Statistical estimation of regional scale climate changes in frequencies of intense precipitation events

在这项研究中，我开发了一个新的短期降水量的概率分布函数。该函数是基于弱降水和强降水特征特征的差异构建的。弱降水和强降水分别用指数和威布尔概率分布函数表示。该函数主要有三个参数。以客观分析数据为预测因子，以三个参数为预测因子，建立了基于回归的统计模型。通过主成分分析对客观分析数据进行修正，使其具有独立性。利用该统计模型进行的试验表明，该模型对强降水频率下的气候变化有较好的估计效果。然而，在技术细节上还存在一些问题。

In this study, I developed a new probabilistic distribution function for short term precipitation amounts. The function was constructed basing on the difference of characteristic features between the weak and strong precipitations. The weak and strong precipitations were approximated by the exponential and Weibull probabilistic distribution functions, respectively. The function has mainly three parameters. A regression based statistical model was developed using objective analysis data as predictors and the three parameters as predictands. The objective analysis data was modified to be independent by the principal component analysis. A test experiment with the statistical model showed better result in the estimation of climate changes in intense precipitation frequencies. However, some problems in details of the techniques remain.