A NASA IMPACTS-based investigation of uncertainties in the GPM combined algorithm estimates of winter precipitation

Although based on dual-frequency space-borne radar observations, precipitation estimates from the GPM combined radar-radiometer algorithm are still characterized by significant uncertainties. Sources of uncertainties specific to the cold-season include: weaker reflectivities (especially above the freezing level) that make the interpretation of the dual-frequency reflectivity ratio challenging, the existence of a large variety of non-spherical ice particles whose electromagnetic scattering properties are hard to quantify, the contamination of the radar observations by ground clutter, and significant variability within the radar observing volumes. In this study, we use the NASA IMPACTS field campaign to investigate the impact of these uncertainties on the NASA GPM combined algorithm estimates of winter precipitation. The IMPACTS field campaign was conducted in the Northeastern US during the winters of 2020 to 2023 and included coordinated observations from instruments onboard the NASA ER-2, and P-3 aircraft, and the NASA GPM Core Observatory satellite. The NASA ER-2 aircraft was used to provide high-resolution triple-frequency radar observations, while the P-3 aircraft was used to provide in-situ microphysical observations. The observations from the airborne instruments are converted into GPM space-borne radar and radiometer observations through accurate retrievals and forward radar and radiometer models. Precipitation estimates are derived from the GPM synthetic observations and systematically evaluated against the reference airborne estimates. The airborne estimates are deemed to be more accurate than those derived from satellite observations given that more complete information is used in their derivation. The results of this study will be used to improve the GPM combined algorithm estimates of winter precipitation.