Accuracy assessment of rainfall prediction by global models during the landfall of tropical cyclones in the North Indian Ocean

Abstract

Listen

Translate article

ABSTRACTThe rainfall prediction accuracy of seven state‐of‐the‐art operational models during landfall (LF) of tropical cyclones formed in the North Indian Ocean was assessed. The rainfall predicted by these models 1–3 days (24–72 h) before the LF day of tropical cyclones was compared with TRMM_3B42 rainfall data. Cyclone induced rainfall occurs along its track; thus rainfall within 600 km along the cyclone track was used for comparison. The error in cyclone track prediction affects the rainfall prediction. This has been taken care of by shifting the rainfall distribution over the cyclone best track positions provided by the India Meteorological Department. The validation statistics for the rainfall prediction accuracy of the models were computed by combining the forecasts generated by these seven models during LF of nine cyclones formed in the North Indian Ocean during 2010–2013. The skill of the models was analysed using different rainfall validation measures/statistics. Based on these validation statistics it was found that, of the seven global models, the models from four centres, namely the National Center for Environment Prediction, the European Centre for Medium‐range Weather Forecasts, the UK Meteorological Office and the Canadian Meteorological Centre, performed closer to the observations. However, there was no single model which was able to exhibit all the features of the observed rainfall.