Research of long-term changes in the coastal territories of the Sea of Azov on the basis of satellite data by the methods of image classification of ArcGIS Pro

Abstract

With the advent of new methods for tracking the Earth’s surface, the priority of the data source for environmental monitoring has shifted in favor of remote sensing (RS) data. The processing of remote sensing data of the territory is the possibility of modern tracking and control of hazardous processes. During the study, by comparing satellite images for different years using various methods of the ArcGIS Pro geoinformation environment, an analysis was made of changes in the coastal zone of the Sea of Azov. Three methods for finding changes in the coastal zone associated with dangerous landslide-abrasion processes were evaluated: detection of changes in the coastal zone by replacing the red channel in an earlier image with the red channel of a later image (Composite Bands method); deep Learning method; compute Change method. The study showed the advantage of the Compute Change method on this sample of first data, which made it possible to assess the impact of landslide and abrasion processes on the coastal zone of the Sea of Azov. The results obtained are consistent with the data of field studies of the coastal zone of the Sea of Azov and the data of environmental control of the Krasnodar territory.