Analyses of dissolved oxygen concentration in Chesapeake Bay over the past three decades suggested seasonally-dependent changes in hypoxic volume and an earlier end of hypoxic conditions. While these studies hypothesized and evaluated multiple potential driving mechanisms, quantitative evidence for the relative effects of various drivers has yet to be presented. In this study, a coupled physical-biogeochemical model was used to conduct hindcast simulations between 1985 and 2016. Additional numerical experiments, in which the long-term trends in external drivers were removed, were analyzed to discern the separate effects of temperature increase, sea level rise and nutrient reduction. After the removal of seasonal and interannual variations, dissolved oxygen concentration in all regions of the estuary showed a statistically significant declining trend: ~0.1 mg/L per decade. Most of this decline occurred during winter and spring while May–August hypoxic volumes showed no changes and September hypoxic volume showed a slight decrease (~0.9 km3). Our simulations show that warming was the dominant driver of the long-term oxygen decline, overwhelming the effects of sea level rise and modest oxygen increases associated with nutrient reduction. There was no statistically significant trend in the initiation of hypoxia in spring, where the potential delay associated with nutrient reduction was offset by warming-induced oxygen declines, and both nutrient reduction and warming contributed to an earlier disintegration of hypoxia in the fall. These results suggest that recent warming has prevented oxygen improvements in Chesapeake Bay expected from nutrient input reductions and support the expectation that continued warming will serve to counter future nutrient management actions.
NOAA Ocean Acidification Program Award NA15NOS4780184