This study utilized a spatio-temporal framework to assess the dispersion and uncertainty of precipitation in Iran. Thirty-one rain gauges with data from 1960 to 2010 were selected in order to apply the entropy concept and study spatio-temporal variability of precipitation. The variability of monthly, seasonal and annual precipitation series was studied using the marginal disorder index (MDI). To investigate the intra-annual and decadal distribution of monthly and annual precipitation values, the apportionment disorder index (ADI) and decadal ADI (DADI) were applied to the time series. The continuous wavelet transform was used to decompose the ADI time series into time-frequency domains. The decomposition of the ADI series into different zones helped to identify the dominant modes of variability and the variation of those modes over time. The results revealed the high disorderliness in the amount of precipitation for different temporal scales based on disorder indices. Based on the DI outcome for all rain gauges, a self-organizing map (SOM) was trained to find the optimum number of clusters (seven) of rain gauges. It was observed from the clustering that there was hydrologic similarity in the clusters apart from the geographic neighborhood.

这项研究利用时空框架评估了伊朗降水的分散性和不确定性。为了应用熵的概念并研究降水的时空变化，选择了31个具有1960年至2010年数据的雨量计。使用边际无序指数（MDI）研究了月度，季节和年度降水量序列的变异性。为了调查月降水量和年降水量的年内和年代际分布，将分配无序指数（ADI）和年代际ADI（DADI）应用于时间序列。连续小波变换用于将ADI时间序列分解为时频域。ADI系列分解为不同区域有助于确定主要的变异性模式以及这些模式随时间的变化。结果表明，基于无序指数，不同时间尺度上降水量的高度无序性。根据所有雨量计的DI结果，对自组织图（SOM）进行了训练，以找到雨量计的最佳簇数（七个）。从聚类中观察到，除地理邻域外，聚类中都存在水文相似性。