Abstract

Purpose. The main goal of the research was development and testing of the methodology of automated processing and analysis of multispectral satellite images of medium spatial resolution for the detection of deforestation. Design / Method / Approach. Experimental testing of the proposed methodology was carried out in the south-west of Ivano-Frankivsk and the south-east of Zakarpattia Oblasts using free archival images of the visible and IR ranges from Sentinel-2A/B satellites. Findings. The results obtained during the conducted research confirmed the possibility and high efficiency of using images from Sentinel-2A/B satellites to detect and assess the dynamics of felling of wild and protected forests. In only one test area, the area of identified fellings in 2 years amounted to more than 300 hectares. Theoretical Implications. The paper shows the advantages of using the normalized difference index of time changes NDI for a given pair of satellite images of different times and selected spectral channels. Thanks to NDI, the accuracy and stability of the results of the detection of temporal changes increases, as well as the influence of such factors as differences in the illumination of different time images, the presence of scattered clouds, aerosol haze, absorption of radiation by the atmosphere, etc., is reduced. Practical Implications. Due to the high degree of automation, the developed technique can be implemented as a software in the form of a geo-informational web service that can function in the interests of a wide range of public services. Originality / Value. During the study, the necessary stages and sequence of processing remote sensing data were determined, and methods, algorithms and software tools for processing images were selected, and proprietary algorithms and programs were developed to increase the degree of automation and efficiency of image processing, as well as to improve the user interface. Research Limitations / Future Research. This study was limited to the use of multispectral imagery from the Sentinel-2A/B satellites (MSI imager). Future studies involve the use of additional data from Landsat-7 (ETM+ imager), Landsat-8 (OLI imager), Landsat-9 (OLI-2 imager) and Terra (ASTER imager). Paper Type. Applied research, methodological article.