The Ocean Barrier Layer in the Eastern Indian Ocean as a Predictor for Rainfall over the Indonesian-Australian Region

Barrier layers in the tropics trap heat in a shallow and stable near-surface layer limiting entrainment of cooler water from below and increasing sea surface temperature (SST). These warmer SSTs possibly enhance atmospheric convection and lead to rainfall. This study investigates whether salt-stratified barrier layers in the eastern Indian Ocean significantly influence rainfall over the Indonesian-Australian Continent (IAC). Argo profile observations show a local maximum of barrier layer thickness (BLT) in the equatorial West Sumatra (WS) region, in November-December-January (NDJ), coinciding with the wet monsoon season December-January-February (DJF) in northern Australia. The high resolution (0.25° atmosphere/land and 0.1° ocean/sea ice) fully coupled pre-industrial Energy Exascale Earth System Model version 0.1 (E3SMv0.1) simulation is used to investigate the seasonal relationships between BLT, local atmospheric convection and remote rainfall. The WS seasonal BLT maximum is accompanied by a seasonal minimum of outgoing longwave radiation at the top of the atmosphere and a maximum of latent heat flux, suggesting intensified atmospheric convection and increased ocean evaporation. The WS region is strategically located in the passage of the East-Asian monsoon moisture flow toward the northern Australia. Linear regression of the regional moisture transport with BLT in the WS region closely resembles the Indo-Australian monsoon dynamics in DJF. This suggests that the presence of the BLT in this region, by intensifying the local convection modifies and amplifies the moisture transported to Australia, thus increasing the rainfall there. The BLT predictive skills are evaluated via partial least square (PLS) regression of the E3SMv0.1 rainfall. The PLS fit is derived using area-averaged WS monthly time series of BLT as well as other ocean predictors: mixed layer depth, sea surface temperature and salinity, and the IOD and Niño3.4 large-scale climate indices. At a one-month lag, the most significant correlations of the rainfall are with the BLT with the largest positive correlations over northeastern Australian continent.