A case study of convectively generated gravity waves over the US

Gravity waves generated by mid-latitude convective storms likely have an effect on the circulation of the middle atmosphere. Using the Weather Research and Forecasting Model (WRF) we model a summer-time squall line over the Great Plains in a three-dimensional, non-linear and non-hydrostatic mesoscale simulation. The model is set up with a high horizontal resolution of 1x1 km and 100 vertical levels extending from the ground up to 10 hPa. As a consequence key storm features such as the spatial structure of precipitation and the associated latent heating field are well resolved.

Previous studies have linked gravity wave properties to characteristics of the latent heating field. The dominant vertical wave length for example has been shown to be about twice the vertical extent of the heating whereas the heating magnitude affects the gravity wave amplitudes. On account of that we use radar precipitation and echo top data to validate the model simulation.

Another important factor that determines how the gravity wave spectrum will be shaped above convection is the background wind shear as damping and critical level filtering occur when the wave approaches a level where the Doppler-shifted phase speed equals the wind speed. The result of this dissipation is a net force on the stratospheric mean flow. The Beres 2004 parameterization uses exactly the heating distribution and magnitude plus the mean winds as input to compute a convective gravity wave source spectrum that can be implemented in general circulation models.

We analyze a total of 11 WRF runs all of which simulate the same storm but differ in their physics parameterizations, causing key features like maximum precipitation rate, storm depth and spatial structure to vary slightly. Comparing their respective wave spectra helps to better understand the wave forcing mechanism, which is a crucial first step towards progress in gravity wave parameterizations.