Observations and modeling of breaking waves in the lee of the Medicine Bow Mountains

In January/February 2006 the University of Wyoming King Air (UWKA) research aircraft, instrumented with the Wyoming Cloud Radar (WCR), obtained measurements as part of a field campaign over the Medicine Bow Mountains in SE Wyoming. One of the campaign's objectives was to investigate the formation of orographic precipitation as air passed up and over the mountain barrier. The data collected through repeated passes of the aircraft along the wind and across the barrier was also well suited to investigate the formation of waves. On two occasions during the campaign (26 Jan and 5 Feb) the UWKA and WCR captured observations of rotor/breaking wave events in the valley 12 to 15 km downstream of the highest peaks.

The WCR is a 95 GHz (W-band) radar capable of measuring microphysical properties and Doppler velocity from up to 4 fixed antennas simultaneously from the UWKA. The data presented in this study were collected from three of the beams, one pointed vertically upward, another pointed vertically downward (nadir) and a third pointed roughly 30 degree forward of nadir. Doppler velocity measurements from the two downward pointing antennas are combined to retrieve a two-dimensional cross-section of the velocity field along the vertical plane (curtain) swept by the beams.

Observations from both days are consistent with the formation of a breaking wave near the base of the lee slope. On 26 January, the UWKA made 3 passes over a 37 minute period during which time WCR observations revealed downslope winds in excess of 30 m s-1 within 300 m of the ground, a single breaking wave with vertical winds greater than 12 m s-1 and numerous small scale, apparently coherent eddy structures embedded within the wave and underneath the crest. These “sub-rotors” vary in scale from the largest being several hundred meters to less than 100 meters and thus not fully resolvable through our analyses. The large scale structure varied remarkably little over the observation period, however, time scales for the smaller features are not able to be resolved through these measurements.

On both days there is some evidence of flow separation beneath the wave crest with a possible reversal of the near-surface flow indicated in the retrieved two-dimensional flow field. The case of 26 Jan evolved rapidly, with a wave that apparently propagated upstream with time. Unfortunately the full evolution of the system was not observed on this day because following the third pass, the UWKA returned to base (due in part to the strong turbulence encountered in the last pass). In contrast, the 5 February case appeared much more stationary. However, there still appeared a transition over the course of 1 hour from a breaking wave with some hint of flow reversal near the surface to a more laminar flow throughout the depth of the wave and finally the development of a downstream wave train not evident in the earliest passes on this day.

To aid in the interpretation of the observations, these cases were modeled using a multi-nested fixed domain method employing the NRL Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS). As many as 7 domains are used with a horizontal resolution of 111 m for the innermost domain. This is comparable to the resolution of the dual-Doppler synthesis provided by the WCR measurements. Preliminary results from the 26 Jan case suggest the model is able to capture the large scale features of the wave. Model predicted maximum horizontal wind speeds, wave crest location, wave amplitude, maximum vertical wind, and wavelength all are in reasonable agreement with the observations. The model does not appear to capture the smaller eddies observed within the wave, but these may be described through turbulence sub-grid scale parameterizations.

In this paper, we will present results from our ongoing analysis and modeling effort for both cases. In particular we will evaluate the model's ability to capture features over a variety of scales. Further, we will use the model analysis to explain the conditions that led to the formation of the waves and the differences observed from the two days. In particular, we will investigate the thermodynamic and dynamic environment under which the waves formed and, in the case of 5 Feb, transitioned from breaking wave to trapped lee-wave. Finally, we hope to gain a better understanding of the utility and limitations of using this model to investigate small scale features within an orographically-induced breaking wave.