The sensitivity of supercritical atmospheric boundary-layer flow along a coastal mountain barrier

Authors
Citation
M. Tjernstrom, The sensitivity of supercritical atmospheric boundary-layer flow along a coastal mountain barrier, TELLUS A, 51(5), 1999, pp. 880-901
Citations number
38
Categorie Soggetti
Earth Sciences
Journal title
TELLUS SERIES A-DYNAMIC METEOROLOGY AND OCEANOGRAPHY
ISSN journal
02806495 → ACNP
Volume
51
Issue
5
Year of publication
1999
Pages
880 - 901
Database
ISI
SICI code
0280-6495(199910)51:5<880:TSOSAB>2.0.ZU;2-M
Abstract
Hypothetical numerical simulations are carried out to investigate the sensi tivity of a coastally-trapped supercritical flow to various forcing. All si mulations are compared with a control simulation, that is based on a real c ase from the coastal waves field experiment off the coast of California in 1996, the 7 June case. This case features a northerly supercritical flow al ong the California coast that triggers an expansion fan at Cape Mendocino ( 40.4 degrees N, 124.4 degrees W). The upstream jet then strengthens as the boundary layer becomes more shallow. Other important features are the terra in at the cape and the SST depression into Shelter Cove in the lee of the c ape. All sensitivity simulations reveal an expansion fan with the associate d dynamics; this appears to be a very stable feature. Altering the local te rrain changes the local structure of the jet and removes the lee-wave that is responsible for the apparent collapse of the boundary layer in Shelter C ove. A realistic SST, instead of a spatially constant value, has only minor effects very locally. Changing the Froude number of the flow by manipulati on of the inversion strength reveals an unexpected feedback. A weaker inver sion increases the entrainment and thus deepens the boundary layer, and vic e versa, while increasing the inversion strength. In the expression for the Froude number, these 2 effects cancel and the end results is very similar to the control run. Even when reducing the background flow speed by 50%, an expansion fan appeared. This case is thus transcritical-subcritical upstre am, turning supercritical at the cape. It experiences the largest changes i n terms of wind speed. Attempts to generate a solid stratocumulus layer (no clouds were present in reality) in the model failed. Within a realistic in crease of the initial boundary layer humidity, it was not possible to maint ain clouds in these runs. In summary, the expansion fan dynamics appear str ong and quite consistent, and it was virtually impossible to find a set of conditions, based on perturbations to the control case, that did not reveal supercriticality and the expansion fan dynamics.