P1.18
Climate of the Mountain Top Station "Sonnblick" in Comparison with Radiosonde Data
Matthias Ratheiser, University of Vienna, Wien, Austria; and S. Schneider, C. Häberli, R. Steinacker, W. Pöttschacher, and W. Gepp
The Sonnblick observatory in the European Alps (47°03’ N, 12°57’ E) with a height of 3105 meters above sea level provides one of the longest time series of meteorological parameters on a mountain top station worldwide (more than 100 years). Parts of this unique dataset are used to compare mountain top measurements of temperature, mixing ratio and wind direction with those in the free atmosphere. For this purpose measurements of about 30 radiosonde stations in Central Europe are interpolated to the mountain station’s position. Interpolation is done for a 25 year period with the objective mesoscale analysis tool VERA, a method similar to the thin plate spline algorithm.
Time series of the difference between the interpolated and the measured values are compared. The radiosonde data have been quality-checked for gross errors before (within the project CALRAS). Therefore it may be assumed that they are “error-free”. Thus we can focus on checking the Sonnblick data for gross and systematic errors.
There are two innovative steps regarding the homogenisation of long time series of mountain top stations on the one hand and to the investigation of differences between values in the free atmosphere and on mountain tops on the other hand: First we do not only compare monthly mean values but also take a look on short term variations due to the current synoptic situation or local influences on the mountain station. Second the values are compared to a representative spatial interpolation of the free atmosphere data. Until now there have only been comparisons with close by mountain stations or a single radiosonde. Thus it is possible to take a closer look at the climate on mountain top stations at a very small time scale (three hourly measurements). Besides inhomogeneous time series (due to instrument deficiencies) may be corrected using this technique.
Poster Session 1, PBL Processes and Modeling (with Coffee Break)
Monday, 17 June 2002, 2:45 PM-4:15 PM
Previous paper Next paper