Performance Evaluation of GPU-Based WRF Model in Simulating a Unique Tropical Cyclone of Arabian Sea:A Case Study of VSCS Vayu

Tropical cyclone (TC) Vayu developed from a low-pressure system on 9 June 2019 near the West coast of India. It underwent rapid intensification (RI) to a very severe cyclonicstorm (VSCS) before weakening into a deep depression on 17 June 2019 with a unique track. The present study aims to evaluate the performance of the GPU-WRF model in simulating the unique tropical cyclone Vayu when initialized with different meteorological boundary conditions and the effect of input of time-varying SST data, the track, and the cyclone's intensity. The study also aims to investigate the cyclone's synoptic parameters during its development and intensification. Four simulations are conducted with GFS and NCEP-FNL data with and without SST input. The four-dimensional data assimilation analysistechnique, the fdda analysis nudging scheme, was used on the GFS data with SST input, which showed a significant improvement in track and intensity. The system skirting the Gujrat coastline on 13 June is skillfully captured. Given the appreciable improvement of track and intensity with GFS data using nudging, further investigation of the cyclone's synoptic parameters is done on the same. Overall, comparing the simulated dynamics with the ERA-5 dataset indicated that the model simulated a stronger TC. WRF can skillfully simulate a well-delineated eye wall at the matured stage (wind speed >40 m/s). An anomalously high mid-tropospheric relative humidity (RH) (~90%) is indicated at the developing stage, indicating the onset of RI, during which the system showed RH ~100% at the mid-troposphere. On 14 June, when the system reached VSCS, the simulated storm's low-level relative vorticity was ~359.93 × 10 ^–5 s ^-1 , whereas ERA-5's was ~175.39 × 10 ^–5 s ^-1 only.The simulated storm cyclone energy (9.35×10 ⁴ knts ² ) was lower than the observed (11.54 ×10 ⁴ knts ² ). The significance of the results obtained from the study is that the model can skillfully simulate the track and intensity ofan Arabian Sea TC and capture TC Vayu's cyclogenesis. The study also provides insight into the cyclone's synoptic parameters, such as mid-tropospheric relative humidity, low-level relative vorticity, and cyclone energy, during its development and intensification. The study's findings can be useful in improving the accuracy of tropical cyclone forecasting and enhancing our understanding of the physical processes involved in their formation and intensification for the Arabian Sea region.