In mountain regions such as Lake Sevan basin, landscape-ecological problems are manifested sharply. Lake Sevan basin lies in the east of the RA （Republic of Armenia） and is characterized by unique natural and econom...In mountain regions such as Lake Sevan basin, landscape-ecological problems are manifested sharply. Lake Sevan basin lies in the east of the RA （Republic of Armenia） and is characterized by unique natural and economic peculiarities. It covers an area of 4,891 km^2, or 16% of the entire territory of the country. The article considers the dynamics of the ecosystems of Lake Sevan basin through remote sensing data. To achieve the stated goal, multi-zonal satellite images Landsat ETM （Enhanced Thematic Mapper Plus）, Landsat TM （Thematic Mapper） and Landsat MSS （Multispectral Scanner） were applied. All satellite imagery data was geometrically and atmospherically corrected. Temporal changes were determined using both a supervised classification approach and NDVI （normalized difference vegetation index） in that time series. Supervised procedure, a hierarchical land cover classification system was used to detect the different land cover classes in the Lake Sevan basin, Armenia. According to this system, four land cover categories exist in this area： （1） water; （2） vegetation-bare; （3） man-altered land; （4） vegetated land, which indicate that for the last 40 years the study area has displayed expansion of man-made landscapes and vegetation-bare sites, ecosystems having lower biomass and reduction of woodlands. NDVI values and the area they covered indicate that between 1973 and 2011, the area of almost non-vegetated lands increased approximately by three times. The main cause of such changes is activation of erosion processes as a result of a climate warming.展开更多
文摘In mountain regions such as Lake Sevan basin, landscape-ecological problems are manifested sharply. Lake Sevan basin lies in the east of the RA （Republic of Armenia） and is characterized by unique natural and economic peculiarities. It covers an area of 4,891 km^2, or 16% of the entire territory of the country. The article considers the dynamics of the ecosystems of Lake Sevan basin through remote sensing data. To achieve the stated goal, multi-zonal satellite images Landsat ETM （Enhanced Thematic Mapper Plus）, Landsat TM （Thematic Mapper） and Landsat MSS （Multispectral Scanner） were applied. All satellite imagery data was geometrically and atmospherically corrected. Temporal changes were determined using both a supervised classification approach and NDVI （normalized difference vegetation index） in that time series. Supervised procedure, a hierarchical land cover classification system was used to detect the different land cover classes in the Lake Sevan basin, Armenia. According to this system, four land cover categories exist in this area： （1） water; （2） vegetation-bare; （3） man-altered land; （4） vegetated land, which indicate that for the last 40 years the study area has displayed expansion of man-made landscapes and vegetation-bare sites, ecosystems having lower biomass and reduction of woodlands. NDVI values and the area they covered indicate that between 1973 and 2011, the area of almost non-vegetated lands increased approximately by three times. The main cause of such changes is activation of erosion processes as a result of a climate warming.
