2024 Theses Doctoral

Optimizing an ocean model to better assess oxygen and carbon cycling in the subpolar North Atlantic

Moseley, Lauren A.

Deep water formation in the Labrador Sea, a marginal sea within the subpolar region of the North Atlantic Ocean, is vitally important to the ventilation of the global ocean interior with atmospheric gases including oxygen (O₂) and carbon dioxide (CO₂). To better understand the current mechanisms of ocean ventilation, and improve predictions of future deoxygenation and anthropogenic carbon uptake, the complex relationships between physical processes, chemical properties, and biological activity must be unraveled. Ocean biogeochemical models (OBMs) can offer a more complete picture of the ocean state than the limited snapshots provided by observations. The overarching goal of this dissertation is to use a data-constrained OBM to examine the processes controlling O₂ and CO₂ variability in the central Labrador Sea.

In Chapter 2, I present the optimization of a data-assimilative regional OBM which simulates the physical and biogeochemical state of the North Atlantic Ocean from 2002 to 2017. The optimization process includes (1) removing the model spin-up to initialize the biogeochemical simulation from GLODAPv2.2016b 1° × 1° and other climatological estimates, (2) adjusting parameterized phytoplankton quantum efficiency, and (3) using a Green’s Functions approach to tune OBM parameters against O(105) in-situ biogeochemical measurements collected by BGC-Argo floats and research hydrography. I find significant model-data misfit reduction in the subpolar North Atlantic which demonstrably improve Labrador Sea modeled O₂, surface ocean pCO₂, and chlorophyll-a against independent satellite data and observation-based products.

Using this data-constrained model, I then investigate the seasonal and interannual variability of central Labrador O₂ and surface ocean pCO₂. The high-frequency SeaCycler mooring dataset provides unique insight into the convective region of the central Labrador Sea over 2016. I use SeaCycler data to better understand the model simulation and, in turn, use the model to expand these biogeochemical insights in space and time. In Chapter 3, I present an oxygen budget of the central Labrador Sea over 2016–2017 by decomposing modeled dissolved O2 into its advective transport, diffusive transport, biological, and air-sea flux terms. We find that the competing effects of air-sea exchange and diffusive mixing are so balanced that there is minimal O₂ storage in the upper 150 m. In Chapter 4, I examine modeled and observation-based estimates of surface pCO₂ against in-situ SeaCycler data.

Our analysis examines the seasonal and interannual variability of pCO₂ and reveals key biases in the non-thermal component of pCO₂, which is the dominant driver of modeled and estimated surface pCO₂ variability in the central Labrador Sea. Across all chapters, my dissertation works to bridge ongoing modeling and observational efforts to expand our understanding of ocean biogeochemical processes.