Numerical Simulation of Slope and Mountain Flows

Richard T. McNider K.E. Johnson Environmental and Energy Center, The University of Alabama in Huntsville, Huntsville, AL 35899

Search for other papers by Richard T. McNider in
Current site
Google Scholar
PubMed
Close
and
Roger A. Pielke Department of Atmospheric Sciences, Colorado State University, Fort Collins, CO 80523

Search for other papers by Roger A. Pielke in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Early descriptive models of mountain-valley circulations indicated that the mountain flow (i.e., the along-valley axis component out of the valley) is a true three-dimensional phenomenon. According to these descriptions, at night shallow-down slope flows on the valley sidewalls directly driven by temperature deficits near the surface produce a pooling of cool air in the valley. This deep pool of cool air in the valley compared with a much shallower surface inversion over the plains (to which the valley opens) produces a secondary flow (the mountain flow) out of the valley driven by a deep hydrostatic pressure gradient. It is this deep secondary flow which is most important to pollutant transport in deep valleys and which has not been previously investigated in a numerical model.

It is the purpose of this investigation to numerically simulate the above-mentioned secondary circulation using a three-dimensional numerical model. The Colorado State University Hydrostatic Mesoscale Model-a hydrostatic, primitive equation model, forced by a surface energy budget-was utilized for the simulations. Both idealized topography and an actual Colorado valley were used in the investigation to emulate the classic mountain-plain configuration discussed above. Special attention was given to the development of the sidewall slope flows which produce the pooling of air in the valley. Results of the simulations indicated shallow slope flows less than 100 m deep started almost immediately after cooling began and reached their maximum velocity after approximately one hour and then slowly decreased through the night. A high level of turbulence was noted near the top of the slope flows. The turbulence was not only due to shear in the slope flow, but to destabilization of the temperature profile because of thermal advection. Definite pooling of cool air occurred in the valley with cooling occurring at all heights to near ridge tops. The secondary flow out of the valley lagged the development of the slope flows by approximately one hour, but was very deep, filling the whole valley to near ridge height.

Abstract

Early descriptive models of mountain-valley circulations indicated that the mountain flow (i.e., the along-valley axis component out of the valley) is a true three-dimensional phenomenon. According to these descriptions, at night shallow-down slope flows on the valley sidewalls directly driven by temperature deficits near the surface produce a pooling of cool air in the valley. This deep pool of cool air in the valley compared with a much shallower surface inversion over the plains (to which the valley opens) produces a secondary flow (the mountain flow) out of the valley driven by a deep hydrostatic pressure gradient. It is this deep secondary flow which is most important to pollutant transport in deep valleys and which has not been previously investigated in a numerical model.

It is the purpose of this investigation to numerically simulate the above-mentioned secondary circulation using a three-dimensional numerical model. The Colorado State University Hydrostatic Mesoscale Model-a hydrostatic, primitive equation model, forced by a surface energy budget-was utilized for the simulations. Both idealized topography and an actual Colorado valley were used in the investigation to emulate the classic mountain-plain configuration discussed above. Special attention was given to the development of the sidewall slope flows which produce the pooling of air in the valley. Results of the simulations indicated shallow slope flows less than 100 m deep started almost immediately after cooling began and reached their maximum velocity after approximately one hour and then slowly decreased through the night. A high level of turbulence was noted near the top of the slope flows. The turbulence was not only due to shear in the slope flow, but to destabilization of the temperature profile because of thermal advection. Definite pooling of cool air occurred in the valley with cooling occurring at all heights to near ridge tops. The secondary flow out of the valley lagged the development of the slope flows by approximately one hour, but was very deep, filling the whole valley to near ridge height.

Save