四种卫星降水数据在黄河源区的适用性分析
2021-03-28张磊磊康颖岳青华
张磊磊 康颖 岳青华



摘 要:基于黄河源区17个气象站1998—2015年逐日降水量资料,分析多源卫星降水数据(TMPA、RT、CMORPH和PERSIANN)在該地区的适用性,并在此基础上评估TMPA数据在水文模拟应用中的潜力。结果表明:PERSIANN数据精度最低,其次为CMORPH和RT,TMPA数据精度最高,其中PERSIANN和RT数据在该地区存在高估降水量的问题。利用气象站插值降水量和唐乃亥水文站1998—2013年逐月径流资料对分布式水文模型(VIC模型)进行参数率定,在此基础上采用精度最高的TMPA数据驱动VIC模型,进一步验证了TMPA在黄河源区分布式水文模拟应用中具有较高潜力,TMPA模拟的径流过程与基于地面气象站插值降水量模拟的径流过程精度基本相当。
关键词:TMPA;RT;CMORPH;PERSIANN;黄河源区
Abstract:The applicability of various remote sensing satellite precipitation data (TMPA, RT, CMORPH and PERSIANN) was estimated based on the daily precipitation data of 17 rainfall stations in the period of 1998-2015 for the source region of the Yellow River. The result shows that the data precision of TMPA is the highest, then is followed by CMORPH, RT and PERSIANN. Among the four datasets, PERSIANN and RT have a problem of overestimating precipitation in this area. Based on the measured rainfall and the observed streamflow of Tangnaihai Hydrometric Station from 1998 to 2013 and interpolated precipitation of weather station, the parameters of distributed hydrological model (VIC model) were determined. Then, the performance of the TMPA data was further verified. The result shows that TMPA has high potentiality for hydrological model. Based on the distributed hydrological model, the performance of the TMPA data is equivalent to that by using actual precipitation.
Key words: TMPA; RT; CMORPH; PERSIANN; source region of Yellow River
青藏高原地区地处我国西南部,是长江、黄河、澜沧江、怒江、雅鲁藏布江等的发源地,被称为“亚洲水塔”[1]。受地形、气候等条件的影响,青藏高原气象站点稀疏且分布极不均匀,这给青藏高原的水资源研究带来了极大挑战,而流域水文模型可通过有限实测降水数据反演径流过程。
在流域水文模拟中,降水是影响模型模拟精度的重要因素之一,模型的空间降水量输入主要是根据气象站实测雨量通过某一插值方法确定,但雨量站点的设立往往受地形、气候、经济、社会等方面的影响,观测到的雨量信息并不能完全代表流域降水特征。随着科技的飞速进步及计算机软硬件的发展,通过卫星探测云层信息反演的遥感卫星降水数据逐步被应用[2],相对于单点观测数据,卫星降水数据具有时间和空间分辨率高、覆盖面积广的特点,虽然卫星降水数据是利用反演方式推求的,与实际降水仍存在一定的偏差,但其对缺乏资料流域的水文过程模拟与验证仍然具有十分重要的意义。……
