Towards operational determination of boundary layer height using sodar/RASS soundings and surface heat flux data

A method for the derivation of boundary layer height from measurements of sodar, RASS and sonic anemometer-thermometer data is presented. Datasets of the years 2004 and 2005 measured at the Meteorological Observatory Lindenberg (Germany) are used. The time resolution is 15 min. Special emphasis is laid on air pollution issues where mixing heights shallower than 500 m are important. A difference to the numerous methods already presented in the literature is twofold. Firstly, not only single vertical profiles of measured or derived parameters are used but also bulk information, e.g. histograms and boundary layer evolution over time intervals, is considered. Secondly, the presented method analyses sodar data and confirms or corrects the results by use of temperature profiles from RASS and by sonic surface heat flux data. The results are presented as frequency distributions for the whole period and for the four seasons, stratified by the hour of the day. The additional analysis of temperature profiles and surface heat flux is particularly helpful for the detection of stable boundary layers and increases the number of shallow boundary layers particularly in the evening. The comparison with radiosonde-derived boundary layer heights shows a good agreement, deviations are mostly due to a complicated boundary layer structure. Boundary layer depths derived after a formalism given in the German administrative regulation TA-Luft show - compared to the sodar/RASS-derived values - too many very small or partly very large values. The height range between 100 m and 300 m which is essential for pollutant dispersion issues is nearly missing. Requirements for a future operational use of the method are formulated.