Snow Cover Mapping with NOAA-AVHRR images in the Scope of an Environmental GIS Project for the Russian Altai (South Siberia)
Erik Hoeppner and Nikolas Prechtel (Nikolas.Prechtel@mailbox.tu-dresden.de)
Presented at Mt. Hood by Sven Etzold (email@example.com)
Aerosensing, Oberpfaffenhofen (Germany)
Institute for Cartography
Dresden University of Technology
01062 Dresden, Germany
An efficient provision of actual, consistent, and reliable geo-information for modelling and map production is a prerequisite of up-to-date environmental protection and ecological planning. This should be accepted not only for the western world, where primary topographic and thematic sources are mostly more transparent and by far easier accessible. For about 6 years, a cooperative Russian-German project is on the way to collect geodata, to set-up an integrative GIS-base, and to derive high-quality maps for the Central Altai Mountains in Siberia, an area with a high environmental status, reflected by Russian national authorities as well as by UNESCO's dedication leading to World Nature Heritage Areas. The principal actors in the project are the Geographic Faculty of Altai-State University, Barnaul, and the Institute for Cartography of Dresden University of Technology.
In the present paper, the general GIS strategy and the actual state of our efforts will be summarised. Then, imbedded in the given project scope, the work package 'Seasonal Snow Cover Mapping for the Russian Altai' (funded by a NATO collaborative linkage grant) will be discussed. The main information was taken from AVHRR satellite images, which have been classified for the years 1997 and 1998. Principles of data selection, and techniques for geometric and radiometric pre-processing will be explained, while the following striking points should be named explicitly: criteria to extract suitable AVHRR scenes (especially orbit, view angle on the target area, and cloudiness), rectification and geocoding of low-resolution imagery, classification using a hierarchical snow assignment scheme, extrapolation for no-data islands, transfer of results to GIS vector coverages, and map design.
The quality assessment is still preliminary. A more precise evaluation is currently planned for the inner study area - the Katoon Range - using high-resolution satellite images and snow pattern analysis.