青藏高原（QTP）作为气候变化高度敏感地区，能够准确估计其降水量的时空分布，获取更加精确的降水数据，对于对QTP地区的关键基础设施、区域经济甚至是人类福祉十分有益。站点实测数据（OBS）往往代表了所在位置的真实降水, 卫星遥感数据能够连续地反映降水的时空分布，数据融合能够将多来源的、不完整的信息加以综合, 形成相对完整一致、准确统一的信息。在本研究中，基于中国气象局运营的104个气象站的地面观测日降水量数据，首先对QTP在2001-2015年三种SPPs（分别为CMORPH、IMERG、TRMM3B42V7）的性能表现进行了评价，然后择优选出了两种SPPs （IMERG和TRMM3B42）与地面站点降水数据进行产品融合，最终获得一套高精度的青藏高原日降水产品。主要研究结果如下：（1）在空间分布方面，3种产品均表现出与站点观测数据一致的降水空间分布特征：降水量从西北到东南逐渐递增的趋势；（2）在季节性分布方面，三种产品都捕捉到了一致的季节性变化：夏季多，冬季少，且IMERG与站点观测有更好的一致性；（3）在精度评价上，相比其它两种SPPs，IMERG性能最好，表现为它年、季度尺度上有较高的相关系数（CC）、KGE和较低的均方根误差（RMSE），表明IMERG与站点实测数据最为接近；（4）经过ANN方法融合后的降水数据在空间分布上更为精细，拥有更高的CC值、KGE值与更低的RMSE值，与实测数据的吻合度更高。

The Qinghai-Tibet Plateau (QTP) is a highly sensitive region to climate change. Accurate estimation of the temporal and spatial distribution of precipitation and the acquisition of more accurate precipitation data will be of great benefit to critical infrastructure, regional economy and even human well-being in the QTP region. Station measured data (OBS) often represent the real precipitation at the location, satellite remote sensing data can continuously reflect the temporal and spatial distribution of precipitation, and data fusion can integrate incomplete information from multiple sources to form relatively complete, consistent, accurate and unified information. In this study, based on the daily precipitation data of 104 meteorological stations operated by the China Meteorological Administration, the performance of QTP in three SPPs (CMORPH, IMERG, TRMM3B42V7) during 2001-2015 was first evaluated. Then, two SPPs (IMERG and TRMM3B42) were selected and integrated with the precipitation data of ground stations to obtain a set of high-precision daily precipitation products over the Qinghai-Tibet Plateau. The main results are as follows: (1) In terms of spatial distribution, the three products all showed the same spatial distribution characteristics as the observed data: precipitation gradually increased from northwest to southeast; (2) In terms of seasonal distribution, the three products all captured consistent seasonal changes: more in summer, less in winter, and IMERG had better consistency with site observations; (3) Compared with the other two SPPs, IMERG has the best performance in terms of accuracy evaluation, as it has higher correlation coefficient (CC), KGE and lower root mean square error (RMSE) on annual and quarterly scales, indicating that IMERG is the closest to the measured data of the site; (4) The precipitation data after the fusion of ANN method has a finer spatial distribution, with higher CC value, KGE value and lower RMSE value, which is more consistent with the measured data.