Gu et al. (2026) Has the latest IMERG V07 from GPM improved the performance of precipitation estimation of regional-scale compared to its predecessor?

Identification

Journal: Journal of Hydrology
Year: 2026
Date: 2026-02-12
Authors: Jingjing Gu, Y T Ye, Haozhe Guan, Zhiyong Zhou, Yin Cao, Yunzhong Jiang
DOI: 10.1016/j.jhydrol.2026.135114

Research Groups

School of Civil Engineering, Tianjin University, Tianjin 300072, China
State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, China
Department of Water Resources, China Institute of Water Resources and Hydropower Research, Beijing 100038, China

Short Summary

This study systematically evaluates and compares the performance of the latest IMERG V07 precipitation product against its predecessor V06 across the Yellow River Basin. It finds that IMERG V07 significantly improves precipitation estimation and rain/no-rain detection capabilities, demonstrating reduced sensitivity to environmental factors compared to V06.

Objective

To systematically evaluate and compare the performance of IMERG V07 (Early, Late, and Final runs) against its predecessor V06 across the Yellow River Basin and its sub-regions, using gauge observations as reference. This evaluation focuses on precipitation estimation accuracy, rain/no-rain detection capability, response to extreme weather, analysis of error sources, and the dependence of product performance on gauge density and altitude.

Study Configuration

Spatial Scale: Yellow River Basin (YRB) and its sub-regions.
Temporal Scale: Comparison of IMERG V06 and V07, covering a period suitable for evaluating the latest version (V07 released July 2023).

Methodology and Data

Models used: Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (IMERG) Version 06 and Version 07 (Early, Late, and Final runs).
Data sources: IMERG satellite precipitation products, Gauge observations (as reference).

Main Results

IMERG V07 consistently outperforms V06 across all continuous and categorical metrics, exhibiting significant regional variability in performance for both versions.
While V07 and V06 show comparable skill in reproducing precipitation frequency during extreme events, V07 demonstrates significantly higher consistency with gauge observations for duration-based indices, specifically maximum consecutive dry days (CDD (max)) and maximum consecutive wet days (CWD (max)).
Except for Late runs, error components (false and missed bias) are consistently higher than the total error for other IMERG V06 and V07 datasets, being the dominant contributors. The systematic error of all datasets is significantly lower than the random error.
IMERG V07 outperforms V06 across most metrics in both flood and non-flood seasons, with all products showing enhanced skill in detecting precipitation events during the flood season.
Most evaluation metrics for both IMERG versions correlate positively with gauge density and negatively with altitude changes; however, V07 demonstrates a reduced sensitivity to these environmental factors compared to V06.

Contributions

Provides the first comprehensive regional-scale evaluation of the latest IMERG V07 datasets, including varying latency runs, and quantifies their improvements over V06.
Offers critical guidance for both users and developers of satellite-based precipitation estimation products, promoting further advancements.
Analyzes the dependence of product performance on gauge density and altitude, and identifies dominant error sources.

Funding

Not specified in the provided text.

Citation

@article{Gu2026Has,
  author = {Gu, Jingjing and Ye, Y T and Guan, Haozhe and Zhou, Zhiyong and Cao, Yin and Jiang, Yunzhong},
  title = {Has the latest IMERG V07 from GPM improved the performance of precipitation estimation of regional-scale compared to its predecessor?},
  journal = {Journal of Hydrology},
  year = {2026},
  doi = {10.1016/j.jhydrol.2026.135114},
  url = {https://doi.org/10.1016/j.jhydrol.2026.135114}
}

Original Source: https://doi.org/10.1016/j.jhydrol.2026.135114