Discrimination of light precipitation from COWVR+TEMPEST, GMI and ATMS passive microwave radiometer observations

In the near future, the number of low-cost, small/cubesat-sized passive microwave (MW) radiometric observations available from low Earth orbit (LEO) is expected to increase substantially, especially high frequency (HF) sounders (lowest operating band near 90 GHz). The number of sensors that include the constant-incidence, dual-polarization low (LF) frequency channels (highest operating band near 37 GHz) will remain steady or decline. Physically, the use of HF-only sensors for precipitation skews the precipitation estimates towards indirect, ice-based (scattering) retrievals, rather than the precipitation closer to the surface. Over certain Earth surfaces, a portion of light precipitation (defined as 2 mm hr-1 or less) may remain undetected, which has ramifications for global satellite-based precipitation products. In this manuscript, observations from the smallsat-sized LF Compact Ocean Wind Vector Radiometer (COWVR) and the HF Temporal Experiment for Storms and Tropical Systems (TEMPEST) duo are contrasted with the Global Precipitation Measurement (GPM) Microwave Imager (GMI) and the Advanced Technology Microwave Sounder (ATMS), for discriminating light precipitation over a variety of Earth surface conditions. Precipitation is discriminated using a surface emissivity-based transformation applied to selected channel combinations from each of these sensors, using the GPM Dual Frequency Precipitation Radar (DPR) products as precipitation references. The results highlight the importance of the LF channels for detection of light precipitation over ocean and lightly vegetated surfaces, and 50-GHz temperature channels over heavily vegetated surfaces. For discrimination from TB scenes that encompass mixed water/land fractions, the omission of the COWVR 18 GHz channel mitigates artifacts owing to resolution differences.