A Modeling Study of a Trapped Lee-Wave Event over the Pyrénées

Abstract

Abstract A trapped lee-wave mountain event in the southern part of the Pyrénées area is analyzed using the Weather Research and Forecasting (WRF) Model. Model experiments are designed to address the WRF predictability of such an event and to explore the influence of the model parameters in resolving the mountain waves. The results show that the model is able to capture a trapped lee-wave event using the 1-km horizontal grid model outputs. Different initial conditions, the vertical grid resolution, and the resolved topography lead to changes in the wave field distribution and the wave amplitude meaning that an ensemble of different model settings may be able to quantify the uncertainty of the numerical solutions. However, the model experiments do not significantly change the wavelength of the generated mountain waves, which is shorter in the three-dimensional real simulations than the one derived from satellite imagery. Comparison with observational data from the surface stations and a wind profiler upstream of the mountain range shows that the model underestimates the horizontal wind speed and this can be the reason for the underestimation of the wavelength. In addition, the valley circulations and the formation of a rotor near the surface are explored. The formation of a low-level rotor in the model is intermittent and brief, and it interacts with other flows coming from multiple directions. The first strong wave updraft is located over the valley aligned with the highest mountain peaks and strong vorticity is captured from the surface up to the first wave crest.