Interannual variability of upper ocean stratification in Bay of Bengal: observational and modeling aspects

Abstract

The annual cycle and interannual variability of stratification in Bay of Bengal (BoB) are studied using both observations and Global Ocean Data Assimilation System (GODAS) analysis during 2003–2012. Annual cycle of stratification and sea surface temperature (SST) evolve coherently, highlighting its role on modulating air-sea interaction over this climatologically important region. Spatial distribution of stratification shows strong seasonality in ARGO observations, whereas it is highly underestimated in GODAS with highest discrepancies during fall and spring. The annual cycle of sea surface salinity (SSS) in GODAS is out of phase with observations implying potential feedbacks. During La Nina years, SSS drop in fall and winter and are lesser than those reported during El Nino years. All these features are misrepresented in GODAS. As stratification modulates air-sea interaction over BoB especially during El Nino and La Nina years, such misrepresentation of ocean stratification may lead to unrealistic thermocline-SST coupling in the models. The mean stratification and its interannual variability in GODAS are weaker than observed even though interannual variability in freshwater flux (P-E) is higher in GODAS. Detailed analysis of GODAS with in situ observations reveals that upper ocean current shear (vertical) is overestimated in GODAS, leading to unrealistically strong mixing which is primarily responsible for the deeper penetration of surface warm and freshwater resulting weaker stratification. As GODAS is used to initialize the ocean component of the coupled forecasting system for seasonal prediction of Asian monsoon, proper representation of stratification is essential. This study advocates the need of accurate representation of upper ocean salinity in GODAS for improved stratification. We speculate that improved stratification and mixing in the BoB improve summer monsoon forecast.