The present study is aimed to investigate sub-surface ocean processes and their contribution to the intensification of a tropical cyclone (TC) from a coupled-modeling perspective. The Coupled Ocean–Atmosphere–Wave–Sediment Transport (COAWST) model was employed to simulate TC Phailin, which originated over the Bay of Bengal and made landfall on the eastern coast of India in October 2013. Three sub-surface ocean condition datasets—viz., (a) the European Centre for Medium-Range Weather Forecast (ECMWF) Ocean Reanalysis, (b) the Climate Forecast System Version 2 (CFSV2) Operational Analysis, and (c) the Hybrid Coordinate Ocean Model (HYCOM) Reanalysis datasets—were used for the initial and boundary conditions for the oceanic component of the coupled model in three different simulations of TC Phailin. All the simulations showed a delay in intensification compared to the observation, and ECMWF simulated the most intensified TC. CFSV2 simulated a deeper mixed layer (ML) and higher mixing, which hindered the intensification. Furthermore, higher entrainment of cold water in the ML led to cold water reaching the surface and, consequently, decreased sea surface temperature, which acted as negative feedback in the intensification of the storm in the cases of CFSV2 and HYCOM. ECMWF realistically simulated the interactions of the TC with a cold-core eddy before landfall. A sudden increase in ML heat content, the addition of heat in the ML due to entrainment, and the prevention of cold water reaching the surface were indicative of the breaking of the barrier layer (BL) in ECMWF, which was further corroborated by the spatial distribution of BL thickness in the simulation. This acted as positive feedback in the intensification of the TC. The findings of this study strongly suggest that not only the incorporation of physical oceanic sub-surface processes in the modeling of TCs but also the proper representation of prevailing mesoscale features and ocean sub-surface temperature, salinity, and current profiles in datasets is essential for realistic simulations of TCs.

Extreme rainfall events (ERF) and their associated regimes are among the major catastrophes that have a worldwide impact including India which result in large-scale flooding and immense socio-economic loss. During the southwest monsoon season, these extremes are becoming a frequent norm in the hilly and mountainous regions of the country such as Assam which received one of the most historical ERFs during June 14-17, 2022. The present study investigates the ERF implementing the Weather Research and Forecasting (WRF) model using two different land use and land cover data sets (i.e. ISRO and USGS) and three different sets of physical parameterization schemes (i.e. planetary boundary layer, cumulus, and microphysics) with a lead time up to 96 hours. Further ahead, the performance of rainfall in bias-corrected ensembles (BCE) along with a suite of model ensembles is rigorously quantified up to the district level in terms of its intensity, and distribution. Results suggest that among all the ensembles (E1-E10), USGS (E6 - E10) has underestimated rainfall (140-260 mm/day) compared to ISRO (150-280 mm/day), including the heavy rainfall (HR), very heavy rainfall (VHR), and extremely heavy rainfall (EHR) categories over the upper Assam division (UAD) and lower Assam division (LAD) of Assam throughout the 96 hours. Afterward, rigorous statistical analysis in terms of frequency distribution, Taylor diagram, and benchmark skill scores is carried out to elucidate the model biases for all the ensembles. Throughout the 96 forecast hour, BCE E5(E10) is noted with the distinct realistic(underestimated) representation of model rainfall bias over all the subdivisions of Assam. The findings of the present study encapsulate the critical role of the ensembles of physical parameterization schemes, along with proper LULC, and the BCE approach is required to overcome challenges to improve the skills of ERF, particularly over complex terrains of Indian subcontinent such as Assam.