Comparative Analysis of Reforestation Indicators on Abandoned Agricultural Lands in the Central Russian Forest-Steppe Based on Remote Sensing Data

Abstract

Natural afforestation of abandoned agricultural lands due to postagrogenic successions leads to changes in the vegetation cover of landscapes in the Central Russian forest-steppe. The article presents a comparative analysis of natural afforestation indicators on abandoned lands, calculated using Landsat OLI, Sentinel-2 MSI and MODIS satellite data. The Sentinel-2-derived indicators are the most informative for assessing the forest cover of abandoned lands. For indicators extracted from Sentinel-2 data, the statistical significance of differences between gradations of forest cover of abandoned lands is highest. The indicators of long-term dynamics of the vegetation index, calculated based on MODIS data, are the most informative for comparing intra-zonal differences in the intensity of abandoned lands afforestation. At the same time, the distribution of abandoned lands by forest cover in the physical-geographical subzones is most pronounced in the histograms of SWIR-reflectance derived from Sentinel-2. Differences in the species composition of forests on abandoned agricultural lands most strongly affect the values of Landsat OLI vegetation index. Its values are more sensitive to differences in the species composition of forests on abandoned lands in comparison with the spectral reflectance.