Assessing aquatic enviromnemt quality of the urban water bodies by system analysis methods based on integrating remote sensing data

Abstract

The work presents the comprehensive methodology for assessment of the state of the urban aquatic environment such as Lakes Opechen, Verbne, and Redkyne in Kyiv using the methods of system analysis. The methodology includes structural-textural analysis of the satellite images and the method based on statistical criteria. The spectral-texture analysis of the satellite images was used to get input information for remote assessment of reservoirs as index images: Normalized Difference Pond Index (NDPI), Normalized Difference Turbidity Index (NDTI), and Normalized Difference Algae Index (NDAI) computed from the Sentinel-2. The surface temperature distribution was estimated from the Landsat 8. The method based on statistical criteria is used for a detailed assessment of the aquatic environment using the obtained indexed images and the corresponding cartographic representation of the water quality. The probabilistic and statistical approaches were used to present the statistical criterion for recognizing classes of objects based on the results of measuring their informative features. These approaches are used to solve optimization problems in statistical theories of identification and recognition. This method allowed the cartographic representing of the change in the water quality and aquatic ecosystem in accordance with the reference areas of the state of the reservoir in 2017.