包含外强迫因子的平潭旬降水量预测

降水量是重要的预报要素之一,长期的降水预测更是能提前预测旱涝分布情况,为国民经济规划提供依据。但目前为止,长期的降水预测仍缺少客观的预报方法。为此,尝试利用非线性预测模型来预测旬降水量,并将该模型应用于福建平潭,分别用与原始数据的差值、与原始数据的相关系数、均方根误差,以及符号显著性检验方法,讨论了包含外强迫因子的平稳性模型与不包含外强迫因子的非线性模型的预测能力,结果表明:包含外强迫因子的模型第一步预测结果与原始观测数据的相关系数为0.73,不包含外强迫因子的模型第一步预测结果与原始观测数据的相关系数则为0.47。无论是从与原始数据的差值及相关系数,还是均方根误差等方面,外强迫模型都是优于平稳性模型,并且通过符号检验方法可看出两种模型存在差异性,这也说明加入外强迫因子可以有效地提高预测技巧,外强迫因子与状态变量在预测中扮演同等重要的角色。

Prediction of Ten-days Precipitation in Pingtan Containing the Effect of the External Forcing Factors

Precipitation is one of the important forecast elements,and the long term precipitation prediction can forecast the distribution of drought and flood,providing the basis for national economic planning.But there are still less objective forecast method of long term precipitation prediction at present.Therefore,in this study,a predictive technique incorporating external forcing factors was used to predict the ten days precipitation of Pingtan.The prediction performance of the stationary model incorporating external forcing factors and the nonlinear prediction model not including external forcing factors were discussed by the difference value,correlation coefficient,root mean square error and the sign significance test.The results showed that: the correlation coefficient of the predication performance of the stationary model which incorporate external forcing factors was 0.73,while the correlation coefficient of the nonlinear prediction model not including external forcing factors was 0.47.The stationary model are better than the nonlinear prediction model according to the difference value,correlation coefficient and root mean square error,and according to the sign test,there are differences between the two models.This also explained that incorporating external forcing factors can effectively improve the forecasting performance and the external forcing factors played the same important role as the state variables in the prediction.