Climatic control of regional ocean deoxygenation: insights from recent trends in the north Indian Ocean

Zouhair Lachkar1, Muchamad Al Azhar1, Michael Mehari2, Marina Levy3 and K. Shafer Smith4, (1)New York University Abu Dhabi, Center for Prototype Climate Modeling (CPCM), Abu Dhabi, United Arab Emirates, (2)New York University Abu Dhabi, Center for Prototype Climate Modeling (CPCM), United Arab Emirates, (3)Laboratoire d'océanographie et du climat : expérimentations et approches numériques (LOCEAN), Paris, France, (4)New York University, Courant Institute of Mathematical Sciences, New York, NY, United States
Abstract:
The oxygen minimum zones (OMZs) of the Arabian Sea and the Bay of Bengal are among the most intense in the world. Observations suggest a decline of O2 over the recent decades in the Arabian Sea accompanied by an intensification of the suboxic conditions there. In contrast, no evidence of a recent intensification of the OMZ in the Bay of Bengal exists. Due to the scarcity of observations, the magnitude and the spatial patterns of these changes remain largely unknown and their drivers and biogeochemical implications unexplored. Here, we reconstruct the evolution of dissolved oxygen in the Arabian Sea and in the Bay of Bengal from 1982 through 2010 using hindcast simulations performed with an eddy-resolving ocean biogeochemical model forced with observation-based winds and heat and freshwater fluxes. We find a significant thermocline deoxygenation in the northern and western Arabian Sea and little O2 change in the Bay of Bengal. The total oxygen inventory declines in the northern Arabian Sea (north of 20N) on average by 3% per decade in the upper 200m and by up to 5% per decade in the underlying (200-1000m) layer, causing an increase of the volume of suboxia (O2 < 4 mmol m−3) and denitrification by up to 20% and 30%, respectively. In contrast, neither the oxygen inventory nor the OMZ intensity shows any significant change in the Bay of Bengal over the study period. Based on a set of sensitivity simulations, our analysis reveals that the deoxygenation seen in the northern and western Arabian Sea is associated with a combination of surface warming and weakening of the winter monsoon winds, resulting in a reduced winter convection and a shoaling of the thermocline depth there. In contrast, the effects of wind and heat flux changes oppose and cancel each other in the Bay of Bengal. Our findings highlight the challenge to predict the future evolution of ocean deoxygenation in regions subject to concurrent disturbances.