Evaluating the Characteristics of Trapped Lee Waves in a High-Resolution Forecast Model using Satellite Imagery

Oral Presentation 

Orographic gravity waves, which are generated by the restoring force of gravity when stably 
stratified atmospheres flow over mountains, may exert frictional forces on the atmosphere and thus 
influence atmospheric circulation. While the effects of these waves may be large-scale, the 
mountains causing them are often of a smaller scale than can be resolved by most weather and 
climate models. Parametrisations are therefore necessary to approximate their effects on 
atmospheric circulation. Vertically propagating orographic gravity waves often are parametrised but 
trapped lee waves – their horizontally propagating counterparts, which occur at low levels and are of 
shorter wavelength – are often not. Before a parametrisation can be designed, however, a study of 
the conditions causing trapped lee waves must be conducted. Because measurements of trapped lee 
waves are scarce, the Met Office’s UK variable resolution model (UKV, 1.5 km grid spacing) could be 
used to investigate the atmospheric conditions associated with trapped lee waves. However, the use 
of this high-resolution model as a substitute for observations requires an evaluation of the model’s 
accuracy.


To this end, the occurrence of trapped lee waves in the UKV is compared to their occurrence in 
geostationary satellite imagery during the year 2023. Continuous wavelet transforms are employed 
to compare the wavelength and orientation of the waves between both sources, showing that the 
UKV manages to accurately reproduce waves of wavelengths above 9 km (~6 times its grid spacing). 
There is very good agreement between the satellite imagery and the UKV on the orientation of the 
waves. In addition, their orientation is shown to be strongly correlated with wind direction. The 
application of linear gravity wave theory to UKV-derived vertical profiles identifies the model’s 
limited vertical resolution as a potential shortcoming in representing the relevant atmospheric 
conditions. Despite this, trapped lee waves are found to be generally well-reproduced in the UKV, 
which suggests a sufficiently accurate representation of the relevant atmospheric conditions in the 
model. It may therefore be considered a suitable substitute for observations in assessing the 
atmospheric conditions associated with trapped lee waves.

Speaker/s