Precipitation, a crucial element of the water cycle, significantly impacts surface streamflow and flooding dynamics. The latest version of Integrated Multi-satellitE Retrievals for GPM (IMERG V07) has garnered global attention for its advancements over its predecessor, IMERG V06. However, the improvement in precipitation rates has not yet been fully quantified, especially when translated into improvements in hydrologic predictions. In this study, we aim to quantify the improvements of IMERG V07 over V06 in the contiguous United States (CONUS) in the aspects of (1) Evaluating the accuracy of precipitation data against Multi-Radar Multi-Sensor (MRMS); and (2) Comparing their hydrologic performance using a hydrologic model, the Coupled Routing and Excess Storage (CREST), against United States Geological Survey (USGS) streamgages. This study mainly finds that: (1) Metrics for both precipitation and streamflow from CREST show that IMERG V07 significantly outperforms IMERG V06. Specifically, the CC improved from 0.391 to 0.443 for precipitation and from 0.487 to 0.515 for streamflow; (2) The improvements in IMERG V07 are region-dependent. Significant improvements are found in basins with small areas (< 1000 km2), in mid-latitudes (41° N to 43° N), at low average elevations (< 800 m), and those located in the northeastern CONUS; (3) In certain cases, IMERG V07 demonstrates a better capability in estimating extreme precipitation, whereas IMERG V06 tends to underestimate it. This is also reflected in the streamflow data, where IMERG V07 better captures flood peaks compared to IMERG V06. This research enhances our understanding of flood dynamics by analyzing IMERG V07′s advancements and their effects on hydrologic predictions. It offers valuable insights into improved precipitation data’s role in hydrological modeling, giving potential benefits for simulating better flood prediction and helping in water management.

中文翻译：

---

与 IMERG V06 相比，最新的 IMERG V07 如何改进 CONUS 的降水估计和水文效用？


降水是水循环的关键要素，对地表径流和洪水动力学有重大影响。最新版本的 Integrated Multi-satellitE Retrievals for GPM （IMERG V07） 因其优于其前身 IMERG V06 的进步而受到全球关注。然而，降水率的改善尚未完全量化，尤其是当转化为水文预测的改进时。在本研究中，我们旨在量化美国本土 （CONUS） IMERG V07 相对于 V06 的改进，包括以下方面：（1） 评估多雷达多传感器 （MRMS） 降水数据的准确性;（2） 使用水文模型、耦合路由和超额存储 （CREST） 与美国地质勘探局 （USGS） 流量计比较它们的水文性能。本研究主要发现：（1） CREST 的降水和径流指标均表明 IMERG V07 的性能显著优于 IMERG V06。具体而言，降水的 CC 从 0.391 提高到 0.443，径流的 CC 从 0.487 提高到 0.515;（2） IMERG V07 的改进取决于地区。在小面积流域（x3C 1000 km2）、中纬度地区（北纬41°至43°N）、平均海拔较低地区（x3C 800 m）和位于北大西洋东北部的流域中，情况得到了显著改善;（3） 在某些情况下，IMERG V07 在估计极端降水方面表现出更好的能力，而 IMERG V06 往往低估了它。这也反映在径流数据中，与 IMERG V06 相比，IMERG V07 更好地捕捉了洪水峰值。这项研究通过分析 IMERG V07 的进展及其对水文预测的影响，增强了我们对洪水动力学的理解。 它为改进的降水数据在水文建模中的作用提供了有价值的见解，为模拟更好的洪水预测和帮助水资源管理提供了潜在的好处。