Kaiser et al. (2026) Evaluation of Gridded Data Sets for Representing Precipitation Event Properties Across 20 U.S. Gulf Coast Locations
⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.
Identification
- Journal: Geophysical Research Letters
- Year: 2026
- Date: 2026-11-10
- Authors: Samiul Kaiser, Ebrahim Ahmadisharaf, Amir AghaKouchak, Gabriele Villarini, Ceyda Polatel, Vasubandhu Misra
- DOI: 10.1029/2026gl125547
Research Groups
Not available in the abstract.
Short Summary
This study evaluates the skill of eight gridded sub-daily precipitation products in representing rainfall event properties (intensity, duration, and volume) across 20 U.S. Gulf Coast locations. It finds that rainfall volume is generally better represented than intensity and duration, often due to compensatory errors, highlighting that accurate volume can mask distorted event structure.
Objective
- To evaluate the skill of gridded sub-daily precipitation products (AORC, MRMS, MSWEP, Stage IV, PERSIANN, IMERG, ERA5, MERRA2) in representing rainfall event properties, specifically intensity (minimum, average, maximum), duration, and volume.
Study Configuration
- Spatial Scale: 20 U.S. Gulf Coast locations.
- Temporal Scale: 2015–2024 (9 years) for sub-daily precipitation events.
Methodology and Data
- Models used: AORC, MRMS, MSWEP, Stage IV (gauge-influenced precipitation products); PERSIANN, IMERG, ERA5, MERRA2 (gauge-independent precipitation products).
- Data sources: 128,141 rainfall events observed across the study locations, used for evaluation against the gridded products. Gauge observations are implicitly used for the gauge-influenced products and for validation.
Main Results
- Rainfall volume is generally represented better by the products than event intensity and duration, particularly for moderate and major events.
- This better representation of volume is partly attributed to compensation between underestimated intensity and overestimated duration.
- Among gauge-influenced products, MRMS and Stage IV best reproduce average and maximum intensities.
- PERSIANN performs best for intensity estimation among gauge-independent products, suggesting its utility in gauge-sparse regions.
- Minimum intensity is poorly represented by most products, indicating difficulty in capturing weak, intermittent rainfall.
- The study concludes that accurate event volume can mask distorted event structure, emphasizing the need to evaluate various event properties when selecting gridded precipitation products.
Contributions
- Provides a comprehensive evaluation of eight widely used gridded sub-daily precipitation products across multiple rainfall event properties.
- Identifies and quantifies the impact of compensatory errors (underestimated intensity, overestimated duration) on overall event volume accuracy.
- Offers specific insights into the performance of different product types (gauge-influenced vs. gauge-independent) for various intensity metrics.
- Highlights the critical importance of assessing a range of event characteristics beyond just total volume for a complete understanding of product skill.
Funding
Not available in the abstract.
Citation
@article{Kaiser2026Evaluation,
author = {Kaiser, Samiul and Ahmadisharaf, Ebrahim and AghaKouchak, Amir and Villarini, Gabriele and Polatel, Ceyda and Misra, Vasubandhu},
title = {Evaluation of Gridded Data Sets for Representing Precipitation Event Properties Across 20 U.S. Gulf Coast Locations},
journal = {Geophysical Research Letters},
year = {2026},
doi = {10.1029/2026gl125547},
url = {https://doi.org/10.1029/2026gl125547}
}
Original Source: https://doi.org/10.1029/2026gl125547