Abstract. It could be shown that the normalised difference Vegetation index (NDVI) can be automatically processed for mountainous terrain characterised by steep slopes and high cloud coverage throughout the year. The quality of the data can be improved if the variables for the atmospheric correction (water vapor, ozone and pressure) are taken from a finer grid resolution and if the actual aerosol optical depth is used. However, even under these improved concitions, the magnitude of the NDVI would differ but the behavior and shape of the graphs in general would remain. Die upward trend of the winter NDVI in the years 1995–1998 is significant for the altitudinal levels 1001–1500 m and 501–1000 m. To be able to find out more about the snow covered areas at these heights, reference was made to the air temperatures of Zürich, Engelberg and Santis. Thus, we were able to show that the increase of NDVI at lower elevations during the wintertime was linked to decreasing snow covered areas. In elevations above 2000 m, a slight increase in the summer NDVI could be detected but the linear trend is not significant. An explanation for this could not be given due to missing meta-data, such as the extent of the snow covered area. In conclusion, it may be said that there is a good correlation between the air temperature and the NDVI, but that the quality of the precipitation database does not allow for a correlation to be made between precipitation and the NDVI.
Extraction of Environment and Resources Information inside a Pixel. A Case Study for Snow Covered Area Extraction from NOAA/AVHRR Data.