The Southern Ocean absorbs approximately 40% of the global ocean sequestration of anthropogenic carbon dioxide (CO2) emissions, and in so doing, provides an important climate regulation service for the global society. Recent research indicates that carbon-climate feedbacks may already be manifesting in the Southern Ocean and suggests that the CO2 emission regulation service provided by the Southern Ocean may operate differently in the future than it does today.

New observations collected via autonomous Lagrangian floats equipped with biogeochemical sensors in the Southern Ocean find a smaller absorption of CO2 in the present-day Southern Ocean than that predicted by Earth System Models (ESMs), including the Department of Energy’s Energy Exascale Earth System Model (E3SM) and other models contributing output to the 6th Coupled Model
Intercomparison Project (CMIP6). This mismatch calls into question the ability of ESMs to accurately project CO2 absorption in this important region.

The project team proposes to investigate the discrepancy in observed and modeled Southern Ocean CO2 absorption through the lens of the E3SM, leveraging recent and upcoming E3SM development advances to simulate Lagrangian Biogeochemical-Argo floats. E3SM simulation output, together with observational data streams and output from the evolving CMIP6 archive, will be used to study the spatiotemporal bias in modeled Southern Ocean CO2 flux. The three research objectives are:

1. To quantify and understand modeled biases in present-day surface ocean partial pressure of CO2 and the corresponding air-sea CO2 fluxes
2. To evaluate the adequacy of the current observational network for quantifying model biases
3. To scale future projections of E3SM and other CMIP6 models according to their skill at replication of observational metrics.

The team will use statistical analysis and machine learning techniques to assess the spatiotemporal patterns of Southern Ocean CO2 flux bias across a range of model resolutions and external forcings, test the sensitivity of bias to observational interpolation, and project the future behavior of Southern Ocean CO2 absorption.

This project will be the first of its kind to simulate and analyze online Biogeochemical-Argo floats within an Earth system model framework. The project will enhance understanding of carbon-climate feedbacks in a rapidly changing, high latitude region critical to the global atmospheric carbon budget, and provide a foundational understanding of modeled Southern Ocean CO2 flux biases. Project results will be made visible through scientific publications and conference presentations. The groundbreaking E3SM model output will be made available to the public and will galvanize the community to consider online modeled Biogeochemical-Argo capabilities in other oceanic regions.