Evaluating surface eddy properties in coupled climate simulations with ‘eddy-present’ and ‘eddy-rich’ ocean resolution

Abstract As climate models move towards higher resolution, their ocean components are now able to explicitly resolve mesoscale eddies. High resolution for ocean models is roughly classified into eddy-present (EP, ∼ 1/4°) and eddy-rich (ER, ∼ 1/12°) resolution. The cost-benefit of ER resolution over EP resolution remains debated. To inform this discussion, we quantify and compare the surface properties of coherent mesoscale eddies in high-resolution versions of the HadGEM3-GC3.1 coupled climate model, using an eddy tracking algorithm. The modelled properties are compared to altimeter observations. Relative to EP, ER resolution simulates more (+60%) and longer-lasting (+23%) eddies, in better agreement with observations. The representation of eddies in Western Boundary Currents (WBC) and the Southern Ocean compares well with observations at both resolutions. However a common deficiency in the models is the low eddy population in subtropical gyre interiors, which reflects model biases at the Eastern Boundary Upwelling Systems and at the Indonesian outflow, where most of these eddies are generated in observations. Despite a grid spacing larger than the Rossby radius of deformation at high-latitudes, EP resolution does allow for eddy growth in these regions, although at a lower rate than seen in observations and ER resolution. A key finding of our analysis is the large differences in eddy size across the two resolutions and observations: the median speed-based radius increases from 14 km at ER resolution to 32 km at EP resolution, compared with 48 km in observations. It is likely that observed radii are biased high by the effective resolution of the gridded altimeter dataset due to post-processing. Our results highlight the limitations of the altimeter products and the required caution when employed for understanding eddy dynamics and developing eddy parameterizations.