论文部分内容阅读
利用统计学方法,根据土壤相对湿度与气象因子、土壤相对湿度自身及环流指数的相关关系,建立了适合辽西干旱区春播期的土壤相对湿度动态预测模型。结果表明:(1)辽西地区4、5月土壤相对湿度与封冻雨和前期底墒相关性较大,且5月土壤相对湿度与同期降水呈显著正相关,与同期气温呈显著负相关;(2)土壤相对湿度与多项环流因子、海温因子及外强迫因子关系较为密切;(3)基于多元线性回归分别建立辽西地区4、5月份土壤相对湿度预报模型,2月末发布首次预报模型,3月末发布4月土壤相对湿度滚动订正模型,4月末发布5月土壤相对湿度滚动订正模型;(4)模型模拟结果表明,各月的拟合准确率均在80%以上,且订正前4、5月份平均相对误差分别为8.5%和5.6%,经模型订正后,平均相对误差分别达到6.6%和4.6%,模拟精度得到提高。
According to the correlation between relative humidity of soil and meteorological factors, the relative humidity of soil and the circulation index, a dynamic prediction model of soil relative humidity suitable for spring sowing period in western Liaoning is established. The results showed that: (1) There was a significant correlation between soil relative humidity in April and May in western Liaoning and frozen soil moisture and soil moisture in the previous period, and the relative humidity in May was significantly and positively correlated with the precipitation in the same period and negatively correlated with the temperature in the same period (3) Based on multivariate linear regression, the forecast models of soil relative humidity in April and May in western Liaoning were established respectively. The first forecast model was released in late February, and the third model At the end of April, the soil relative humidity rolling correction model was released in April, and the soil relative humidity rolling correction model in May was released at the end of April. (4) The model simulation results showed that the fitting accuracy of each month was above 80% The average relative errors of the months were 8.5% and 5.6% respectively. After the model was revised, the average relative errors reached 6.6% and 4.6% respectively, and the simulation accuracy was improved.