Generation of internal gravity waves by a katabatic wind in an idealized alpine valley

Abstract

Abstract:The dynamics of the atmospheric boundary layer in an alpine valley at night or in winter is dominated by katabatic (or down-slope) flows. As predicted by McNider (1982) oscillations along the slope are expected to occur if the fluid is stably-stratified, as a result of buoyancy and adiabatic cooling/warming effects. Internal gravity waves must also be generated by the katabatic flows because of the stable stratification. The aim of the present paper is to identify and characterize the oscillations in the katabatic flow as well as the internal gravity wave field emitted by this flow. Numerical simulations with the ARPS code are performed for this purpose, for an idealized configuration of the Chamonix valley. We show that the oscillations near the slope are non propagating motions, whose period is well predicted by the single particle model of McNider (1982) and equal to 10 to 11 mn. As for the wave field, its frequency is close to 0.85N, where N is the value of the Brunt-Väisälä frequency in the generation region of the waves, consistently with previous academic studies of wave emission by turbulent motions in a stratified fluid. This leads to a wave period of 7 to 8 mn.

Interactions Between Downslope Flows and a Developing Cold-Air Pool

“…• with the vertical, which agrees with 0.88 < ω ′ /N xz < 0.92 (see Appendix C), and that 2 π/ k z y = λ z y ≈ 1 km, which are very similar to the results of Chemel et al [2009] and Largeron et al [2013], and supports their finding that λ z is set by the depth of the topography. An interesting feature of the flow are the vortices between the regions of upward and downward motions.…”

“…C.2a), which permits the use of a single representative ω ′ . Using a similar model set-up to that used here, Chemel , the frequency with the largest amplitude in the spectrum of the time series, after t = 60 min, of (a) ∂θ v /∂t and (b) w, across an (x, z) slice taken half-way along y. study by Largeron et al [2013] extended the work by Chemel et al [2009] and found 0.8 < ω ′ y /N 0 < 0.9 for a similar location above the valley atmosphere, where y indicates an average across y. These results were found to correspond to IGWs radiated by any turbulent field with no dominant frequency component.…”

On Adding Thermodynamic Damping Mechanisms to Refine Two Classical Models of Katabatic Winds

“…In addition, the analytical wind speed of the Prandtl model is inversely proportional to the ambient buoyancy frequency. The analytical solutions of these layer-averaged models show that buoyancy-type oscillations are possible when the air parcel moving along the slope overshoots its buoyant equilibrium position; these oscillations appear to provide a satisfactory explanation for some fluctuations observed in katabatic winds (Doran and Horst 1981;Stone and Hoard 1989;Helmis and Papadopoulos 1996;Bastin and Drobinski 2005;Chemel et al 2009). Fleagle (1950) took a different approach by formulating a time-dependent, layer-averaged model to analyze katabatic winds in terms of net radiation, friction, and angle of the slope.…”

Turbulent Mechanical Energy Budget in Stably Stratified Baroclinic Flows over Sloping Terrain

“…The presence of the Brunt-Väisäla frequency and the restoring role of the buoyancy force may tempt us to associate the mechanism described above with pulsations occurring in katabatic winds (e.g., McNider 1982;Princevac et al 2008), gravity waves associated with slope winds (e.g., Chemel et al 2009) and intermittency of turbulence caused by breaking gravity waves (e.g., Staquet 2004). Apparently, these are separate mechanisms that may, but need not necessarily, coexist and affect one another, e.g., by driving changes of shear and stress.…”