Abstract
Here, we investigated the impact of assimilating the satellite-based product of Global Positioning System (GPS) radio occultation (RO) refractivity profiles data on the simulation of selected extreme rainfall events in three states of southern India: Tamil Nadu, Telangana, and Kerala. We assimilated the GPS RO data into the weather research and forecasting model using a 3DVar assimilation technique and evaluated the results against unassimilated (control) simulations. Various observations (e.g., rainfall measurements from AWS/rain-gauge) and observation-based gridded rainfall were used. The assimilation of the data yielded improved prediction of the spatial distributions of extreme rainfall regions and the amounts of rainfall. The analysis of the simulated dynamical and thermodynamic processes indicated that the assimilation of the data enabled the model to simulate significantly deep convection, high instability, and strong vertical motions. A vorticity budget analysis confirmed the marginally strengthened low-level convergence. The vertical motions because of assimilation facilitated an increased vertical advection of vorticity, which enhanced the extreme conditions in storms. Moreover, the assimilation of the data resulted in enhanced water vapor condensation and high levels of ice, cloud, and rain water in clouds, all of which contributed to extreme rainfall.