High-resolution 3D numerical simulations of conditionally unstable flows over a ridge

Numerical simulations of conditionally unstable flows impinging on a mesoscale mountain ridge have been performed with an explicitly resolving cloud model in order to investigate the statistically stationary features of the solutions. The simulations are performed in a three-dimensional domain and at high resolution (grid spacing = 250 m) in order to resolve properly cellular-scale features. The environmental conditions are given by an idealized conditionally unstable sounding; even with this simplified atmospheric profile, quite a few external parameters come into play, so only a limited portion of the state space was explored. Numerical solutions were first carried out for different uniform-wind profiles impinging on a bell-shaped ridge 2000m high. In the experiments with weaker environmental wind speeds (U = 2.5 m/s), the cold-air outflow, caused by the evaporative cooling of rain from storm cells, is the main mechanism for cell redevelopment and movement: the outflow produces convective cells near the head of the up- and down-stream density currents, which rapidly propagate far from the ridge, so that no rainfall is produced close to the mountain at later times. For a larger wind speed (U = 10 m/s), the evaporation is effective in generating a cold pool only on the downstream side of the mountain, in a region where the air is unsaturated; for even larger wind speeds (U = 20 m/s), the air beneath the thunderstorm over the mountain remains saturated, so that rain cannot evaporate and cool the sub-cloud layer and thus no cold pools form in this case. Further experiments with different mountain heights h and half-widths a were carried out in order to analyze their effect on the distribution and intensity of precipitation. Finally, the idealized environmental sounding was modified to change the value of the convective available potential energy (CAPE) but to keep the lifted condensation level (LCL) constant. Dimensional analysis reveals that the maximum (nondimensional) rainfall rate mainly depends on h/LCL, h/a and a/U*(CAPE)**1/2, while the Froude number, U/Nh (where N is the buoyancy frequency), turns out to be a less relevant parameter.