The objective of this thesis is to investigate the vertical structure of the atmosphere and to assess the performance of several radiosonde- and lidar-based methods to detect the boundary layer height within the Inn Valley. All of these methods are developed for horizontally homogeneous and flat terrain. Radiosonde methods include different versions of the parcel and Richardson number methods and are, therefore, based on vertical profiles of virtual potential temperature, potential temperature, and horizontal wind speed, while lidar methods are based on vertical profiles of velocity variance, dissipation rate of turbulent kinetic energy, and backscatter intensities. The investigation of the performance of these methods as well as the vertical structure is based on data collected during the CROSSINN field campaign in summer 2019 in the lower Inn Valley near Kolsass and mainly focuses on two case studies.
Potential temperature profiles measured by radiosondes during daytime confirm previous studies with regard to the existence of two nearly neutral layers, separated by a region of enhanced stability. The case studies illustrate that the depth of the upper layer is larger than the depth of the lower layer. In both case studies, the elevated near-neutral layer extended above mean crest height. The case studies also indicate that the radiosonde methods and the lidar methods based on turbulence determine mixing heights that can be related to the top of the lower well-mixed layer during convective conditions, while mixing heights derived from backscatter profiles are not determined by unique but rather case-dependent processes. From the radiosondes, it is found that the surface connected mixed layer only extends up to around 300 to 500~m~above ground level on average. Nevertheless, warming occurred above this mixed layer, which is too strong to be explained by large-scale advection alone. This result demonstrates the importance of additional transport and mixing processes besides surface-driven convection in complex terrain.
None of the methods tested in this study can detect the multilayer structure of the atmospheric boundary layer within the valley. While for the radiosonde-based methods this can be solely attributed to methodological simplicity, the lidar methods suffer from the limited range of current boundary-layer Doppler lidars. Due to the low aerosol concentration above the surface connected mixed layer, lidars cannot probe the whole atmospheric boundary layer within complex terrain.