In order to get a more homogenous data set in the Alpine region, where dozens of data provider exist, a quality control should be applied prior to its use.

A sophisticated quality control scheme, based on a variational spatial data self consitency check has been applied to the MAP conventional (surface) data. Atmospheric pressure, temperature, humidity and horizontal wind have been treated. The results are available at the MAP database and consist of a graphical time series display of the differences between observed and interpolated values and of a statistical evaluation of the number of gross errors, of the bias and the interquartile range as a function of day time. It can be shown, that much of the characteristics of a station can be easily interpreted, when metadata are available, indicating if a station is on a valley floor, at a slope, on a mountain peak, at the shore of a lake or the sea, in a downtown area or at a rural site, etc. Winds are further sensitive on the height of the sensor above ground level or on the roughness of the surrounding surface. The knowledge of this station characteristics is vital when a comparison between high resolution models and observations is made or when high resolution analysis or diagnostic studies are being carried out. Some prototypical examples will be shown, which highlight the importance of data quality control for mesoscale meteorology.