Advancing eddy parameterizations: Dynamic energy backscatter and the role of subgrid advection and stochastic forcing

Modeling oceanic eddies requires incorporating physical processes through additional equations. While the overall understanding of the ocean is clear, the models tend to lose too much kinetic energy, resulting in systematic errors. Our goal in this study is to explore how to prevent false energy loss by sending the energy back to where it originated via dynamic backscatter. Our research shows that by adding an advection and a random element, the current method can better capture the turbulent nature of the flow. We tested the implementation on the channel and the double-gyre setups and observed an increase in eddy activity and an improvement in flow characteristics.

The FESOM2 model output data is uploaded here. The names of the data files contain the information about the diagnostic type, resolution (20km=eddy-permitting, 10km=eddy-resolving) and type of backscatter setup (with/without backscatter, with/without advection, with/without stochastic component), and the general type of the setup (DG=double-gyre, Channel=channel setup).