Evaluating the Gap-Filling Capabilities of Satellite and Model Precipitation Products within the Multi-Radar Multi-Sensor (MRMS) System

The current Multi-Radar Multi-Sensor (MRMS) system generates quantitative precipitation estimation (QPE) products from a number of sources (radar,gauge,environmental data). These high resolution products provide a vital source of information for flash flood prediction, along with input to hydrological models. In the western US, the mountainous terrain causes coverage gaps in the radar and gauge products. The coverage limitations in this region lower the accuracy of the QPE products. Similar limitations are expected as the MRMS QPE products are expanded to include outside of continental US (oCONUS) domains such as Hawaii, Alaska, and the Caribbean. Satellite QPE and short-term numerical weather prediction (NWP) model quantitative precipitation forecast (QPF) data have the continuous coverage and low latency necessary to fill in the gaps in these regions. These new data sources also have their own limitations and sources of uncertainty which need to be taken into account when merging them with radar and gauge data into a new multi-sensor QPE product. A long-term statistical evaluation and analysis of the satellite and NWP model data sources is presented here for different climatological regimes and precipitation types. Verification datasets include quality-controlled gauge observations and the National Centers for Environmental Prediction (NCEP) stage IV precipitation product. Overall, the evaluation results indicate the IR-based satellite QPE is relatively more accurate in convective precipitation while the NWP model QPF performs best in large-scale stratiform precipitation.