盐城市小麦产量与气象因子的相关性分析及产量预测模型建立

Correlation Analysis of Wheat Yield with Meteorological Factors in Yancheng City and Establishment of Yield Prediction Models

摘要: 利用2010~2024年盐城市小麦产量数据和气象数据，通过指数平滑法计算得出小麦逐年的趋势产量和气象产量，接着对气象产量与气象因素进行相关分析从而得到关键气象因子，在此基础上构建小麦产量预测模型并验证。结果表明：(1) 利用指数平滑系数法建立小麦产量预测模型，当平滑系数为0.4时，气象因子与小麦产量的相关系数最高同时标准误差最低，预测模型拟合效果最优；(2) 盐城市小麦气象产量与当地气象因素有一定的相关性。其中，11月份的月日照时数和1月份的月均最高气温与小麦气象产量的相关性最高；(3) 小麦预测产量可以较好模拟小麦的逐年变化趋势，除2012年外，小麦预测产量与实际产量的相对误差在±3%以内，可作为预测盐城市小麦产量的参考。

Abstract: Using wheat yield and meteorological data from Yancheng between 2010 and 2024, this study employed exponential smoothing to calculate annual trend yields and meteorological yields. Correlation analysis between meteorological yields and meteorological factors identified key climatic variables, which were then used to construct and validate a wheat yield prediction model. Results indicate: (1) The exponential smoothing coefficient method established the optimal wheat yield prediction model. With a smoothing coefficient of 0.4, meteorological factors exhibited the highest correlation coefficient with wheat yield and the lowest standard errors, yielding the best model fit. (2) Meteorological yield of wheat in Yancheng showed significant correlations with local meteorological factors. Among these, monthly sunshine duration in November and monthly average maximum temperature in January demonstrated the strongest correlations with meteorological yield. (3) The predicted wheat yield effectively simulates the annual variation trend. Except for 2012, the relative errors between predicted and actual yields remained within ±3%, making it a reliable reference for forecasting wheat production in Yancheng.