Impact of initial conditions and cloud parameterization on the heavy rainfall event of Kerala (2018)

In Aug 2018, the extreme precipitation event had created devastation over Kerala's state for 1 week and led to more than 483 deaths, and destructions of properties cost more than 50,000 Crore rupees. We simulated the event with the Advanced Research Weather Research and Forecasting model (WRF-ARW). Two most popular initial conditions (ICs), i.e., the National Centre for Environment Prediction (NCEP-FNL) and European Centre for Medium-range Weather Forecasting (ECMWF-ERA5), have been used to investigate the impacts of ICs on the forecast skills (up to day1) with particular reference to rainfall. Using these ICs, the study simulated the event with four cloud microphysics parameterization schemes (CMPS), namely, Milbrandt–Yau, Thompson, Aerosol aware Thompson, and WDM6. The study examined the role of ICs in modulating the cloud microphysical properties and associated mechanisms impacting the rainfall characteristics, hydrometeor formation, and related convective mechanism over the region. A total of 24 simulations (2 ICs × 4 CMPS × 3 initial days 13–15 Aug 2018) are conducted. Results show that the ERA5 IC with Milbrandt–Yau CMPS has performed best among all the simulations. Nevertheless, in general, there is an underestimation of rainfall in the model. It is found that though an ample amount of moisture available at the lower levels of the troposphere in the model, however, due to the lack of intense updraft, this moisture could not reach the middle levels, resulting in robust dryness at these levels and adversely impacting the convection and rainfall. This process shows a deficiency in the formation of frozen hydrometeors in the model's upper levels, leading to the lack of cloud heating and weakly developed vertical pressure gradient. These chain processes resulted in weak convection and reduced rainfall in the model experiments. These findings are highly relevant to understand the limitations of the model, influences of ICs, and cloud microphysical parameterizations to forecast extreme rainfall events accurately.