Evaluating Mesoscale NWP Models Using Kinetic Energy Spectra

TLDR

Observations of free‑atmosphere kinetic‑energy spectra show a k⁻³ scaling at large scales that transitions to a k⁻5/3 dependence at mesoscale. The study defines the effective resolution of NWP models by quantifying departures between model‑derived and observed spectra. WRF simulations test various dissipation mechanisms and assess their impact on effective resolution, including spectrum spin‑up and variability under different synoptic regimes. WRF spectra match observations across scales but decay more rapidly at the highest resolved wavenumbers, revealing energy removal by the model’s dissipation processes.

Abstract

Kinetic energy spectra derived from observations in the free atmosphere possess a wavenumber dependence of k−3 for large scales, characteristic of 2D turbulence, and transition to a k−5/3 dependence in the mesoscale. Kinetic energy spectra computed using mesoscale and experimental near-cloud-scale NWP forecasts from the Weather Research and Forecast (WRF) model are examined, and it is found that the model-derived spectra match the observational spectra well, including the transition. The model spectra decay at the highest resolved wavenumbers compared with observations, indicating energy removal by the model's dissipation mechanisms. This departure from the observed spectra is used to define the model's effective resolution. Various dissipation mechanisms used in NWP models are tested in WRF model simulations to examine the mechanisms' impact on a model's effective resolution. The spinup of the spectra in forecasts is also explored, along with spectra variability in the free atmosphere and in forecasts under different synoptic regimes.