Friday, 13 July 2012: 11:15 AM
Representation of the Afternoon Transition in Numerical Weather Prediction Models: Evaluation with BLLAST Data
Essex Center (Westin Copley Place)
The Boundary Layer Late-Afternoon and Sunset Turbulence (BLLAST) field campaign took place in June and July 2011 in southern France. It provides a new dataset combining measurements from various platforms in order to simultaneously document the temporal and spatial evolution of the Planetary Boundary Layer (PBL) during this transition. The present study evaluates two numerical weather prediction (NWP) models (AROME and ARPEGE), focusing on the representation of the vertical profiles of temperature and humidity and the time evolution of near surface atmospheric variables as well as the radiative and turbulent fluxes. ARPEGE is a global model with a stretched resolution of about 10 km over France with a 4D var assimilation system. AROME is a limited area non-hydrostatic model with a higher resolution of 2.5 km that is forced laterally by the ARPEGE analysis. Some comparisons to ECMWF operational outputs are also shown. A special emphasis is given on IOP days for which high-frequency soundings were acquired (about 1h time interval from 1300 to 2000 UTC). In general, the models exhibit a cold bias, the ARPEGE model also indicates a dry bias. The high-resolution model AROME resolves the vertical structures better, in particular the strong inversion during the day and the thin stable boundary layer during the evening. This model is also capable to capture typically observed features, as subsidence and a well-defined maximum in water vapor mixing ratio in the upper part of the residual layer that occurs during the evening and that is linked to mesoscale advection. The representation of the various surface characteristics by the NWPs is also evaluated, showing that at least part of the observed variability is reproduced. Even the global model satisfactorily represents higher sensible heat fluxes over forest. The properties of the observed boundary layers are linked to local surface fluxes, but they also involve mesoscale (orographic) atmospheric circulations which are only captured by the finer resolution models.