Sensitivity analysis of WRF model PBL schemes in simulating boundary-layer variables in southern Italy: An experimental campaign

The sensitivity of boundary layer variables to five (two non-local and three local) planetary boundary-layer (PBL) parameterization schemes, available in the Weather Research and Forecasting (WRF) mesoscale meteorological model, is evaluated in an experimental site in Calabria region (southern Italy), in an area characterized by a complex orography near the sea. Results of 1km×1km grid spacing simulations are compared with the data collected during a measurement campaign in summer 2009, considering hourly model outputs.Measurements from several instruments are taken into account for the performance evaluation: near surface variables (2m temperature and relative humidity, downward shortwave radiation, 10m wind speed and direction) from a surface station and a meteorological mast; vertical wind profiles from Lidar and Sodar; also, the aerosol backscattering from a ceilometer to estimate the PBL height.Results covering the whole measurement campaign show a cold and moist bias near the surface, mostly during daytime, for all schemes, as well as an overestimation of the downward shortwave radiation and wind speed. Wind speed and direction are also verified at vertical levels above the surface, where the model uncertainties are, usually, smaller than at the surface. A general anticlockwise rotation of the simulated flow with height is found at all levels. The mixing height is overestimated by all schemes and a possible role of the simulated sensible heat fluxes for this mismatching is investigated. On a single-case basis, significantly better results are obtained when the atmospheric conditions near the measurement site are dominated by synoptic forcing rather than by local circulations.From this study, it follows that the two first order non-local schemes, ACM2 and YSU, are the schemes with the best performance in representing parameters near the surface and in the boundary layer during the analyzed campaign.