Validating GPM IMERG Precipitation using SMAP Soil Moisture

Abstract:

Hydrologic ground validation (Hydro-GV) is a unique approach to validate remotely sensed precipitation products, such as the Integrated Multi-satellitE Retrievals for GPM (IMERG) products. Using data from synergistic missions to characterize detection and false alarms at a global-scale to assess the performance of satellite rainfall products is needed to gain a better understanding of the uncertainties related to the use of such products in hydrologic model forecasts. Using IMERG in concert with the Soil Moisture Active Passive (SMAP) soil moisture estimates, a retrospective regional and global analysis using an independent dataset to validate the occurrence of precipitation in IMERG is conducted. This method will follow a multi-pronged approach for the assessment of IMERG. First, using a traditional network GV approach to validate IMERG over CONUS with the Multi-Radar/Multi-Sensor (MRMS) data to get a baseline skill for detection of precipitation. Following the traditional approach, our Hydro-GV method to detect precipitation over the same region will be conducted with changes in soil moisture from overpass-to-overpass from SMAP as a proxy for precipitation to build confidence at the regional-scale for the method. Upon successful completion of the regional-scale analysis, a subsequent expansion of the domain to the global-scale will be conducted with SMAP being used to validate IMERG. Early results indicate two key sources of uncertainties in the method that lead to increased false alarms: 1) light precipitation being evaporated before the following SMAP overpass, and 2) decreases in SMAP soil moisture that are within the range of SMAP uncertainty. Altering the categorization technique from the tradition 2x2 contingency table to a 3x3 contingency table, there was an enhance ability to detect precipitation occurrence, with over 70% of the domain experiencing increased skill. Following a successful demonstration of this method, the Hydro-GV method can be implemented in near-real-time to validate IMERG products.