Computational models employed for simulations of weather and climate have limited spatial resolution, currently around 2 km for regional weather models and 100 km for global climate models. Such models cannot include processes on smaller scales, including the effects of viscous dissipation on the scale of 0.1 mm or less. Including dissipation effects requires a parametrisation of turbulent energy flows at each grid point, depending on the type of motion and the geographical constraints. Despite the importance of such energy transfers, their distribution as well as their temporal and spatial behaviour is known only crudely. Models may often not even get the correct direction of the energy transfers, from large scales to small, or vice versa.
In a new paper, LML Fellow Davide Faranda and colleagues exploit a local energy budget derived using a weak formulation of the Navier-Stokes equations to characterize the distribution of instantaneous and local sub-grid energy transfers in the atmosphere using 3D velocity fields obtained from reanalysis of historical datasets. They find that these energy transfers are highly correlated with baroclinic eddies occurring at mid-latitudes and also with severe tropical cyclones. Their computations provide the direction of the local energy cascade at any scale in physical space, and yield important information which can be used to interactively adjust the energy fluxes in weather and climate models so they account more accurately for the energy conservation laws in the atmosphere.
The paper is available at https://journals.ametsoc.org/doi/abs/10.1175/JAS-D-17-0114.1