Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Subantarctic Stratification in Climate Models Controlled by Ocean Circulation


A common bias, or consistent error, found in models collected in the fifth phase of the Climate Model Intercomparison Project (CMIP5) is a shallow mixed layer depth in the Subantarctic zone of the Southern Ocean in what should be part of a zone known as the Deep Mixing Band (DMB). This shallow bias is especially prevalent in the wintertime. Many model biases around the Southern Ocean are interconnected and linked to the bias in the DMB. Researchers funded in part by CPO’s Climate Variability & Predictability (CVP) program hypothesized that the DMB’s shallow bias is due to erros in how the models simulate ocean circulation. In particular, how circulation drives heat and salinity from low-latitudes into the Subantarctic zone. Their study, published in Climate Dynamics, tests this hypothesis using two experiments to investigate how changes in circulation affect the stratification, or layering, of the Southern Ocean. The first experiment operates at a finer resolution than the standard CMIP5 models (i.e. the model can represent smaller-scale properties of ocean currents) while the second experiment operates at the same resolution but simulates currents that match observations. Both experiments improved DMB representation, with the increased deeper mixed layers coming from increased surface heat loss (experiment 1) or salinity advection (experiment 2). Both results also indicate that ocean circulation is a key factor in Southern Ocean mixing. However, while there were improvements, not all of the biases were eliminated. Still, while there is a need for continued research, the authors suggest that a surefire way to improve the simulation of DMB in global ocean models is to utilize a finer model grid resolution. 

Read the study »

More News

Scroll to Top