Development, Assessment, and Applications of a New Multi-Sensor High-Resolution Dataset of QPEs for Southern California

The mountainous regions of Southern California experience both wet and dry extremes in precipitation. During the wet extremes, short periods of localized high intensity precipitation often occur within the context of a larger storm. In orographic areas with significant topographic slope, hydrometeorological impacts (namely, shallow landslides and flash flooding) are often triggered by these types of precipitation extremes. More precisely characterizing the triggering conditions of these land surface impacts is critical in improving hazard mitigation strategies. This requires high-resolution precipitation information in both space and time, which presents a challenge in topographically complex areas due to sparse sub-daily rain gauge density and issues with radar measurements in these regions. Additionally, better understanding the characteristics of precipitation events most likely to result in these hydrometeorological impacts at a higher spatiotemporal resolution may yield improvements in event predictability. With this goal in mind, this work has developed a decadal (2014-2023) record of high-resolution (4km spatial, 6-hourly) quantitative precipitation estimates (QPEs) by merging bias-corrected satellite precipitation estimates with in-situ gauge measurements. Thorough characterization of remaining dataset uncertainty is critical information for any QPEs used for the purposes of hydrologic model forcing data. To this end, the stability of the bias correction methodology in topographically complex regions is evaluated using a Monte Carlo cross validation methodology. Additionally, the merged dataset is validated in areas of differing topographic characteristics through an examination of uncertainty metrics and distribution characteristics. An application of this high-resolution dataset is also outlined concerning upstream atmospheric precursor conditions to mountainous extreme precipitation events of varying duration. The high resolution of this dataset in space and time, preservation of precipitation variability, and quantification of remaining uncertainty makes it suitable for use in other hydrometeorological applications, as well.