Answers are grounded in this paper and cite their sources.
Current operational numerical weather prediction models potentially have sufficient spatial and temporal resolution to resolve much of the inertia‐gravity wave spectrum (e.g. horizontal wavelengths in the region of 100–1000 km typically, and even longer zonal wavelengths in the Tropics). The effects of gravity wave dynamics are usually considered to be subgrid‐scale and parametrizable, either as stationary, mountain waves or travelling, non‐orographic waves. Simple techniques for isolating these waves in both the Met Office and ECMWF forecast models are presented and their associated wave fluxes (including three‐dimensional Eliassen–Palm fluxes) are computed for August 2006. In addition, a limited‐area configuration of the Met Office Unified model with 4 km horizontal resolution is run over a region that includes the southern half of the Andes mountain range. Spectral decompositions of the wave energy and momentum fluxes show a distinct peak in power at horizontal wavelengths of about 400 km, irrespective of whether a 4 or 40 km grid is used, although about one half of the power resides in wavelengths shorter than 200 km. The computed Eliassen–Palm fluxes are shown to be similar to the net parametrized gravity wave drag, although the latter are too widespread and uniform over the oceans, reflecting the absence of geographical source variation. Comparison of the time mean temperature amplitude of the waves in the mid‐stratosphere with estimates derived from limb‐sounder satellite measurements from the HIRDLS project suggests that forecast models are capable of capturing the correct overall strength and distribution of gravity wave activity.
No takes yet. Share an insight, caveat, or question.
Shutts et al. (2011) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: