Performance of Cloud Microphysical Parameterization In Simulating A Heavy Rainfall Event

This study evaluates the performances of four different cloud microphysical parameterization (CMP) schemes of WRF model at 3km horizontal resolution (lead time up to 96h) for the Heavy Rainfall Event (HRE) of Kerala during August 2018. The goal is to evaluate the ability of these sophisticated CMPs in replicating the event with special emphasis on their rainfall prediction skills. It is found that the choice of CMP has considerable impact on the rainfall forecast characteristics and associated processes. Results are validated against the IMD station data and GPM observations and it is found that WDM6 is the best performing scheme. Milbrandt and Thompson Aerosol Aware (WDM6 and WSM6) schemes were unable (able) to predict the cloud ice realistically leading to large (less) errors in the rainfall prediction. Further, it is also noted that mixed phase processes were dominant and the rainfall characteristics are profoundly modulated by different schemes due to their inherent variations in prediction of mixing ratios of the different hydrometeors. Moisture budget analysis suggests that horizontal moisture flux convergence (MFC) was the major (minor) driver of convection for higher (weaker) threshold of rainfall with WDM6 predicting the most consistent peaks of MFC in comparison with the TRMM rain rate peaks. The Contiguous Rain Area (CRA) analysis also found the WDM6 scheme to have the least volumetric error. Results clearly indicate that, the moisture convergence was prominent in the foothills of the Western Ghats and it was highly influential in facilitating Orography driven lifting of moisture over the region. Overall results infers substantial influence of CMPs on the forecast of the HRE, however, more accurate representation of cloud processes/parameters are required to reduce uncertainties in the parameterization for better prediction of rainfall.