Use of aerospace methods and image processing methods of remote sensing of the Earth for environmental monitoring of Kakhovka reservoir

摘要

The creation of a cascade of reservoirs on the Dnieper led to the emergence of many environmental problems, in particular, flooding of adjacent lands to reservoirs, shore abrasion, “flowering of water” etc. In recent decades, in the study of environmental problems in the biosphere, aerospace research methods have been increasingly attached importance. The specificity of the application of space imagery and the obtaining of new information with their help is due to their visibility, the possibility of studying the Earth's surface at various levels of generalization. Changes in the water regime negatively affected the oxygen regime of reservoirs, quite often leads to the death of fish. The Kakhovka Reservoir is characterized by fairly high indicators of total solar radiation, which contributes to its overgrowing by higher aquatic vegetation, especially in the shallow upper part, and the “flowering” of water caused by the rapid reproduction of blue-green algae in a large part of the water area (up to 80–95%). The peak of this process, as a rule, falls on July–August. The analysis of the ecological state of the Kakhovka reservoir using remote sensing data of the Earth and methods for their digital processing, in particular, the method of synthesis of the multispectral image (Composite), creation of a mosaic of images, radiometric and atmospheric correction, spectral transformations of pixel values of space images and zonal statistics of ArcGIS are performed in this article. The study was carried out on the basis of the Landsat satellite images for 1977–2017. An analysis of the temperature regime of the Kakhovka Reservoir showed that the highest temperature was recorded in 2010 and allowed to allocate on the space images the sections of the reservoir with the highest temperature. At the same sites, the most active processes of “flowering” of water were detected using the calculated values of vegetation indices. This confirms the efficiency and feasibility of using remote sensing data processing to perform environmental monitoring and assessment of the state of large-scale water bodies, in particular reservoirs.