选出关联度较大的气象因子,并采用逐步回归法建立产量预测模型。结果表明：安徽花生产量与生育期气象因子关联度5月份平均气温>7月份光照时数>5月份光照时数>6月份光照时数>7月份平均气温>8月份光照时数>8月份平均气温>6月份平均气温>8月份降水量>7月份降水量>5月份降水量>6月份降水量,应用建立的花生单位产量预测模型对历年产量进行回测,结果显示预测值与实际值均方根误差为815 kg/hm²、拟合指数为0.81,总体较好,具有一定的应用价值。

Peanut Yield in Anhui: Correlation with Meteorological Factors and Forecast Model

To construct a local peanut yield forecasting model based on the influence of meteorological factors in Anhui, and provide reference for exploring the economic benefits of peanut and coping with the risk of meteorological disasters, grey correlation analysis on peanut meteorological output and meteorological factors from 2000 to 2017 of all cities in Anhui was conducted. The meteorological factors with greater correlation were screened out and a yield prediction model based on stepwise regression was established. The results showed that the correlation between Anhui peanut production and meteorological factors during growth period was followed the order of average temperature in May > light hours in July > light hours in May > light hours in June > the average temperature in July > light hours in August > the average temperature in August > the average temperature in June > precipitation in August > precipitation in July > precipitation in May > precipitation in June. Furthermore, peanut production over the years was tested based on the peanut per unit yield prediction model, revealing that the root mean square error between the predicted value and actual value was 815 kg/hm² and the fitting index was 0.81. The prediction model is proved to have certain application value.