Observed versus simulated mountain waves over Scandinavia – improvement of vertical winds, energy and momentum fluxes by enhanced model resolution?
Atmospheric Chemistry and Physics
<p><strong>Abstract.</strong> Two mountain wave events, which occurred over northern Scandinavia in December 2013 are analysed by means of airborne observations and global and mesoscale numerical simulations with horizontal mesh sizes of 16, 7.2, 2.4 and 0.8<span class="thinspace"></span>km. During both events westerly cross-mountain flow induced upward-propagating mountain waves with different wave characteristics due to differing atmospheric background conditions. While wave breaking occurred
... e breaking occurred at altitudes between 25 and 30<span class="thinspace"></span>km during the first event due to weak stratospheric winds, waves propagated to altitudes above 30<span class="thinspace"></span>km and interfacial waves formed in the troposphere at a stratospheric intrusion layer during the second event. Global and mesoscale simulations with 16 and 7.2<span class="thinspace"></span>km grid sizes were not able to simulate the amplitudes and wavelengths of the mountain waves correctly due to unresolved mountain peaks. In simulations with 2.4 and 0.8<span class="thinspace"></span>km horizontal resolution, mountain waves with horizontal wavelengths larger than 15<span class="thinspace"></span>km were resolved, but exhibited too small amplitudes and too high energy and momentum fluxes. Simulated fluxes could be reduced by either increasing the vertical model grid resolution or by enhancing turbulent diffusion in the model, which is comparable to an improved representation of small-scale nonlinear wave effects.</p>