|This study investigates the atmospheric boundary layer (ABL) height over complex terrain in the Bavarian Alpine foreland. Data was collected during the field experiment SALSA 2005 at Hohenpeißenberg in August and September of 2005. Sodar/RASS measurements, tethered balloon soundings and turbulence measurements were performed at the foot of the mountain, which overtops the surrounding terrain by about 300 m.
In the first step, thermally induced wind systems were identified. The mesoscale circulation system between the mountains and the foreland, referred to as Alpine pumping, was observed on four consecutive fair weather days. On the smallest scale of thermally induced wind systems, slope winds occurred.
The ABL height was determined for three selected days that were assumed to represent different weather conditions (easterly winds, westerly winds, Alpine pumping). Emphasis was placed on sodar measurements. Visual inspection of reflectivity profiles as well as automatic detection using profiles of reflectivity and the standard deviation of the vertical wind velocity component were applied to determine the ABL height. The applied automatic method performed reasonably well but could not replace the visual inspection method, which turned out to be the most reliable. Sodar reflectivity profiles frequently revealed a complex ABL structure. Tethered balloon measurements for one of the selected days were compared to sodar data which confirmed this complex structure. ABL heights determined with various methods using profiles of meteorological variables from routine radiosonde ascents at three sites in the vicinity turned out to be of only limited value due to their low temporal and vertical resolution. The applicability of parameterisations of the stable boundary layer (SBL) depth and of slab models for the growth of the convective boundary layer (CBL) using surface measurements was investigated in complex terrain. If the SBL top is assumed to be horizontal and the difference in elevation between measurement sites is taken into account, simple parameterisations reasonably agreed with SBL depths from sodar measurements, whereas more complicated parameterisations underestimated the SBL depth. Results from slab models agreed well with measurements for two of the three days, whereas these models largely overestimated the CBL depth for the day when Alpine pumping developed. This overestimation was attributed to the influence of factors such as advection and subsidence that were neglected in the model but could not be neglected under the prevailing conditions. Similar performance was observed for the Lokal-Modell.
The ABL structure was strongly influenced by Alpine pumping with distinct layering as well as suppressed CBL growth. Processes triggered by Alpine pumping such as cold air advection from the lower parts of the Alpine foreland as well as subsidence that replaces boundary layer air which is pumped into the Alps strongly affected the ABL structure.