The performance of the high‐resolution Advanced Research Weather Research and Forecasting (ARW) model for tropical cyclone prediction over the Bay of Bengal region of the northern Indian Ocean is assessed through study of 21 cyclone systems. Error metrics in the predicted fields of maximum sustained winds (MSW), central sea‐level pressure (CSLP) and the vector track position are computed by comparison with corresponding tropical cyclone estimations from the India Meteorological Department (IMD). From 65 sensitivity experiments for five severe cyclones the combination of Kain–Fritsch (KF) convection, Yonsei University (YSU) planetary boundary layer (PBL), LIN explicit microphysics and NOAH land surface schemes are found to provide the best simulations for intensity and track prediction. It has been found that the KF convection scheme gives higher convective warming with stronger vertical motions relative to other tested cumulus schemes and that the YSU scheme simulates more realistic winds in the inflow region than other tested PBL schemes. Results of simulations with the best physics for all 21 cyclones reveal that the model had a tendency to overestimate the intensity, with mean errors ranging from −2 to 15 hPa for CSLP, 1 to 22 m s−1for MSW corresponding to 24 to 72 h predictions. The mean vector landfall position errors are found to be 122 km at 12 h, 170 km at 24 h, 244 km at 48 h and 250 km at 72 h, and 67% of the landfall errors are less than 135 km, indicating fairly good forecasts. Further, the predictions are found to be best for northward moving cyclones followed by northwestward, westward and northeastward moving cyclones.
The Red Sea is a narrow, elongated basin that is more than 2000 km long. This deceivingly simple structure offers very interesting challenges for wind and wave modeling, not easily, if ever, found elsewhere. Using standard meteorological products and local wind and wave models, this study explores how well the general and unusual wind and wave patterns of the Red Sea could be reproduced. The authors obtain the best results using two rather opposite approaches: the high-resolution Weather Research Forecasting (WRF) local model and the slightly enhanced surface winds from the global European Centre for Medium-Range Weather Forecasts model. The reasons why these two approaches produce the best results and the implications on wave modeling in the Red Sea are discussed. The unusual wind and wave patterns in the Red Sea suggest that the currently available wave model source functions may not properly represent the evolution of local fields. However, within limits, the WAVEWATCH III wave model, based on Janssen’s and also Ardhuin’s wave model physics, provides very reasonable results in many cases. The authors also discuss these findings and outline related future work.