A spatial model for predicting North Indian Ocean tropical cyclone intensity: Role of sea surface temperature and tropical cyclone heat potential

Abstract

Tropical cyclones (TCs) are extreme weather events that may result in enormous loss of life and property especially for countries surrounding the North Indian Ocean (NIO) rim. In this study, a regional scale spatial statistical model has been developed to define the relationship between TCs and sea surface temperature (SST)/tropical cyclone heat potential (TCHP) for the period 1979 to 2017 over the NIO region. The spatial model employed here tessellates the NIO basin with hexagonal grids (area of each hexagon is approximately 213,961 km2) in order to analyze the relationship between cyclone intensity and the two predictors-SST and TCHP. The role of SST and TCHP contribution to the NIO cyclone intensity is determined by using a geographically weighted regression (GWR) method. This study postulates that a hexagon with positive coefficient signifies a direct relationship between the cyclone intensity and the predictors, for example, between SST (TCHP) and the Bay of Bengal (Arabian Sea) TCs. Based on robust model verification and test of significance, it is attributed that the spatial model using GWR has the potential to model NIO cyclone intensity, particularly using SST as the predictor for Bay of Bengal and TCHP for the Arabian Sea.

Keywords

• Tropical cyclone intensity
• Sea-surface temperature
• Tropical cyclone heat potential
• North Indian ocean
• Spatial model