How does subgrid‐scale parametrization influence nonlinear spectral energy fluxes in global NWP models?

The paper examines horizontal wind variance (kinetic energy spectra) and available potential energy spectra in simulations conducted with a state‐of‐the‐art global numerical weather prediction (NWP) model: the Integrated Forecasting System (IFS) of the European Centre for Medium‐Range Weather Forecasts. The formulation of the spectral energy budget of the atmosphere by Augier and Lindborg (2013) is used to illustrate how the nonlinear spectral fluxes differ for a hierarchy of reduced models, adiabatic dynamical core, Held‐Suarez dry, and idealized moist aquaplanet simulations, compared to NWP simulations with full complexity. The results identify surface drag and momentum vertical mixing as the key processes for influencing the transfer of energy in a stratified atmosphere. Moreover, the circulation generated by topography plays a significant role in these transfers. Given that subgrid‐scale vertical mixing is parametrized, and that the treatment of orography filtering varies vastly between NWP models, the magnitude and scale of the nonlinear interactions can differ substantially from model to model, and depends on the choices made for the physical parametrizations. The need to appropriately parametrize the essential influence of subgrid‐scale processes in global NWP and climate simulations has the effect that the physical energy cascade is replaced by a parametrized energy transfer. This explains the seeming failure of the IFS to produce a shallower mesoscale energy spectrum. In contrast, neither the horizontal filtering, typically applied in NWP models to avoid a spectral blocking at the smallest scales, nor implicit numerical dissipation significantly constrain, at sufficiently high resolution, the nonlinear interactions or the dominant slope of the energy spectra at synoptic and mesoscales.