Mapping Land Cover Changes Using Landsat TM: A Case Study of Yamal Ecosystems, Arctic Russia

Polina Lemenkova (1) , Bruce C Forbes , Timo Kumpula

1 OUC - Ocean University of China

This paper details changes in land cover types and vegetation distribution in tundra landscapes during the past two decades. The main method of the work is classification of the Landsat TM scenes for land cover change detection. The new approach of the current work is application of GIS and remote sensing tools for Bovanenkovo region, since there is no previous remote sensing and GIS-based studies performed in the same area focusing research problem of land cover changes. The research area is geographically located on the Bovanenkovo region, the northwestern part of Yamal Peninsula, West Siberia, Russia. The Yamal Peninsula is the world's largest high-latitude wetland system covering in total 900,000 km² of peatlands, since lowland region creates ideal conditions for the development of wetlands, dense lake and river network (Kremenetski et al. 2003). The geomorphology of Yamal Peninsula is flat homogeneous land and low-lying plains with maximal elevations lower than 90 meters (Walker et al. 2009). Such environmental settings of Yamal facilitate seasonal flooding, active erosion processing, permafrost distribution and intensive local landslides formation.The dominating vegetation types on Yamal include different types of shrubs and willows, heath, grasses, moss, and lichens. Changes in land cover types in the Russian North are caused by various reasons. These include multiple ecological and social factors, such as permafrost degradation, reindeer grazing and gas-field development, as well as overall environmental changes, including climate (Walker et al. 2009). One of the factors causing changes in vegetation types in landslide formation. Thus, the early-stage vegetation, such as pioneering mosses or lichens usually follows recent landslide formation, while meadows and willow shrubs with high canopy points indicate later stages of vegetation regeneration after landslide activities. Therefore, distribution of the willow shrubs on bare slopes may indicate that these areas were landslide-affected in the past (Ukraintseva and Leibman, 2007; Leibman and Kizyakov, 2007). Current research focuses on application of remote sensing data and GIS methods for land cover change detection in Bovanenkovo region. Technically, the data processing was performed in ILWIS GIS, using methods of image interpretation and supervised classification applied for Landsat TM scenes (1988 and 2011). The classification results indicate changes in land cover types in Yamal ecosystems, namely, the overall increase in woody plants, such as willows and shrub tundra (e.g. "short shrub tundra", "sparse short shrub tundra" and "dry short shrub tundra"), and slight decrease in grasses, heath and peatland. The main detected trend in these changes demonstrates process of greening in Arctic tundra, which indicates structural variations in ecosystems within the Bovanenkovo district. These changes can be explained by the complex ecological processes as well as anthropogenic influence, caused by Bovanenkovo gas field exploration and its consequences. Methods The research methods consist of the application of the remote sensing and geoinformation tools for land cover studies. The technical implementation of the imagery processing, spatial analysis and GIS mapping was performed in ILWIS GIS. The research data include two Landsat TM scenes for the years 1988 and 2011. The Landsat data were selected due to their suitability for land cover mapping they enable recurrent remote sensing observations, interpretation and assessment of temporal changes in land cover types in the high north. The research methodology includes supervised classification of the Landsat scenes, GIS mapping highlighting changes in land cover types. And calculation of the areas of various land cover types in 1988 and in 2011. The pre-processing of the imagery includes data import and visual settings of colours and contrast. The images were imported to ILWIS from .img into ILWIS raster map format (ASCII) using GDAL and processed. Final maps are displayed showing land cover changes in 1988-2011.

Mots clés: Image processing -- Digital techniques; Image processing Pattern recognition; Image processing Image recognition; Landsat TM 5 TIR; Landsat TM; Landsat TM and ETM data; Landsat TM and ETM images; GIS Climat Environnement Société; GIS Geographic Information System; GIS MAPPING; GIS MODEL; GIS Analysis; GIS

Domaines: Apprentissage [cs.LG]; Traitement des images [eess.IV]; Calcul [stat.CO]; Applications [stat.AP]; Autres [stat.ML]; Statistiques [stat]; Traitement du signal et de l'image [eess.SP]; Environnements Informatiques pour l'Apprentissage Humain; Algorithme et structure de données [cs.DS]; Synthèse d'image et réalité virtuelle [cs.GR]; Recherche d'information [cs.IR]; Planète et Univers [physics]; Sciences de la Terre; Méthodologie [stat.ME]

Polina Lemenkova, Bruce C Forbes, Timo Kumpula. Mapping Land Cover Changes Using Landsat TM: A Case Study of Yamal Ecosystems, Arctic Russia. XI th International Conference on Geoinformatics - Theoretical and Applied Aspects, National Academy of Sciences of Ukraine, May 2012, Kiev, Ukraine. ⟨10.6084/m9.figshare.7434242.v1⟩. ⟨hal-02022108⟩