Precipitation’s Role in the 2017 Oroville Dam Crisis Authors: Kwo-Sen Kuo (Bayesics, LLC); Mike P Bauer (Bayesics, LLC); Carrie Vuyovich (NASA Goddard Space Flight Center) Corresponding author email: kuo@bayesics.com; kkuo@umd.edu Abstract: This study investigates the meteorological and hydrological factors contributing to the 2017 Oroville Dam crisis in California. Managed as the tallest dam in the U.S. and a critical water supply, the Oroville Dam experienced a catastrophic failure of its main spillway in February 2017 following a series of intense winter storms. This event led to the evacuation of nearly 200,000 people and hundreds of millions of dollars in infrastructure damage. Using NASA's MERRA-2 reanalysis data and the NOAA SNODAS data product, this research examines the large-scale context of the 2017 winter, characterized by the interplay between extratropical cyclones, atmospheric rivers, and precipitation. The analysis highlights the role of the precipitation phase—specifically, the transition from snow to liquid rainfall—triggered by warm atmospheric rivers originating from the southwest, and the critical timing of this transition, which coincides with rising snowpack temperatures.

Bridging the Sensitivity Gap in Precipitation Estimates from Spaceborne Radars using Passive Microwave Observations Authors: Simon Pfreundschuh (Colorado State University); Christian D. Kummerow (Colorado State University) Corresponding author email: simon.pfreundschuh@colostate.edu Abstract: Current global passive microwave (PMW) precipitation retrievals in NASA’s and JAXA’s Global Precipitation Measurement (GPM) mission rely on reference precipitation estimates from the GPM Dual-Frequency Precipitation Radar (DPR), whose sensitivity to light and frozen precipitation is limited. Consequently, PMW retrievals calibrated against DPR reproduce reduced precipitation rates in these regimes even when microwave observations retain sensitivity to them. This discrepancy is particularly evident at high latitudes, where satellite retrievals differ systematically from reanalysis products. We develop a precipitation retrieval for the GPM Microwave Imager that combines CloudSat-based reference estimates for light and frozen precipitation with DPR-based estimates for heavier precipitation. Validation against shipborne disdrometer measurements shows improved precipitation detection and reduced high-latitude accumulation deficits relative to retrievals trained solely on DPR reference data. However, the inclusion of CloudSat information increases instantaneous retrieval errors, including mean absolute error, mean squared error, and linear correlation. We therefore compared CloudSat and DPR reference estimates with ground-based precipitation radar measurements and with each other. CloudSat precipitation rates show a positive bias, large random errors, and low correlation even within the shared sensitivity range of the two radars. The low correlation stems from categorial errors in which CloudSat seems to estimate very high rain rates (not seen by DPR or the ground-based radars) or CloudSat decreases rain by a factor of more than 10 when the surface echo appears to contaminate the signal. Combined with the retrieval results, this strongly suggests that CloudSat observations provide useful precipitation occurrence information but with limited quantitative accuracy. This explains why their inclusion improves the detection of light precipitation in PMW retrievals but does not lead to improved correlations with in-situ observations. Our results show that passive microwave observations can recover precipitation regimes missed by current spaceborne radars, improving the representation of light and frozen precipitation in global satellite products. At the same time, the large quantitative uncertainties in CloudSat precipitation rates, widely used to constrain global precipitation estimates, indicate that significantly higher levels of quality control are needed if CloudSat is used to constrain current global precipitation estimates.

Characterizing precipitation system variability from multi-decadal spaceborne radar records Authors: Masafumi Hirose (Meijo University) Corresponding author email: mhirose@meijo-u.ac.jp Abstract: The Tropical Rainfall Measuring Mission Precipitation Radar (TRMM PR) and the Global Precipitation Measurement Core Observatory Dual-frequency Precipitation Radar (GPM DPR) provide more than a quarter century of near-continuous observations. These records enhance the statistical representativeness of precipitation across diverse environmental regimes and improve the characterization of spatiotemporal variability, as well as associated observational and retrieval uncertainties. Advances in sampling and algorithms have further increased the scientific value of these datasets, despite challenges posed by short-lived, localized precipitation events that strongly influence long-term statistics and sampling limitations from low–Earth orbit sensors. Combined radar records remain uniquely valuable for representing mean precipitation patterns and their variability. Although the pursuit of higher statistical robustness and estimation accuracy accentuates existing limitations, sustained operational and algorithmic improvements have progressively enhanced the continuity and reliability of precipitation climatologies. It has become increasingly evident that precipitation variability is closely linked to environmental fluctuations across scales, and that local topography exerts a strong influence on retrieval errors. Increasing GPM DPR data enable systematic studies of seasonal variability in high latitudes, while measurement limitation—arising from surface clutter masking and sensitivity limitations—continue to cause underestimation, complicating interpretation. In low latitudes, longer records have refined understanding of seasonal and interannual variability, including additional rare high-impact precipitation events. Therefore, continued data accumulation and systematic evaluation of sensor characteristics and orbital changes are essential for improving the quality of datasets that represent precipitation variability. This study analyzes a high-resolution gridded precipitation dataset from TRMM PR and GPM DPR Ku-band precipitation radar for the period 1998–2025, showing that long-term records refine precipitation climatology. Findings include detection of rare events impacting spatial means, characterization of kilometer-scale spatial features, improved spatial coherence of diurnal precipitation variations, and identification of seasonal shifts in peak precipitation timing linked to localized systems.