Causes of uncertainties in the representation of the Arabian Sea oxygen minimum zone in CMIP5 models

Abstract

Abstract. Open ocean oxygen minimum zones (OMZs) occur in regions with high biological productivity and weak ventilation. They restrict marine habitats and alter biogeochemical cycles. Global models show generally a large model-data misfit with regard to oxygen. Reliable statements about their future development and the quantification of their interaction with climate change are currently not possible. One of the most intense OMZs is located in the Arabian Sea (AS). We give an overview of the main model deficiencies with a detailed comparison of the historical state of ten climate models from the 5th coupled model intercomparison project (CMIP5) that present our present-day understanding of physical and biogeochemical processes. Considering a threshold of 60 μmol l−1, we find a general underestimation of the OMZ volume in the AS compared to observations, that is caused by a too shallow layer of oxygen-poor water in the models. The deviation of oxygen values in the deep AS is the result of subduction of higher oxygenated waters in the Southern Ocean in the models compared to observations. In addition, model deficiencies related to the coarse resolution of the abyssal ocean, are identified in the deep water mass transport from the Southern Ocean northward into the AS. Differences in simulated water mass properties and ventilation rates of Red Sea Water and Persian Gulf Water cause different mixing in the AS and thus influence the intensity of the OMZ. These differences also point towards variations in the parametrisations of the overflow from the marginal seas among the models. The results of this study are intended to foster future model improvements regarding the OMZ in the AS.

Oxygen, carbon, and pH variability in the Indian Ocean

“…Perhaps the low nitrogen to phosphorus ratios in the western Bay of Bengal are a harbinger of what is to come (Figure 1). This OMZ expansion is consistent with global modeling studies (Stramma et al, 2008;2010;Doney, 2010; Figure 3) but uncertainties in global model predictions are large (McCreary et al, 2013;Bopp et al, 2017;Rixen et al, 2020;Kwiatkowski et al, 2020;Schmidt et al, 2020;Schmidt et al, 2021). Given the importance of the Indian Ocean OMZs in the global carbon and nitrogen cycles, including the production of radiatively active greenhouse gas, it is essential to understand the biogeochemical variability associated with these regions, their rates of change and the potential implications for global warming.…”

“…Indian Ocean Central Water forms through convective mixing as Subantarctic Mode Water in the southern Indian Ocean and is advected northward into the OMZs in both the Arabian Sea and Bay of Bengal (Fine, 1993;McCartney, 1979;Sverdrup et al, 1942). Oxygen-enriched Persian Gulf Water is introduced into the Arabian Sea OMZ after its outflow from the Persian Gulf (Rixen et al, 2005;Schmidt et al, 2020;Schmidt et al, 2021;Tchernia, 1980;Lachkar et al 2019). The negative water balance of the Persian Gulf drives this localized deep-water formation by increasing the salinity and hence the density of Persian Gulf Water.…”