五种降水产品在疏勒河上游山区和中下游月尺度降水的适用性对比研究

降水产品为理解降水时空分布提供了新的视角，然而其在不同地区适用性差异很大，目前对不同降水产品在西北内陆河流域上游山区和中下游平原适用性的对比研究还十分有限。利用疏勒河流域及周边9个国家气象站2001-2013年和4个山区自建气象站2008-2013年降水观测数据，基于皮尔逊相关系数（R）、均方根误差（RMSE）、平均绝对误差（MAE）和相对偏差（BIAS）指标，对CMFD、ERA-Interim、GPCC、GPCP V2.3和JRA-55共五种常用降水产品的月尺度降水数据在疏勒河上游山区和中下游平原区的适用性进行了评估对比，结果表明：五种降水产品都可以较好地模拟疏勒河流域多年平均降水的空间分布，总体来说，降水产品在上游的评估的相对误差优于中下游，但绝对误差比中下游大，这与上游降水远大于中下游有关。各产品在上游和中下游的精度排序结果一致，均为CMFD > GPCC > ERA-Interim > GPCP V2.3 > JRA-55。CMFD效果最佳可能与融合了中国更多的地面降水资料有关，其在自建站的效果也表现最佳。不同产品在疏勒河上游山区和中下游月尺度的降水评估中表现较好，但尚不能满足流域水文过程分析和模拟的需要。

Evaluation on the performance of five precipitation datasets of monthly precipitation in the upper reaches,middle and lower reaches of Shule River basin

Multiple global precipitation datasets provide a new perspective for understanding the temporal and spatial distribution of precipitation. However, their applicability varies widely in different regions. At present, the study on the performance of different datasets in mountainous areas of upper reaches and plains in the middle and lower reaches is still very limited. The observed precipitation data from 9 national meteorological stations from 2001 to 2013 and 4 self-built meteorological stations from 2008 to 2013 were used in this study, and Pearson Correlation Coefficient (R), Root Mean Squared Error (RMSE), Mean Absolute Deviation (MAE) and Relative Deviation (BIAS), were used to evaluate the performance of five precipitation dataset monthly precipitation including CMFD, ERA-Interim, GPCC, GPCP V2.3 and JRA-55 in the upper reaches, middle and lower reaches of Shule River basin, respectively. The results suggested that the five precipitation datasets could better reproduce the spatial pattern of yearly average annual precipitation in the Shule River basin. The performance of precipitation dataset in the upper reaches are higher than that in the middle and lower reaches. The relative performance of five precipitation datasets are same in the upper reaches, middle and lower reaches, both are CMFD > GPCC > ERA-Interim > GPCP V2.3 > JRA-55. The best performance of CMFD may be related to a variety of factors. The evaluation on the performance of different precipitation dataset in the Shule River basin provides an important clue for hydrological process simulation.