Preindustrial ocean circulation alone drives over half of modeled Southern Ocean heat uptake

Why it matters

Most of the heat the ocean absorbs from global warming enters in the Southern Ocean, but the reasons for its pattern, size and timing were unclear. This work gives a simple, process-based account using parameters measured in the control run. The authors suggest it could serve as a starting point for explaining why models differ in ocean heat uptake. This is a preprint and has not been peer reviewed.

What they did

The authors used the CESM1 climate model with an abrupt 4×CO2 increase. They coupled a second 4×CO2 atmosphere to passive copies of temperature and salinity carried by a preindustrial control ocean, which keeps circulation and sea ice fixed. They also replaced the air–sea heat exchange over the Southern Ocean with a linear restoring law. They then built a minimal model: a well-mixed surface layer coupled to a one-dimensional advective–diffusive equation along the upwelling isopycnals. All parameters were measured in the control run.

Key findings

• With circulation fixed, centennial-mean Southern Ocean uptake was 0.25 PW, against 0.44 PW with free circulation. The slow decline over the century was reproduced.
• Uptake is strongest where upwelling renews the mixed layer with water not yet exposed to the warm air. A new exposure-time tracer matched the uptake pattern, with a rank correlation of −0.76.
• Long-term uptake is set by how fast the residual overturning renews the mixed layer. Doubling the air–sea exchange coefficient changed uptake by less than 5 %.
• Heat diffusing down the upwelling isopycnals explains the multidecade decline. The minimal model, with no fitted parameters, captured the magnitude and timing of CESM’s uptake.
• A linear restoring law reproduced the Southern Ocean uptake pattern with a pattern correlation of 0.97, but it fit the tropics poorly.

Limitations

• The results come from one coarse-resolution model, CESM1, in which overturning and isopycnal diffusion are heavily parameterized. Eddy-resolving models and the real ocean may behave differently.
• The fixed-circulation case misses sea-ice and circulation changes, which affect the pattern near the ice edge. The minimal model underestimates early uptake by about 25 % near year 25.
• The linear restoring law works over the Southern Ocean but not in the tropics, and the step-response shortcut assumes nearly uniform air warming there.

Glossary

• Residual overturning: The net circulation that carries water along density surfaces, combining resolved currents and the effect of eddies.
• Isopycnal: A surface of constant water density, along which water moves and mixes most easily.
• Redi diffusion: A model scheme for mixing of heat and other properties along isopycnals.
• Uptake efficiency (κ): Heat absorbed by the ocean per unit area for each kelvin of air warming.