基于时空结构指标的中国融合降水资料质量评估

Analysis of the high-resolution merged precipitation products over China based on the temporal and spatial structure score indices

中文关键词:  资料评估  逐日融合降水  空间技巧评分  时间技巧评分  结构函数

英文关键词:Data evaluation  The daily merged precipitation  Spatial skill score  Temporal skill score  Structure function

中文摘要:

      引入空间、时间技巧评分以及结构函数3种指标,通过对比中国国家气象信息中心研制的逐日融合降水资料和美国国家海洋大气局(NOAA)气候预测中心卫星反演降水资料(CMORPH)、热带测雨卫星反演降水资料(TRMM)在中国区域的适用性和误差分布,着重考察融合降水资料的质量。结果表明,中国区域平均的融合降水资料时空精度远高于CMORPH和TRMM卫星降水资料,且融合资料和卫星资料在夏季的质量优于冬季；在中国东南区域的模拟精度普遍好于西部地区,融合降水质量最高的两个区域为江淮和华南,较差的区域则在青藏高原和西北地区。融合后降水资料比融合前CMORPH卫星降水在空间及时间技巧评分均有较大提高,其提高幅度冬季大于夏季。通过计算结构函数,发现在中国江淮、华南、华北和东北等地区,随着网格区域内任意两点距离的增大,融合产品与观测降水的结构函数曲线始终十分接近。在西北、青藏高原等区域,融合产品与观测降水的结构函数则偏离较大。西南地区地形复杂,卫星资料无法精确反映实际降水情况,高密度观测资料尤为重要。江淮、华南、华北地区的融合降水结构函数曲线增长率大于东北,也从侧面反映江淮、华南、华北地区降水分布的非均一性比东北强,降水可能受中小尺度天气系统影响较大。

英文摘要:

      Based on the three indices of spatial and temporal skill score and structure function, the investigation into the applicability of the daily merged precipitation, and the two datasets of precipitation respectively retrieved from the satellites of CMORPH and TRMM in China are performed. The spatial-temporal distributions of errors of these precipitation data are also analyzed with the focus on the quality of the merged precipitation data. The results indicate that, for the whole area of China, the spatial and temporal precision of the merged precipitation is observably promoted compared with that of both the CMORPH and TRMM precipitation. For the merged and satellite data, their performances on simulating the spatial and temporal distribution in summer are all superior to in winter. The simulation qualities in the southeastern region of China are generally better than in the northwestern region. The Yangtze-Huaihe River Valley and South China are the two regions showing the best quality of the merged precipitation. However, the Tibet Plateau and northwest China have the poorest effect of the improvement. Another significant feature after precipitation merging is that both spatial skill scores (S_S) and temporal skill scores (S_T) are improved obviously. Compared to summer, the values of S_S and S_T are raised more remarkably in winter. By comparing the structure function values of the three sets of evaluation data and observation data in the different regions, we can clearly see that, in the Yangtze-Huaihe River Valley, South China, North China and northeast China, with the increase of the distance between any two points within the regions, the structure function curves of the observed and merged precipitation are always very close. In northwest China and the Tibet Plateau, the structure function of the merged precipitation has large deviation between the merged and observed precipitation. In southwest China, due to the complex terrain, the satellite data are unable to give the accurate simulation of actual precipitation. So the observed data plays a more important role in this region. In the Yangtze-Huaihe River Valley, South China, and North China, the growth rates of structure function curves of the merged precipitation are larger than that in the northeast China. It reflects that the heterogeneities of precipitation distribution in the three regions of the Yangtze-Huaihe River Valley, South China, and North China, in which the precipitation is more influenced by the mesoscale weather system than that in northeast China.