Assessment of WRF microphysics schemes in simulation of extreme precipitation events based on microwave radiative signatures

Passive microwave radiometric signatures are affected by microphysical properties and hydrometeor distributions. Various microphysical properties of hydrometeors assumed in numerical simulations of precipitating clouds cause the different emission and scattering signals. This study compares the radiometric signatures obtained from five microphysics (MP) schemes of the Weather Research and Forecasting (WRF) model during the simulation of eight typhoons with respect to Tropical Rainfall Measuring Mission Microwave Imager observations. The MP schemes include the Thompson (MP8), Morrison two moment (MP10), WRF Double Moment 6 Class (MP16), National Severe Storms Laboratory 2 moment (MP17), and Thompson aerosol-aware (MP28) schemes. The results show that most of the schemes produce insufficient emission/scattering signals from liquid/ice particles over the same rain rate compared to the observations. It was also confirmed that, depending on the schemes, the scattering signals appear to be more variable than the emission signals. In general, MP16 is more frequently identified as the scheme with the smallest biases in both low- and high-frequency channels for individual orbits. However, the results demonstrate that one scheme does not work best in all cases examined here. This suggests that the characteristics assumed in MP schemes should be carefully understood to simulate precipitating clouds. Moreover, comparisons of the radiometric signatures from simulations and observations might be used to understand the uncertainties caused by incorrect assumptions of the microphysical properties of precipitating clouds.