Passive microwave precipitation measurements for tropical cyclones based on a priori databases including various cloud microphysics schemes

The accuracy of a physically based passive microwave precipitation retrieval algorithm is affected by the quality of a priori databases consisting of precipitation information from cloud resolving models. In particular, various microphysics parameterizations assumed in cloud resolving models cause the different a priori microwave emission and scattering signatures. It is also found that for the same rainfall intensity emission and scattering signatures of tropical cyclones observed from the microwave imager onboard the Global Precipitation Measurement (GPM) satellite vary depending on the stages of development, growth and extinction of tropical cyclones. It suggests that using appropriate microphysics parameterization for a priori information is critical for retrieving precipitation from tropical cyclones at various intensity stages. Precipitation retrievals in this study begin with building a priori databases based on the Weather Research Forecasting (WRF) model simulations with five different microphysics schemes including Thompson, Morrison two moment, WRF Double Moment (WDM6), National Severe Storms Laboratory 2 moment (NSSL-2), and Thompson aerosol aware schemes. In the retrieval process, a specific a priori database is then selected for every target tropical cyclone scene by comparing the similarities in simulated and observed microwave emission and scattering signatures. Results show that WDM6 scheme tends to have the most similar characteristics of emission and scattering signals from weak to moderate rainfall regions. However, over the strong rainfall region WDM6 shows weaker scattering signatures and the most frequently selected scheme is found to be the Thompson scheme. Accordingly, the retrievals based on the selected a priori databases demonstrate improved statistics in comparison with GPM Dual-frequency Precipitation Radar (DPR) rainfall products.