Challenges in Improving the Representation of Mesoscale Kinetic Energy in NWP Models

A necessary condition for improving the prediction of high-impact weather events is for the prediction model to correctly represent the mesoscale kinetic energy spectrum. To this end many modelers compare their models’ mesoscale KE spectral slope with the -5/3 slope reported in Nastrom and Gage’s 1986 study of upper tropospheric aircraft observations. But is this the “true” slope? We have revisited this issue in light of the vastly greater number of observations that have been analyzed since 1986. Specifically, we have examined the 200-hPa KE spectra in several high-resolution global reanalysis datasets, including NCEP GFS (resolution T1534 and T254), ERA-Interim (T255), ERA5 (T639), and JRA-55 (T319). We find that the mesoscale portions of the global spectra are highly mutually inconsistent. This is primarily because the global mesoscale KE has a large contribution from the KE in convective regions, which differs greatly among the various reanalyses.

In addition to this lack of consensus on the slope of the “true” spectrum, a model’s simulated spectrum and its slope also has some sensitivity to its physical parameterizations and dynamical horizontal diffusion. We assessed this sensitivity in two ways: 1) by stochastically perturbing the physical tendencies and 2) by decreasing the hyper-viscosity coefficient, in large ensembles of 10-day forecasts made using the NCEP GFS (T254) model. Both changes increased the mesoscale KE and decreased the steep spectral slope. The effect was stronger in the modified hyper-viscosity experiment. We conclude that improvement in the representation of the mesoscale KE spectrum in NWP models is challenged by the facts that (a) we do not really know the “true” spectrum, and (b) the models’ KE spectra are sensitive to and can be manipulated by perturbing the physics and tuning the horizontal diffusion. Increasing model resolution alone cannot adequately address these challenges.