This study conducted evaluation and analysis on various precipitation products over the eastern Tibetan Plateau (ETP), including four sets of satellite precipitation data (i.e., IMERG_Uncal, IMERG_Cal, GSMaP_MVK, GSMaP_Gauge) and one set of model reanalysis data (i.e., ERA5-land, hereafter ERA5-L). We evaluated the spatial-temporal distribution of their quality at an hourly temporal scale and 0.1° spatial scale, with a special focus on capturing different types of precipitation. The results show that: (a) GSMaP_Gauge exhibits the highest correlation with ground-based gauges, while IMERG_Uncal and IMERG_Cal perform best in the estimation of the amount of precipitation. Satellite products generally perform better during summer while ERA5-L sometimes outperforms satellite products in spring and autumn. (b) The evaluation results for different precipitation types reveal that all the QPE products face significant challenges in accurately describing convective precipitation. They tend to underestimate convective precipitation and fail to properly capture the intensity and location of heavy precipitation. (c) In heavy convective precipitation cases, the evaluated QPE products show various issues in accurately capturing the intensity and spatiotemporal variation of precipitation. Almost all QPE products underestimate maximum precipitation (both hourly precipitation and accumulated precipitation) and small-scale (about 50 km or less) spatial variability of precipitation. IMERG_Uncal, IMERG_Cal, and GSMaP_MVK perform better than other products in heavy convective precipitation cases. This study provides new insights into the quality of QPE products in different types of precipitation. The analysis of the quality of these QPE products serves as a valuable indicator of their potential applications, particularly in flash flood simulations, while also underscoring the critical need for improving precipitation product quality.

中文翻译：

---

卫星或再分析降水产品能否描述青藏高原东部山区山洪规模的降雨位置和强度？


本研究对青藏高原东部 （ETP） 的各种降水产品进行了评价和分析，包括 4 组卫星降水数据（即 IMERG_Uncal、IMERG_Cal、GSMaP_MVK、GSMaP_Gauge）和 1 组模式再分析数据（即 ERA5-land，以下简称 ERA5-L）。我们在小时时间尺度和 0.1° 空间尺度上评估了其质量的时空分布，特别关注捕获不同类型的降水。结果表明：（a） GSMaP_Gauge 与地面测雨量仪的相关性最高，而 IMERG_Uncal 和 IMERG_Cal 在降水量估计中表现最好。卫星产品通常在夏季表现更好，而 ERA5-L 有时在春季和秋季的表现优于卫星产品。（b） 不同降水类型的评价结果表明，所有 QPE 产品在准确描述对流降水方面都面临着重大挑战。他们往往低估了对流降水，无法正确捕捉强降水的强度和位置。（c） 在强对流降水情况下，评估的 QPE 产品在准确捕捉降水的强度和时空变化方面表现出各种问题。几乎所有的 QPE 产品都低估了最大降水量（每小时降水量和累积降水量）和小尺度（约 50 公里或更短）降水的空间变率。IMERG_Uncal、IMERG_Cal 和 GSMaP_MVK 在强对流降水情况下的性能优于其他产品。本研究为不同类型沉淀中 QPE 产品的质量提供了新的见解。 对这些 QPE 产品质量的分析是其潜在应用的重要指标，尤其是在山洪模拟中，同时也强调了提高降水产品质量的迫切需求。