%0 Journal Article
%T 卫星降水在怒江流域的精度及径流模拟评估<br>Evaluation of Accuracy and Streamflow Simulation of Satellite Precipitation on Nu Basin
%A 周星
%A 倪广恒
%A 仲志余
%A 郭生练
%J Journal of Water Resources Research
%P 125-135
%@ 2166-5982
%D 2019
%I Hans Publishing
%R 10.12677/JWRR.2019.82015
%X <div style=\"text-align:justify;\">
	<span style=\"font-size:14px;line-height:1.5;\">高时空分辨率的卫星降水产品弥补了山区流域地面雨量站点布设密度低、空间结构不合理的局限性，在水文研究及应用上有很大的发展前景，但是其精度是制约其应用的瓶颈，也是研究的热点。本文以地面雨量站点观测降水数据为参照，评估不同尺度下常用三种降水产品TRMM 3b42 V7(简称TRMM)、TRMM 3b42RT V7(简称RT)以及CMORPH(简称CMO)，在典型的少资料山区流域——怒江流域(云南段)的精度，并进一步评估了不同降水产品驱动水文模型(Hydro-Mount)模拟日径流的效果。研究表明：站点尺度上三种卫星降水产品与站点数据相关性较好，相关系数在0.65以上，降水正确预报率在0.71以上，但CMO存在超过40%的低估降水，TRMM存在0.42%的低估，RT存在9.37%的低估降水；流域尺度上卫星降水的精度明显优于点尺度上，精度趋势和站点尺度一致；TRMM数据和RT数据日径流模拟效果较好，纳西效率系数(NSE)指标上与雨量站点数据相当，但表现较雨量站点稳定，CMO数据日径流模拟效果较差，但在消除一定系统偏差后，也表现出一定的模拟精度。本研究为TRMM数据的精度订正提供相关思路，并分析了其在径流模拟应用中的可行性。本研究可为卫星降水数据的精度修正以及区域水文应用提供参考借鉴。<br />
High temporal-spatial resolution satellite based precipitation products has been extensively used in ungauged or sparsely gauged areas to bridge the gap between the need for precipitation estimates and the scarcity in gauge observations. The accuracy of satellite precipitation products which restricts its application has become a focus. In this study, three satellite precipitation products (i.e., TRMM 3B42 V7 short for TRMM, TRMM 3B42 RT V7 short for RT, CMORPH short for CMO) were evaluated at different spatial scales via comparing with rain gauge precipitation observations in Nu River basin of Yunnan part, a typical mountainous area with sparse gauge observations. These four rain products were then adopted to drive the distributed hydrological model (Hydro-Mount) to simulate daily runoff. The results showed that under gauge scale, all three satellite precipitation products had a good correlation with rain gauge precipitation data with the correlation coefficient over 0.65 and probability of detection over 0.71 while on the amount of precipitation, CMO underestimated over 40%, and TRMM underestimated 0.42%, and RT underestimated 9.37%; it had similar but better accuracy results under watershed scale than it did at gauge scale. The TRMM-driven and RT-driven daily streamflow simulation performed litter better relative to gauge-driven streamflow with equivalent but more stable NSE values over the simulation period; the CMO-driven streamflow simulation had poor performance, but after eliminating certain system deviations, it also showed a certain simulation accuracy. The present study will hopefully be a reference for future accuracy correction and hydrological utilizations of satellite-based precipitation products.</span> 
</div>
%K 降水数据，TRMM，CMORPH，精度评估，径流模拟<br>Precipitation
%K TRMM
%K CMORPH
%K Accuracy Evaluation
%K Streamflow Simulation
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=29685