Energetically consistent scale-adaptive stochastic and deterministic energy backscatter schemes for an atmospheric model

Abstract

Current climate models still suffer from many biases which are partly due to excessive subgrid-scale dissipation. Here we systematically develop energetically consistent stochastic energy backscatter (SEB) and deterministic energy backscatter (DEB) parametrization schemes. We implement our schemes in a simplified spectral atmospheric General Circulation Model (GCM). It is shown that the SEB scheme performs better than the DEB scheme at low horizontal resolutions (T21 and T31), whereas the performance of both schemes becomes comparable as the resolution increases to T42 when comparing with our reference simulation at T127 resolution. The energy backscatter parametrization schemes improve eddy variability in low-resolution models and correctly capture the dominant mode of zonal-mean zonal wind variability. The autocorrelation time-scale of low-resolution models is also found to be more consistent with the reference simulation when applying the SEB and DEB parametrizations. Our schemes are scale-adaptive and computationally efficient.