Monitoring the operational changes in surface reflectances after logging, based on popular indices over Sentinel-2, Landsat-8, and ASTER imageries

Abstract

Revealing the status of forests is important for sustainable forest management. The basis of the concept lies in meeting the needs of future generations and today's generations in the management of forests. The use of remote-sensing (RS) technologies and geographic information systems (GIS) techniques in revealing the current forest structure and in long-term planning of forest areas with multipurpose planning techniques is increasing day by day. Significant technological advances are in allowing programmers to modernize how they manage data. Sentinel-2, which is a relatively new addition to Earth observing satellites, is a new-generation satellite that has enabled classification and monitoring of land cover change with high precision at ease. Visible R, G, B, and near-infrared (NIR) bands have offered exceptional 10-m spatial reasolution, making them suitable for vegetation monitoring along with the additional 20-m bands to spare especially in chlorophyll content analyses. On the contrary, Landsat-8 and ASTER which have been longer lasting in Earth observation were rougher results especially in forestry studies. In this study, Landsat-8 and ASTER satellite images were compared against the Sentinel-2 images as a reference in conjunction with GIS techniques to monitor and assess the impact of various logging procedures, including selective logging and regeneration silviculture. The investigation employed a range of plant vegetation indices, including NDVI, GNDVI, and SAVI, to evaluate the efficacy of image resolution in detecting forest cover changes in the Kastamonu region, where the timber production is the hightest in Turkey. For selective and regeneration activities, satellite images were taken pre-harvesting and immediately post-harvesting, and index maps were produced. NDVI, GNDVI, and SAVI indices were the most accurate indicators of green vegetation change in the Sentinel-2A imagery. Similarly, for the Landsat-8 imagery, the SAVI, NDVI, and GNDVI indices were found to be satisfactory indicators. As for ASTER imagery, the success sequance was like SAVI, GNDVI, and NDVI. Based on the findings of this study, it has been noted that the ASTER imagery closeness to Sentinel-2A was more remarkable in detecting changes in green vegetation in forested areas. The data derived from ASTER imageries demonstrated superior efficacy compared to Landsat-8 in generating forest cover maps, owing to their proximity to those produced by Sentinel-2. The findings also indicated that ASTER imagery, with suitable spatial and spectral resolution, could still be utilized as efficienly as Landsats to generate forest cover density maps and monitor long-term forest conservation practices, particularly in professionally managed forests. Thus, this methodology demonstrated the capacity for efficient worldwide forest management.