Award: OCE-1737158

This project aims to better understand the loss of dissolved oxygen from the oceans under global warming, termed as ocean deoxygenation, through a combination of observational and theoretical approaches. The uptake of heat from the atmosphere causes ocean temperature to rise, causing a decline of solubility of oxygen in the seawater. Ocean warming can cause changes in physical circulation, mixing and biochemical processes, further changing the distribution of oxygen in the sea. In this project, we examined the loss of oxygen in the historic dataset over the past 50 years together with computational modeling and theoretical analysis. The loss of oxygen is significantly stronger than expected oxygen loss from the warming-induced solubility change, which implies an important role for circulation and biogeochemical processes. The observed oxygen-heat ratio measures how many moles of oxygen have been lost per degree of warming of the seawater. This ratio can be used to estimate oxygen loss in relation to ocean heat uptake, and it can be an important benchmark for Earth system model simulations. This project examined how this ratio is determined globally and regionally. In the North Atlantic where deep waters form, this ratio primarily comes from the vertical gradients of temperature and oxygen. Vertical exchange of water between the surface and interior ocean covary according to their vertical gradients, which modulates the air-sea fluxes of heat and oxygen. Regionally, there is no simple relationship that links changes in ocean temperature and oxygen content as they are controlled by distinct physical and biochemical processes. Modes of climate variability play an important role determining the regional pattern of observed oxygen changes. The patterns of oxygen changes associated with Pacific Decadal Oscillation (PDO), for instance, were analyzed in observations and a computational model. The PDO pattern dominates the structure of natural variations in upper ocean oxygen changes in the North Pacific basin through the modulation of mixed layer depths, shifts in water masses, and biological consumption of oxygen in the subsurface waters. Globally, the vertical and isopycnal mixing of waters regulates the ventilation of oxygen and anthropogenic heat into the interior ocean. Through the analysis of a large number of model runs with varying mixing coefficients, a pattern emerged that the stronger ocean mixing leads to higher level of heat and oxygen inventories in the mean state, and the loss of oxygen under global warming is also stronger under stronger ocean mixing. These variations in the rate of ocean deoxygenation have a quasi-linear relationship with the rate of ocean heat uptake in the global scale. This implies that correct representation of the mean state is a requirement for better representation of oxygen changes under ocean warming. The impacts of dissolved oxygen loss on marine ecosystems must be considered in the context of warming, as the metabolic rates of many marine organisms depend on temperature. This project used a combination of metabolic theory, empirical data describing physiological traits, and numerical models to map variability and change in the ocean's capacity to support aerobic metabolisms. Understanding these metabolic constraints is a critical need to quantify the impacts of climate variability and change on marine ecosystems, including those supporting fisheries. Last Modified: 01/04/2022 Submitted by: Matthew Long