Investigation of Landsat satellite image change detection of snow and ice cover: A seasonal and multi annual time scale approach to evaluate this technique as a tool for water resource management

Abstract

94 Landsat Satellite images were analyzed regarding snow cover and perennial snow or ice patches in the mountains close to Oppdal. Special attention was paid to the ablation period. The Landsat scenes were processed and analyzed with ERDAS Imagine. Detailed analysis was conducted in ESRI ArcGIS.

The study aimed to analyze the temporal development of snow cover distribution in the study area. Of particular interest were perennial snow or ice patches as they might indicate climate change in the study area and because of their little researched character. The gathered infor-mation regarding the snow cover might be used to adjust existing management tools and strat-egies.

The snow cover was detected with the normalized differential snow index (NDSI) and a fixed threshold. It was a goal to develop a robust process that had the potential to be applied auto-matically. For detailed analysis the binary snow cover map was further used in GIS. It was applied together with a digital elevation model (DEM) to increase the level of information.

Landsat proved to be suitable to map the snow cover extent in the study area. The NDSI proved its reliability and it is possible to use it in an automated process.

The analysis with change detection techniques revealed large losses in area covered by peren-nial snow patches. The average loss for selected snow patches was 55%. Snow patches are usually located on the east facing slope of mountains, with variations towards north- or south-east. Finally snow cover is disintegrating in a similar pattern each year.

As a recommendation for water resource management in the area two things must be men-tioned: climate change is altering the snow cover towards the end of the ablation season and the snow cover is disintegrating in similar patterns. Therefore this information can be incorpo-rated in existing management tools.

Keywords: Natural resource management; remote sensing; normalized differential snow in-dex, NDSI, Landsat, Oppdal, snow patch, ice patch, snow cover detection, change detection