Short-Term Predictability of the Bay of Bengal Region Using a High-Resolution Indian Ocean Model

Abstract An ocean circulation model, Nucleus for European Modelling of the Ocean (NEMO version 3.6) is customized to run at high-resolution over a regional domain [30oE-105oE; 20oS-30oN] in the Indian Ocean. It uses horizontal resolution of 1/12° in longitude/latitude and 75 levels in the vertical direction. The model well captures the observed space-time variations of temperature and salinity at the surface and subsurface, and the surface currents and eddy kinetic energy. The short-term spatio-temporal predictability of the Bay of Bengal (BoB) region is quantified using the model currents. The Lagrangian measure of predictability, Finite Time Lyapunov Exponent (FTLE) is compared with the Eulerian measure (Okubo-Weiss parameter). The regions of chaotic stirring are identified in the BoB. The FTLE analysis reveals that the predictability on a biweekly time scale in the BoB is minimum during October-November, and the highest during May to July. The FTLE is shown to serve as a useful tool for planning targeted observations in the BoB region.