采用2017—2023年7—9月安徽砀山县梨树集中种植区域的A、B、C、D、E 共5个果园每日最高温度数据和砀山县国家基本气象观测站每日最高气温数据，通过一元一次线性回归方程，分别建立这5个果园酥梨果实二次膨大期日最高温度气象预报模型。结果表明，这5个果园最高温度预报模型分别为y =1.008 9x + 0.422 9，y = 0.998 3x + 0.649 1，y = 1.013 4x + 0.164 6，y =1.092 7x - 1.761 2，y = 1.000 4x + 0.363 4。模型检验R ²均大于0.9，表明模型预报效果好；5个果园模型预报结果与实况的平均绝对误差分别为0.5、0.5、0.6、0.8、0.5 ℃，均小于1 ℃，且比使用砀山县国家基本气象观测站日最高气温作为果园日最高温度的平均绝对误差更小，预报更精准；应用预报结果与实况的平均绝对误差分别为0.5、0.6、0.9、0.9、0.5 ℃，均小于1 ℃，预报精准。说明果园最高温度预报模型使用便捷，预报效果好，可应用于实际气象预报业务。

Based on the daily maximum temperature data of A, B, C, D and E orchards and the daily maximum temperature data of Dangshan National Basic Meteorological Observatory from July to September, 2017 to 2023, the daily maximum temperature meteorological forecast models of crisp pear fruit in A, B, C, D and E 5 orchards were established by linear regression equation. The results showed that the maximum temperature prediction models of these 5 orchards were y =1.008 9x + 0.422 9，y = 0.998 3x + 0.649 1，y = 1.013 4x + 0.164 6，y =1.092 7x - 1.761 2，y = 1.000 4x + 0.363 4. R² of model test was greater than 0.9, which indicated that the model had a good forecasting effect. The average absolute errors of the 5 township models were 0.5, 0.5, 0.6, 0.8 and 0.5 ℃ respectively. Both of them were less than 1 ℃, and the average absolute error was smaller and the prediction was more accurate than that of using the daily maximum temperature of Dangshan National Basic Meteorological Observatory as the orchard daily maximum temperature. The average absolute errors between the predicted results and the actual situation were 0.5, 0.6, 0.9, 0.9 and 0.5 ℃ respectively, all of which were less than 1 ℃, so the prediction was accurate. The forecasting model of orchard maximum temperature was convenient to use and had good forecasting effect, which can be applied to practical meteorological forecast business.