Remote monitoring of natural afforestation processes on abandoned agricultural lands is crucial for assessments and predictions of forest cover dynamics, biodiversity, ecosystem functions and services. In this work, we built on the general approach of combining satellite and field data for forest mapping and developed a simple and robust method for afforestation dynamics assessment. This method is based on Landsat imagery and index-based thresholding and specifically targets suitability for limited field data. We demonstrated method’s details and performance by conducting a case study for two bordering districts of Rudnya (Smolensk region, Russia) and Liozno (Vitebsk region, Belarus). This study area was selected because of the striking differences in the development of the agrarian sectors of these countries during the post-Soviet period (1991-present day). We used Landsat data to generate a consistent time series of five-year cloud-free multispectral composite images for the 1985–2020 period via the Google Earth Engine. Three spectral indices, each specifically designed for either forest, water or bare soil identification, were used for forest cover and arable land mapping. Threshold values for indices classification were both determined and verified based on field data and additional samples obtained by visual interpretation of very high-resolution satellite imagery. The developed approach was applied over the full Landsat time series to quantify 35-year afforestation dynamics over the study area. About 32% of initial arable lands and grasslands in the Russian district were afforested by the end of considered period, while the agricultural lands in Belarus’ district decreased only by around 5%. Obtained results are in the good agreement with the previous studies dedicated to the agricultural lands abandonment in the Eastern Europe region. The proposed method could be further developed into a general universally applicable technique for forest cover mapping in different growing conditions at local and regional spatial levels.

The capabilities of Sentinel-2 optical multispectral satellite data for modeling nitrogen (N) and carbon (C) contents, their ratio (C : N), and stocks in the litter horizons of forest soils were assessed. The study was conducted in the Bryansk Forest Nature Reserve and its buffer zone. The organic horizon samples were taken on 33 plots selected with due account for the tree species diversity of the reserve’s forests. Two layers of the organic horizon—L and FH—were sampled separately. The main variables for geospatial modeling were derived from a time series of eight Sentinel-2 multi-seasonal satellite images. Basic terrain characteristics and pixel coordinates were also added to variables’ stack. We used random forest to build regression models and the corresponding standard methods to assess their performance. The best results were obtained for the C : N ratio: the coefficient of determination R² = 0.71 with a scaled root-mean-square error RMSE = 12.5% in the L layer, and R² = 0.83 with RMSE = 10.6% in the FH layer. For other models, the values of R² ranged from 0.23 to 0.61, and the RMSE ranged from 15.8 to 48.6% with the least reliable results for the N and C stocks. Satellite-based variables were most informative for the contents of N and C, and, notably, for the C : N ratio. The most significant periods in the time series were early spring, summer, and snowy winter. To conclude, Sentinel-2 satellite imagery can be successfully used for estimation and mapping the contents and stocks of N and C in the forest soil organic horizon as a free and relevant alternative to thematic data on the species composition and related properties of stands.

The Caucasus Mountains are recognized as a global center of biodiversity but currently face significant risks of degradation due to intensified economic development and the effects of climate change. Forest inventory and mapping are essential for biodiversity conservation in the Caucasus region. Geospatial modeling is a common method of thematic mapping, but its reliability depends heavily on the initial classification of reference data used for model training. Modern vegetation science features various classification approaches, most of which were developed independently of digital mapping practices and are rarely assessed for their suitability in geospatial modeling. To fill this gap, we classified the same dataset of vegetation relevés from mixed and broadleaf forests in the northwestern Caucasus using two approaches, based on floristic and dominant concepts, and compared the predictive performance of geospatial models trained on these datasets. We considered multiple types of geospatial variables, including optical satellite imagery, a digital elevation model (DEM), and bioclimatic and soil features, to evaluate their informativeness for spatial differentiation of the resulting forest types and to identify optimal variable combinations for modeling via multistage feature selection. We trained several models using different variable sets and machine learning methods for both classifications and evaluated their accuracy via nested cross-validation. The forest types produced by the two approaches scarcely matched, and the selected variable sets for model training differed accordingly. Unexpectedly, bioclimatic and soil variables were more effective than DEM- and satellite-derived variables, despite their coarser spatial resolution. Floristic-based geospatial models outperformed dominant-based models in terms of forest-type separability and predictive accuracy. Therefore, a floristic classification approach may be preferable for forests with complex species composition, both ecologically and in terms of the reliability of geospatial modeling and the derived mapping results.

Geospatial modeling based on a combination of remote sensing data and machine learning methods currently is a common way for thematic mapping of forest ecosystems. However, the reliability of such mapping strongly depends on the initial classification of the reference data used for model training. In modern vegetation science, there are different classification approaches that coexist, but most of them were originally developed regardless of digital mapping practices, and were rarely analyzed on comparative suitability and benefits for geospatial modelling tasks. To fill this gap, we compared the performance of geospatial models for mixed and broadleaf forest types, provided by two different classification approaches, based on floristic and dominant concepts. Both classifications were applied on the same dataset of field plots, collected in the North-Western Caucasus region. We considered geospatial variables of different types, including optical satellite imagery, DEM and its derivatives, bioclimatic and soil features, to analyze their informativeness, and to obtain their optimal combinations via multistage feature selection. We trained the set of models with different variables’ sets and machine learning methods (Random Forest, CatBoost, LDA, and kNN) for both classifications, and evaluated their accuracy via nested cross-validation. As a result, forest types, obtained by different approaches, had very little matching, just like the compositions of the respective variables’ sets, selected for models’ training. Bioclimatic and soil variables were more effectie, than DEM-based and optical satellite-based ones, despite their coarser spatial resolution. Floristic-based geospatial models clearly outperformed the dominant ones in terms of forest types’ separability and potential predictive accuracy. Therefore, a floristic classification approach may be preferable for forests with complex species composition not only in terms of common ecological sense, but also in terms of reliability of geospatial modelling and its derivative mapping results.

Marmots weren’t part of the ecosystems of Bryansk regions for a long time, because they disappeared under the pressure of the plow and uncontrolled hunting. In this area, the first colony of marmot was noticed in 2013 on the gentle slopes of the southern exposure. The aim of the work is to estimate the role of Marmota bobak activity for intraforest steppe meadows in the east of the Bryansk region. Route survey showed that plants communities transformed under their activity. Based on geobotanical relevés and statistical analysis, two groups of vegetation were identified in the study area: short-grass and tall-grass meadows. Tall-grass meadows are characterized by low biodiversity indices, because highly competitive grasses dominate the community (Bromus inermis, Calamagrostis epigeios, Elytrigia repens etc.). Small-grass meadows are associated with marmot settlements and are sustained by animal activity. Due to the grazing of dominant grasses, less competitive species can occupy more advantageous positions in the grass stand. The activity of marmots allows plant species of different ecological-coenotic groups to coexist in the meadows. To maintain structural and species diversity of the vegetation cover of intraforest meadows, the following features of marmots’ behaviour are important: formation of burrows with fresh throwings, which are characterized by an exposed substrate necessary for seed and vegetative propagation of plants; creation and maintenance of pasture or forage areas near the burrows. The above determines the development of vegetation patches (micro-groupings) in meadows and their spatial redistribution in communities.

The research shows the possibility of identifying a key biotope of boreal floodplain meadows in the Smolenskoe Poozerye National Park on the basis of spectral and topographic characteristics of the territory. The key biotope E3.47 according to EUNIS is a meadow along large rivers with areas of calm flow; it freeze every winter and affected by flooding in spring. This category includes areas that are open or poorly overgrown with woody and shrubby vegetation. Sentinel-2 multispectral satellite images (ESA Sentinel-2, 2019) are used to determine areas of floodplain meadows. The geospatial assessment of boreal floodplain meadows is carried out using the method of uncontrolled classification of “K-means” of ten spectral channels of Sentinel-2 satellite images obtained over six seasons, including the flood period. As a result of cluster analysis, classes that are spectrally close to meadow communities are selected. These classes located along flood-plains of rivers, with the elevation difference from the water line not exceeding 5 m. This elevation corresponds to the maximum possible rise of water in rivers during a flood for the studied area. A topographic index using a digital elevation model (DEM) and a hydrographic terrain network are used to correct the selected classes. The accuracy of recognition of the floodplain meadows class is determined using a matrix of error in 133 segments along the 4 transects located in floodplain and dry meadow communities. The convergence of result between the floodplain meadows classification and ground surveys data was 75% (Cohen’s kappa equals 0.53).