Windstorms along the Western Side of the Washington Cascade Mountains. Part II: Characteristics of Past Events and Three-Dimensional Idealized Simulations

Publisher Website
Google Scholar
Abstract
This paper examines the sensitivity of windstorms along the western Washington Cascades to critical-level height, cross-barrier pressure gradient, crest-level stability, and the magnitude of the cross-barrier flow. A statistical analysis of 55 past events shows that the crest-level (850 mb) flow and cross-barrier sea level pressure gradient together explain 82% of the variance of the observed windstorm severity. Although many windstorm cases had a critical level in the middle troposphere, several cases had easterly flow from the surface to the tropopause or only within the Cascade gaps. Three-dimensional idealized simulations using the Pennsylvania State University–NCAR Mesoscale Model Version 5 at 3-km horizontal resolution show that damaging winds are favored along the Cascade foothills for strong cross- barrier flow cases (>13 m s−1) when an environmental critical level exists below 400 mb. However, even without a critical level, reverse shear or a sharp reduction in static stability above crest level can lead to enhanced mountain wave amplification and strong downslope winds for strong cross-barrier flow events. An environmental critical level is not necessary for windstorm development for cases with weaker cross-barrier flow (−1), which is representative of a majority of the observed windstorm events. For these cases, wave breaking is favored in the lower troposphere, which results in shallower easterly cross-barrier flow and a transition to supercritical flow over the lee slopes. Idealized simulations also show that significant leeside winds (20–30 m s−1) do not develop for cases without initial cross-barrier flow until the pressure difference across the Cascades approaches 15 mb and the depth of the cold air exceeds the height of the major elevated gap in the barrier (Stampede Gap). Lastly, the devastating windstorm of 24 December 1983 (gusts greater than 50 m s−1) was successfully simulated with the idealized setup.

This publication has 0 references indexed in Scilit: