Optimization of hybrid data assimilation for cases of very heavy rainfall events over the Indian region

This study endeavors to enhance the performance of Hybrid Data Assimilation (DA) system, specifically focusing on the complex meteorological conditions of the Indian region. The optimization process involves adjusting five key parameters: (i) incorporating Background Error Statistics dependent on flow (FB); (ii) selecting different assimilation datasets; (iii) fine-tuning the optimal sample size for generating FB ensembles; (iv) establishing a detailed weighting scheme for FB against Static Background Error Statistics (SB); and (v) refining convergence criteria for the DA process. Numerical DA experiments (using WRF-4.3.3) were carried out with a 30‐km horizontal grid spacing, covering ten cases from 2018 to 2022, with a particular emphasis on significant rainfall events during the South West Monsoon season, associated with Depressions or more significant weather systems over the Bay of Bengal region. Three distinct FB formulations were formulated using ensembles: (i) B-MPCU (varying microphysics & cumulus schemes) (ii) B-RAD (varying microphysics, cumulus, short and long wave radiation schemes), and (iii) B-PERT (only perturbing initial conditions). The SB was derived using the National Meteorological Center method. A comparative analysis against ground truth sources, including radio sonde, real-time metar observations, and model forecasts with GPM 0.1°X 0.1° rainfall estimator outputs, established the superiority of Hybrid DA techniques under specific conditions: (i) using B-MPCU as FB; (ii) setting FB's contribution against SB at 99%; (iii) fixing the ensemble sample size for FB at 45; (iv) simultaneously assimilating varied set of observational data (both conventional and satellite); and (v) setting convergence criteria at the exacting threshold of 0.0001. This scientific inquiry not only advances our understanding of assimilation techniques but also provides a robust framework for their optimal deployment in the complex meteorological context of the Indian region.