Successional pathways after peatland draining: remote sensing and predictive modelling of landscape dynamics

Successional pathways after peatland draining: remote sensing and predictive modelling of landscape dynamics | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Successional pathways after peatland draining: remote sensing and predictive modelling of landscape dynamics Vladimir Ivanov, Evgeniya Soldatova, Milyaev Ivan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7012873/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Peatlands are key contributors to carbon storage and hydrological regulation but their role and ecosystem functions and services have been substantially altered by anthropogenic interference, primarily through drainage and peat extraction. This study focuses on the Tarmanskoe peatland in Western Siberia, where large areas were drained for peat extraction and agricultural use from the 1960s to 1970s. Using Landsat satellite imagery from 1984 to 2024 - complemented by high-resolution Unmanned Aerial Vehicle (UAV) data - we applied object-based classification (Random Forest) to assess historical land-cover changes. We then employed a hybrid CA-Markov (Cellular Automata-Marcov) model to project future landscape transformations over the next three decades (2034–2054). Results indicate that formerly drained peatlands followed two main successional pathways: an initial phase of meadow formation with varying levels of waterlogging, followed by a gradual expansion of mixed forests. By 2024, about half of the drained peatland areas transitioned from meadows to forest cover, suggesting a dominant trend toward forest succession. Simultaneously, lakes in the region underwent significant water losses - nearly a 50% reduction in total area since 2013 - driven by natural aging processes, drainage-induced lowering of water levels, and rising mean annual temperatures. The CA-Markov projections reveal a continued, albeit slower, increase in forested areas and a further reduction in water bodies, reaching only 17.4% of their 1984 extent by 2054. These findings underscore the lasting ecological impacts of drainage and peat extraction, as evidenced by spatially heterogeneous successional processes and widespread fragmentation of ecosystems. They also highlight emerging risks, including further water-level declines, increased fire hazard, and ongoing landscape fragmentation. From a conservation perspective, proactive management and the restoration of hydrological functions in disturbed peatlands may help mitigate long-term ecological and climate-related vulnerabilities. Peatland Land-Cover Change Remote Sensing CA-Markov Model Succession Peat Extraction Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 1. Introduction Wetlands are valuable ecosystems that perform multiple essential ecological functions. Their widespread geographic distribution and diversity make them habitats for a vast number of plant and animal species (Ma et al., 2022 ; Ritson et al., 2021 ). The geomorphological structure and organic composition of wetland soils enable them to store large volumes of water, preventing erosion and positively influencing the frequency and intensity of floods (Åhlén et al., 2022 ; McLaughlin and Cohen, 2013 ). The vegetation, structure, and microbiological activity of wetlands contribute to their ability to retain and absorb pollutants from water. Additionally, their role in climate changerelated processes is noteworthy. Peatlands cover approximately 3% of the Earth’s land surface but store one-third of the planets soil carbon and exert an overall cooling effect on the atmosphere, at least at a local level (Taillardat et al., 2020 ; Tassi et al., 2021 ). On the other hand, wetlands also have significant potential as sources of carbon dioxide, methane, nitrous oxide, and dissolved organic compounds (Hugelius et al., 2020 ; Leng et al., 2018 ; Loisel et al., 2020 ). The risk of wetlands transitioning from carbon sinks to emitters of climate-active gases increases with the intensification of anthropogenic factors, such as land-use changes and the growing incidence of landscape fires (Leifeld et al., 2019 ; Mickler, 2021 ; Turetsky et al., 2011 ). An increasing number of studies examining the effects of land-use changes on ecosystems have highlighted that landscape recovery dynamics, plant community biodiversity (Gomes et al., 2020 ; Luan and Liu, 2022 ; Malek et al., 2019 ) and ecosystem services (Gomes et al., 2021 ; Luan and Liu, 2022 ), including carbon sequestration and storage (Kurganova et al., 2015 ; McDaniel et al., 2019 ), are significantly influenced by the nature and intensity of previous land use (Fu et al., 2015 ; Provost et al., 2020 ). Complementing these findings with recent analyses of contemporary and paleoecological data that identify peatlands as crucial determinants of the future global climate, the detailed study of peatland land-cover and land-use dynamics and their impact on ecosystem services has become more relevant than ever (Edvardsson et al., 2022 ; Zhao and Zhuang, 2023 ). The obtained data will provide a foundation for developing policies for the management and conservation of peatlands, aimed at mitigating the effects of climate change in both natural and previously industrially and agriculturally utilized areas (Chen et al., 2022 ; Humpenöder et al., 2020 ; Rappaport et al., 2015 ). Moreover, given the challenges in accurately assessing the extent of degraded peatlands, current data remain approximate (Sirin et al., 2021 ), while precise estimates of the area and current condition of disturbed lands are either fragmented or entirely absent-this is particularly true for the southern part of Western Siberia. Over the past 200 years, large-scale land-use changes have occurred in northeastern Eurasia, affecting various ecosystems, including peatlands (Monier et al., 2017 ; Malek et al., 2019 ). From the extensive settlement-driven land cultivation in the 19th century to the collectivization and prolonged extensive agricultural practices of the 20th century, followed by the collapse of the Soviet Union and the large-scale abandonment of developed lands in the late 20th and early 21st centuries, vast territories have undergone profound anthropogenic transformation (Lapytskaya, 2020 ; Lysenko, 2021 ; Rada et al., 2020 ; Shcherbakova, 2021 ). Drainage reclamation of wetlands was frequently employed to expand agricultural lands (Baisheva et al., 2022 ; Bruisch, 2020 ; Maslov and Maslova, 2020 ). However, plowing was not the sole reason for wetland drainage. Since many thermal power plants operated on peat, extensive peatlands were drained for extraction and use as fuel for electricity and heat production (Akhmeteva et al., 2019 ; Bruisch, 2020 ; Moskalenko et al., 2020 ). The subsequent development of extraction and transportation technologies in the oil and gas sector rendered natural gas a more economically viable energy source, leading to a gradual decline in peat extraction (Bondar, 2022 ; Karpov, 2010 ). Peatland reclamation, primarily for agricultural purposes, was conducted in the Soviet Union until around 1990 but has since ceased (Minaeva and Sirin, 2009 ; Sirin et al., 2021 ), resulting in most disturbed peatlands undergoing spontaneous succession. In Western Siberia, agriculture has historically been highly dependent on state support, leading to a sharp response to socio-political changes (Nekrich and Lyuri, 2019 ; Nguyen et al., 2018 ). Peat was widely used both as fertilizer and as an energy source, leading to widespread peat extraction sites, most of which were abandoned without proper reclamation. Consequently, large areas of former agricultural lands interwoven with peat extraction sites continue to form fragmented ecosystems under the influence of natural, anthropogenic, and post-agricultural factors, exhibiting diverse successional pathways (Lvov, 1995 ; Filippova et al., 2023 ; Krasnov et al., 2005 ). The situation is further complicated by the expansion of urban peripheries and the emergence of new settlements on lands previously used for agriculture or peat extraction. The Tarmanskoe peatland, located west of Tyumen and partially within the city limits, exemplifies a site where all these processes manifest with varying intensity. According to detailed surveys, the exploration of the Tarmanskoe peat deposit was the largest in the history of Soviet peat industry development. Drainage of the Tarmanskoe wetland began in the 1960s and continued until the 1970s, primarily for peat extraction to supply CHPP-1. In addition to peat mining, drained lands were used for arable farming, haymaking, and livestock grazing. Currently, most of the reclaimed area is no longer managed and is undergoing various types of succession, primarily afforestation and shrub encroachment, with localized secondary waterlogging. Considering the environmental conditions and the complex, dynamic history of industrial and agricultural land use, we propose two hypotheses regarding existing and future ecosystem development trends. First, ecosystems on former peat extraction sites will remain fragmented due to the combined effects of natural and anthropogenic factors, including hydrological changes and climate impacts. Second, the predominant successional pathway on previously drained lands will be forest encroachment, with the possibility of shifting toward wetland formation depending on hydrological conditions and human intervention. Thus, this study aims to identify the main trends and patterns of ecosystem dynamics in the Tarmanskoe peatland, shaped by anthropogenic transformation and natural succession, and to model potential landscape changes over the next 30 years using the hybrid CA-Markov model. 2. Materials and Methods 2.1. Study Area The Tarmanskoe peatland is located within the West Siberian Plain, on the southern periphery of the subtaiga zone, on the second abovefloodplain terrace of the left bank of the Tura River (Khozyainova et al., 1999 ). The terrain in this area is predominantly flat, with weakly expressed features of fluvial erosion and accumulation, which contributed to the development of wetland conditions. The Tarmanskoe peatland extends from west to east for 130 km, with a width varying between 10 and 50 km. The region is characterized by an extensive distribution of lakes across its territory, with the highest concentration in the central part of the wetland, which has influenced peatland moisture levels and facilitated peat accumulation. The lakes within the Tarmanskoe complex are predominantly shallow and overgrown, with an average depth of approximately 1.5 meters. Their basins often have indistinct and irregularly shaped shorelines, which are frequently swampy. In addition to lakes, the landscape features numerous drylands distributed throughout the peatland (Giprotorfrazvedka, 1955 ). These drylands are covered by mixed coniferous and small-leaved forests, primarily composed of birch and pine, or used for agricultural purposes, mainly hayfields and pastures, with arable land being less common. The Tarmanskoe peatland is underlain by heterogeneous, predominantly loamy alluvial deposits of the Upper Pleistocene Sartansk horizon (Ogorodnov, 1971 ). The clayey composition of the underlying surface is considered one of the most significant factors contributing to the formation of the Tarmanskoe peatland (Giprotorfrazvedka, 1955 ). Water supply is primarily sourced from atmospheric precipitation and runoff from the catchment area, as well as unconfined groundwater. From a landscape perspective, the undisturbed portion of the peatland is predominantly occupied by flat, level, and slightly hummocky sedge-hypnum fens with floating mats, as well as sedge-buckbean fens, with an average peat deposit thickness of approximately 2 meters, classifying them as shallow-peat fens. The climate of the region is continental, with an average annual temperature of + 1.5 ͦ C. The warmest month, July, has a mean temperature of + 18.5 ͦ C, with an annual maximum reaching + 37 ͦ C. The coldest month, January, has a mean temperature of -16.4 ͦ C, with an annual minimum of -40 ͦ C and an absolute minimum of -50 ͦ C. The total annual precipitation is approximately 460 mm, with the majority occurring during the warm season (Kazakov, 2024). Since the collapse of the Soviet Union, the drainage network has not been maintained, leading to the partial failure of some channels, frequent secondary waterlogging, and the flooding of residential areas and agricultural lands. Peat extraction is currently conducted on a limited scale, primarily for agricultural purposes. 2.2. Methods 2.2.1. Data Sources, Processing, and Accuracy Assessment For remote sensing data analysis, we utilized multispectral satellite imagery from the Landsat TM, ETM+, and OLI missions, provided by the U.S. Geological Survey (USGS). These sensors measure Earth’s reflected radiation in the visible and infrared spectral ranges with a spatial resolution of 30 meters (Ihlen and Zanter, 2019 ). Landsat mission data have been widely used in numerous studies on land use and land cover change (LULCC) (de Waard et al., 2024 ; Kurbatova et al., 2021 ; Sirin et al., 2020 ). For time series analysis, we selected atmospherically corrected, low-cloud Landsat Collection 2 Level 2 Surface Reflectance (SR) images for the longest possible observation period. Image processing was conducted using Google Earth Engine (GEE) (Gorelick et al., 2017 ). Google Earth Engine is a cloud-based geospatial analysis platform that leverages Google’s computational power to address critical global issues, including deforestation, droughts, natural disasters, epidemics, food security, water resource management, climate monitoring, and environmental protection (Gorelick et al., 2017 ). 2.2.2. Data Selection and Classification Methodology Despite the overall availability of cloud-free and low-cloud satellite images, for certain years, high-quality images were unavailable, preventing a uniform temporal resolution between observation periods. To address this, image selection involved spatial filtering, cloud masking, and, when multiple images were available, their merging into unified median mosaics (Habib and Connolly, 2023 ; Phan et al., 2020 ). A total of 67 images from the period 1984–2024 (captured between May and September) were obtained. Given the absence of satellite imagery for certain years and variations in the quality of available cloud-free images, we selected images with a predominantly threeyear interval, which is considered optimal for time-series analysis (Habib and Connolly, 2023 ). We then conducted object-based classification for two selected images, corresponding to 1984 and 2024, as the initial and final observation points. The first stage involved generating a set of training points for supervised classification through expert assessment and the interpretation of false-color composites (SWIR-NIR-Red), which provide detailed visual information on land cover (Asare et al., 2021 ; Sali et al., 2021 ). High-resolution Google Earth satellite imagery and Soviet-era topographic maps were used as auxiliary materials for compiling the training and validation datasets. To refine classification accuracy and landscape characterization, we incorporated detailed multispectral and LiDAR data obtained using Unmanned Aerial Vehicle (UAV) surveys conducted in summer 2024. UAV surveys were conducted for five representative landscapes, which include all the land cover classes referenced in this study. The models of the drone and cameras used in this study are described in Table 1 . Table 1 Specifications of Technical Equipment Used in the UAV Survey Model Name Equipment Type Purpose of Use DJI Matrice 300 RTK UAV Platform for transporting and stabilizing remote sensing payloads DJI Zenmuse L1 LiDAR sensor Acquisition of elevation data and generation of a Digital Elevation Model (DEM) MicaSense RedEdge MX Multispectral camera Acquisition of multispectral imagery and generation of orthomosaics and vegetation indices (e.g., NDVI, NDMI) A total of eight land cover classes were identified, with three additional classes (settlements, mixed forests on former peat extraction sites, and driedup lake areas) manually delineated at later stages. The dataset was randomly split into training (70%) and validation (30%) subsets. For model training, we used the spectral bands blue, green, red, near-infrared (NIR), and both short-wave infrared (SWIR) bands, following previous methodologies (Habib and Connolly, 2023 ). The second stage involved applying image segmentation using the Simple Non-Iterative Clustering (SNIC) algorithm, integrated into Google Earth Engine. This technique partitions the image into superpixels-groups of pixels with similar characteristics-enhancing landscape contour detection (Gxokwe et al., 2022 ; Shafizadeh-Moghadam et al., 2021 ; Tassi et al., 2021 ). For land cover classification, we employed the widely used random forest (RF) algorithm-a robust ensemble learning method known for its resistance to overfitting and high classification accuracy (Jin et al., 2018 ; Millard and Richardson, 2015 ; Phan et al., 2020 ; Teluguntla et al., 2018 ). The model utilized 100 decision trees, a number determined based on prior research recommendations (Ghimire et al., 2012 ; Phan et al., 2020 ). The classification results underwent accuracy assessment using confusion matrix calculations and Kappa coefficient analysis (Congalton and Green, 2019 ). 2.2.3. Vector Data Processing and Mapping The classified segments were merged into a single layer using the intersect tool in ArcGIS Pro 3.0 (Esri), followed by visual inspection and manual correction of boundaries and land cover classifications assigned by the algorithm. Based on these layers, additional vector layers were manually created to represent the study area’s condition at different time intervals (2–3 years apart) by visually interpreting satellite imagery. A total of 15 vector layers were produced, each containing 10–11 land cover classes, as the exposed peat class was present only in the first half of the study period. A list of land cover classes and their corresponding descriptions is provided in Table 2 . Table 2 Description of LULC Classes Land use and land cover classes Description Water bodies Water surface of lakes, ponds and reservoirs Shallow areas of lakes Periodically flooded areas of lake basins Fen Waterlogged areas accumulating peat, primarily occupied by herbaceous vegetation Meadow Drained areas with diverse meadow vegetation, sometimes including shrubs Arable land Areas used for the cultivation of agricultural crops Active peat extraction sites Peat extraction sites that are operational as of the date of the imagery Waterlogged meadows and hayfields on peat extraction sites Former extraction sites covered with meadow vegetation with varying drainage Mixed forest on peat extraction sites Former extraction sites primarily covered with broadleaf and mixed forest vegetation Small-leavedandmixed forests Areas with broadleaf and mixed (subtaiga) vegetation Pine forests Areas with predominantly coniferous (pine) forest vegetation Residential areas Suburban and cottage development The integration of topographic maps from 1976–1981 allowed for more precise delineation of the drained areas. By analyzing multi-temporal land cover data, a landscape dynamics map of the Tarmanskoe peatland was created, providing a detailed representation of the changes that occurred, including their nature and spatial distribution from 1984 to 2024. Similar approaches have been employed in previous studies (Ludwig et al., 2019 ). The accuracy of the classification is presented in Table 3 . Table 3 Classification accuracy assessment for land cover maps. Year Training Overall Accuracy Test Accuracy Kappa 1984 0.99 0.84 0.82 2024 0.99 0.77 0.70 2.2.4. Spectral Data Analysis Spectral channel dynamics of Landsat satellite imagery were analyzed over the period (June-August) for four land cover classes: Mixed forests on former peat extraction sites Waterlogged and meadow grasslands on former peat extraction sites Small-leaved and mixed forests Wetlands These classes were chosen because the first two represent anthropogenically modified peatlands, while the last two cover areas minimally affected or unaffected by human activity. For mixed forests and waterlogged meadows on former peat extraction sites, areas with observed peat extraction in 1984 were selected. Spectral values from six bands (Blue, Green, Red, NIR, SWIR1, SWIR2) were extracted and three spectral indices computed (computational formulas are provided in Table 4 ): Normalized Difference Vegetation Index (NDVI); Normalized Difference Water Index (NDWI); Normalized Difference Moisture Index (NDMI). Table 4 Spectral indices calculating formulas Spectral Index Formula Reference NDVI NIR − Red NIR + Red (Rouse et al., 1974 ) NDWI Green − NIR Green + NIR (Gao, 1996 ) NDMI NIR − SWIR1 NIR + SWIR1 (Jin and Sader, 2005 ) 2.2.5. Time Series Analysis and Statistical Validation The extracted time series data for spectral bands and vegetation indices were analyzed and compared. To assess the statistical significance of trends over specific time periods, we applied the non-parametric Mann-Kendall test (Kendall, 1948 ; Mann, 1945 ). 2.2.6. CA-Markov Analysis for Land Use Change Modelling To predict future land use changes for 2034, 2044, and 2054, we employed a hybrid CA-Markov model. This method calculates a transition probability matrix using a spatial neighborhood influence algorithm and is widely used for land use and land cover (LULC) forecasting based on past trends and inherent spatial patterns (Guan et al., 2008 ; Selmy et al., 2023 ; Yang et al., 2008 ). In recent years, the application of the hybrid CA-Markov model for landscape change prediction has significantly increased (Eva et al., 2024 ; Kumar et al., 2024 ; Selmy et al., 2023 ). The model was trained using 15 land cover maps (produced at 2–3 year intervals). The simulation was conducted in Python, utilizing the GeoPandas, Matplotlib, and NumPy libraries (Harris et al., 2020 ; Van den Bossche et al., 2024 ). 2.2.7. Model Validation Model accuracy was evaluated by comparing predicted land cover changes with observed changes during the validation period. The results of accuracy assessment are presented in Table 5 . We used standard Kappa statistics and overall classification accuracy to assess model performance. The Kappa coefficient helps distinguish quantitative errors and spatial misclassification errors between two categorical maps (Selmy et al., 2023 ; Singh et al., 2015 ). Table 5 Kappa coefficients for land cover maps at 10-year intervals. Year Kappa coefficient 1993 0.915 2003 0.711 2013 0.768 2024 0.877 2.3 Limitations and Uncertainties of the Methodology Despite the methodology employed in this study being widespread and well-established, several limitations should be acknowledged. First, the classification accuracy is inherently influenced by the quality and temporal consistency of the remote sensing datasets. Seasonal differences in vegetation phenology and water levels may affect spectral signatures, especially in dynamic wetland environments, potentially leading to misclassifications. Although field validation was conducted during multiple seasons, some areas remained inaccessible due to waterlogging, limiting ground-truth observations. The absence of historical ground-truth data presents a challenge for validating past imagery, thereby introducing uncertainty in the assessment of satellite-detected changes. The algorithms used for classification are sensitive to the quality and representativeness of the training data. Given the heterogeneity of post-extraction peatland surfaces and the ongoing secondary succession processes, the training samples may not fully capture all transitional states, which can reduce classification reliability in ecotonal zones. Lastly, the long-term analysis spanning four decades’ years relies in part on archival satellite imagery, which varies in spatial, spectral, and radiometric quality. These discrepancies can introduce temporal inconsistencies, despite preprocessing efforts such as atmospheric correction and co-registration. 3. Results 3.1. General Dynamics of Natural Complexes Figure 3 illustrates the overall dynamics of land cover and land use over the study period. Active peat extraction areas have shown a steady decline in extent since the earliest time interval. Drained lands without evidence of active peat extraction have been observed throughout the entire study period. The area of croplands has gradually decreased, with some fluctuations, although at no point they were entirely absent. Residential zones have exhibited continuous expansion, partially replacing drained and natural grassland ecosystems and, less frequently, forests. Water bodies, primarily lakes, experienced minor fluctuations in total area between 1984 and 2013. However, from that point onward, a declining trend emerged, reaching its lowest recorded extent in 2024. Natural wetland areas and pine forests have remained largely unchanged throughout the observation period. Overall, the most substantial land cover changes were recorded between 1984 and 2003. Among the general patterns of land cover dynamics in the Tarmanskoe peatland, as illustrated in Fig. 3 , the expansion of mixed-species forest cover on former peat extraction sites is particularly pronounced. This trend is especially noticeable south of the Bolshoye and Sredneye Tarmanskoe lakes in the central part of the study area, as well as in the northeastern and eastern regions, where it occurs in the vicinity of undisturbed forests and wetlands. However, the majority of former peat extraction sites are now occupied by waterlogged and meadow grasslands, covering an area of 126 km 2 (or 59.2% of the total peat extraction area), compared to 73 km 2 of forest (34.5%). Minor areas of flooding, amounting to 0.17%, are observed in the eastern part of the study area. As previously noted, the reduction in surface water area has become a significant phenomenon due to lake shallowing and wetland expansion. This process is most evident in the lakes located in the eastern part of the study area, as well as in Sredneye Tarmanskoe Lake, where a substantial portion has dried up. According to calculations, nearly 49% of the total lake area within the study region has undergone shallowing. Based on the data presented in Fig. 3 , most of the pasture, hayfields, and croplands (11.2% of the study area or 92 km 2 ) are concentrated in the southwestern part of the region, where numerous fragmented patches of secondary small-leaved and mixed forests have developed on former agricultural lands. The northwestern territories are dominated by stable, undisturbed landscapes consisting of wetlands, coniferous, mixed, and small-leaved forests. Settlements are concentrated in the southern part of the study area, closer to the city of Tyumen and along the highway running parallel to the Tura River floodplain. The majority of these settlements have been established on land previously used for agriculture (49% of total settlement area), peat extraction (18.2%), and forests (11.9%). The highest frequency of land cover changes occurs in areas subject to economic activity, particularly peat extraction and agriculture (Fig. 4 ). Pasture, hayfields, and croplands exhibit the greatest variability, as periodic shifts in land use practices are observed. Figure 4 serves as an indicator of the stability of landscape conditions within the study area. The map of polygon distribution within the study area highlights a pronounced landscape mosaic, with a substantial proportion of polygons belonging to classes with minimal areas (up to 0.41% of the total area). The largest polygons correspond to natural wetland, forest, and lake areas, primarily concentrated in the northern part of the study site, where no drainage or peat extraction has taken place. These polygons cover between 4% and 18% of the total area. An exception is found in the eastern part of the site, where abandoned, unforested peat extraction sites account for 12% of the total area. It is important to note that this area is conditionally homogeneous, as drainage channels were not classified separately. As a result, the map does not reflect the subdivision into checks, which, in reality, segment the area into smaller components. The highest degree of land fragmentation is observed in the western and central parts of the study area, where peat extraction was most intensive, and multiple attempts at agriculture were made. Some of these agricultural practices were partially successful and have persisted to the present day. In these areas, polygons of the smallest size classes account for 65.72% and 68.95% of the total area, respectively. Figure 7 presents the dynamics of areas subjected to drainage. The largest extent of active peat extraction was recorded in 1984, followed by a steady decline, ultimately leading to the complete cessation of peat extraction by 1997. Over time, mixed forests progressively replaced meadow vegetation. Until 2011, this transition occurred at a rapid pace, as evident from the data shown in Fig. 7 , but thereafter, the trend persisted while slowing considerably. By 2024, the area of meadows on former peat extraction sites reached 15.2%, which is nearly the same as in 1984 when they covered 15.9% of the study area. In contrast, mixed forests expanded significantly, now occupying 9.1% of the total area in 2024, whereas at the beginning of the study period, they were just emerging and covered less than 0.5% of the territory. Figure 8 illustrates the frequency and total area of successional changes that occurred over the study period. The highest frequency and extent of changes were recorded between 1987 and 2006, followed by a period of relative stabilization. 3.2. Variability of Spectral Characteristics As expected, land cover classes unaffected by human activity during the study period exhibit relative stability in their spectral channel values (Fig. 9 ). Only wetlands show a slight downward trend in near-infrared reflectance (NIR) and an even weaker increase in shortwave infrared reflectance (SWIR1, SWIR2). For classes resulting from secondary succession on former peat extraction sites, an increase in NIR reflectance is observed until 1998–2000, after which values stabilize. Waterlogged and meadow grasslands exhibit greater fluctuations in this spectral range compared to mixed forests. Additionally, after 2015, there is a noticeable decline in NIR reflectance and an increase in SWIR reflectance, influencing trends in NDVI and NDMI. In mixed forests, all spectral channel values have remained stable since the early 21st century, a trend also reflected across all spectral indices. Between 1984 and 2000, NDVI values in secondary mixed forests increased from 0.15 to 0.40, NDMI rose from − 0.10 to 0.24, while NDWI declined from − 0.20 to -0.38. In waterlogged and meadow grasslands, NDVI increased from 0.15 to 0.30, NDMI from − 0.10 to 0.15, and NDWI decreased from − 0.15 to -0.30. The Mann-Kendall statistical test for trend significance in NDVI and NDMI from 2016 to 2024 for the class of waterlogged and meadow grasslands on former peat extraction sites yields a p-value near the significance threshold (p = 0.04) for both indices-NDVI (declining from 0.31 to 0.21) and NDMI (declining from 0.13 to -0.01). Given this result and the short observation period, the downward trend cannot yet be confirmed with high confidence. 3.3. Markov Chain Model Analysis The CA-Markov model projection reflects expected land cover changes at three future time points: 2034, 2044, and 2054. The simulation predicts the continued expansion of forest ecosystems, particularly mixed forests on former peat extraction sites, following the trend identified in the LULC change analysis based on Landsat imagery from 1984 to 2024. Since 1990, forests have increasingly occupied drained wetland areas, and according to the model, this process will continue. By 2054, forests are projected to cover 13.7% of the total study area. However, the model also forecasts increased stability of meadow ecosystems on former peat extraction sites. Over the 27-year period from 1997, when meadow ecosystems on former peat extraction sides reached its maximum area, to 2024, these meadows have significantly declined, from 23–15.2%, primarily due to forest encroachment. However, the CA-Markov model predicts that further replacement of meadows by forests and other land cover types will be limited to an additional 4.6%, indicating a remarkable slowing trend in meadow loss. The model also projects further expansion of residential areas. By 2034, built-up areas are expected to increase by 2.2% of the total study area compared to 2024. This trend is expected to continue in later time periods, reaching a maximum of 14.3% of the total area by 2054, representing a 6,3% increase from 2024. The decline in water bodies is forecasted to persist, following the previously observed trend. Starting at 3.5% of the total area in 2024, their extent is projected to shrink continuously, and by 2054, only 31.9% of their 2024 extent will remain-equivalent to 17.5% of their original area in 1984. The most stable land cover types, according to the model, are natural wetlands, coniferous forests, and croplands. Their areas are expected to experience only minor fluctuations within a ± 0.5% range over time. 4. Discussion 4.1. Main Successional Pathways The presence of a significant proportion of waterlogged and meadow grasslands on former peat extraction sites between 1984 and 1990 suggests a prolonged period of site abandonment or difficulties in organizing peat extraction on some drained lands. The first occurrence of forest stands on these sites only in 1993 further indicates the absence or extremely low intensity of afforestation efforts. Artificial reforestation is a common method for rehabilitating disturbed peatlands (Laudon and Maher Hasselquist, 2023 ). However, the long-term effectiveness of large-scale afforestation remains debatable, as without proper maintenance of the drainage network, the gas exchange and hydrological properties of afforested areas tend to revert to their natural state, leading to waterlogging and methane emissions (Inisheva et al., 2021 ). In contrast, localized afforestation measures, such as tree belts along drainage channels (Inisheva et al., 2023 ) and the creation of forest-meadow-wetland agro-landscapes, are considered more sustainable and ecologically safe approaches (Ulanov et al., 2023 ). The sporadic and unsystematic appearance of croplands on former peat extraction sites reflects two key factors: the degradation of the drainage network, which led to waterlogging and secondary wetland formation, and the overall decline in agriculture and rural depopulation in Russia at the turn of the 20th-21st centuries, driven by socio-political changes (Sheludkov et al., 2020 ). Over the past 40 years, the dominant successional pathways in the study area have been ’peat extraction → waterlogged meadows and hayfields’ and ’peat extraction → waterlogged meadows and hayfields → mixed forests’. The first pathway was predominant for most of the study period, likely due to poor drainage conditions and soil properties. However, from 1997 to 2007, the transition from meadows to forests on former peat extraction sites accelerated significantly. After this period, forest expansion continued until 2024, but at a noticeably slower rate. The onset of rapid mixed forest growth on these sites coincided with the cessation of large-scale industrial peat extraction. This correlation is likely not coincidental, as the discontinuation of direct anthropogenic influences, such as heavy machinery operations, created favourable conditions for rapid forest encroachment on well-drained areas with suitable soil properties. This process aligns with natural succession patterns characteristic of the regional climatic conditions. The expansion of forests is a typical process in recovering peatlands (Habib and Connolly, 2023 ; Sirin et al., 2020 ), and the spread of woody and shrub vegetation in the study area has been previously noted (Pashnina and Simakova, 2017 ; Simakova et al., 2023 ). 4.2. Dynamics of Water Surface Area A particularly interesting aspect is the dynamics of lakes within the Tarmanskoe peatland. An analysis of water bodies area revealed a significant reduction since 2013, with a more than 40% decrease over the following decade due to progressive shallowing and overgrowth. All lakes in the study area exhibit floating-mat shoreline vegetation expansion (Solodovnikov, 2021 ). In the Tarmanskoe peatland, this process was already observed during peat deposit surveys in the 1950s (Giprotorfrazvedka, 1955 ). Drainage reclamation also influenced the hydrological regime and water levels, causing some lakes-such as Lake Svetloe in the southern part of the wetland-to completely disappear, as confirmed by satellite imagery analysis. Observations at Lake Srednee Tarmanskoe, where monitoring by the Tyumen Fen Station began in 1960, indicate that active water level decline started in 1971, following the drainage of adjacent wetland areas. This trend continued for about a decade (Materials from Fen Station Observations, 1974, Fig. 11 ). However, after a few years, the lake’s water level nearly returned to its original state and remained stable until the late 1980s, when peat extraction sites were abandoned. Similar or even more pronounced trends of water surface reduction following drainage activities have been observed in other human-modified wetland complexes, though they tend to slow down over time (Kurbatova et al., 2021 ). However, in the case of the Tarmanskoe peatland, lake area reduction has intensified again in recent decades. Analysis of satellite imagery from 1984 to 2024 shows that the process of shallowing and overgrowth became particularly evident in the 2000s, more than a decade after industrial peat extraction ceased, despite the near-total degradation of the drainage system. Notably, water bodies in parts of the wetland that were never subjected to intensive human impact have also experienced overgrowth and shallowing, albeit to a lesser extent. Thus, the causes of lake area reduction appear to be complex, involving both natural and anthropogenic factors. Most likely, the natural aging of water bodies was compounded by drainage reclamation, which led to lower groundwater and wetland water levels, as well as climate change, manifested in rising mean annual temperatures (Fig. 12 ). The stabilization of the hydrological regime a few years after drainage began, along with a partial rebound in groundwater and wetland water levels following the abandonment of peat extraction sites and drainage system degradation, likely slowed the shallowing process. However, the continued rise in air temperature has reaccelerated this trend (Figs. 13 , 14 ). 4.3. Landscape Mosaic and Fragmentation Small-scale, mosaic-like landscapes are primarily concentrated in areas previously subjected to peat extraction, indicating a lasting impact of human activity on natural ecosystems, even after anthropogenic pressure has ceased. As these areas undergo recovery, they exhibit a diversity of ecosystems, with some sections becoming forested, while others remain open landscapes with varying degrees of water saturation. Forested areas that have developed on former peat extraction sites display different stages of successional processes. Analysis of high-resolution Esri Imagery and UAV surveys revealed a heterogeneous structure of forest communities formed on drained lands. These forests often consist of a mosaic of tree stands with varying canopy density, age, and species composition, interspersed with shrub-dominated patches, as well as open grassland and wetland ecosystems. This structural heterogeneity affects their ecological state, biodiversity, and spatial organization, creating a range of microhabitats that, in turn, support overall biological diversity (Erds et al., 2018 ; Mobaied et al., 2016 ; Redon et al., 2014 ). However, such mosaic landscapes have both positive and negative consequences. On the one hand, ecosystem diversity can enhance species richness and resilience to external disturbances, providing varied ecological niches for flora and fauna (Hitchman et al., 2018 ). On the other hand, high landscape fragmentation can disrupt migration routes, alter ecosystem processes, and create challenges for land management (Haddad et al., 2015 ; Hanski, 2015 ; Liu et al., 2018 ). Some former peat extraction sites have been converted for residential, industrial, and agricultural use. However, most drainage systems are no longer maintained, leading to frequent flooding. Drainage reduction has also contributed to rising groundwater levels, which, combined with new construction and road development, has further complicated land use. In addition to the direct risks of property damage, such flooding introduces various contaminants that primarily affect soil and water quality. 4.4. Spectral Information Natural ecosystems unaffected by anthropogenic disturbance exhibit stable spectral channel and index values. In contrast, ecosystems that have emerged through secondary succession on former peat extraction sites show a gradual decline in reflectance in SWIR channels and a corresponding increase in NIR reflectance, influencing the dynamics of spectral indices. It is well established that SWIR bands correlate with water reflectance and moisture content (Burdun et al., 2020 ; Koley and Jeganathan, 2020 ), while NIR reflectance is associated with chlorophyll content and vegetation biomass (Holben, 1986 ). NDVI and NDMI are linked to chlorophyll and water content in plants, with NDMI particularly sensitive to water stress, while NDWI reflects the amount of liquid water in plant canopies (Jin and Sader, 2005 ). Thus, increasing NDVI and NDMI values during secondary successionas former peat extraction sites become overgrown with forest and meadow vegetation-indicates rising vegetation biomass, while a declining NDWI reflects reduced water saturation and increased drainage. However, the recent trend of rising NDWI in meadow ecosystems suggests increasing water saturation, likely due to rising groundwater levels, which may negatively impact vegetation. This stress effect is reflected in declining NDVI and NDMI values. Peatland recovery may be gradually shifting plant communities toward helophytic peatland vegetation (Haapalehto et al., 2011 ). The observed spectral index dynamics, combined with previous studies on the Tarmanskoe peatland (Pashnina and Simakova, 2017 ), may serve as indirect evidence of rising groundwater levels, resulting in secondary wetland formation and the development of meadow-wetland ecosystems. This raises the question of whether the decline in lake area since 2013 is linked to the increasing water saturation of abandoned peat extraction sites, a trend that has been detectable in spectral data since 2015–2016. 4.5. CA-Markov Modelling Results The results of the CA-Markov model predicting future changes in the land cover structure of the Tarmanskoe peatland provide important insights into potential landscape dynamics over the next 30 years. The data confirm existing trends and reveal new possible successional pathways, which is particularly relevant in the context of global challenges related to managing disturbed peatlands and mitigating the impacts of climate change. The forecast for the period up to 2054 indicates a continued expansion of forest ecosystems, particularly mixed and small-leaved forests on former peat extraction sites. This process aligns with observations from the past 40 years and confirms the hypothesis that forest succession dominates on drained peatlands. Forest ecosystems play a critical role in carbon sequestration and the stabilization of hydrological conditions, which is vital for reducing greenhouse gas emissions. However, the model also projects a slowing of forest expansion by 2044 and 2054, suggesting that the landscape may be reaching a certain level of stabilization. Waterlogged meadow and hayfields on former peat extraction sites will continue to cover significant areas despite their gradual replacement by forest vegetation. This reflects the second key successional pathway, which involves the preservation of resilient meadow communities. Over the 30-year forecast period (2024–2054), the model predicts a 3.7% reduction in meadow area, compared to a 7.8% decline during the 40-year observation period from 1984 to 2024, largely due to forest encroachment. This indicates increasing ecosystem stability. Meadows on former peat extraction sites remain ecologically significant, contributing to moisture retention, soil erosion prevention, and biodiversity preservation. However, without proper management, these areas could be at risk of secondary waterlogging or desiccation, depending on changes in drainage or climatic conditions. This would lead to the loss of ecosystem functions or increased fire hazards. The forecast highlights the need for a comprehensive management strategy for these territories. Of particular concern is the projected increase in urbanized areas. The growth of settlements on former agricultural lands and peat extraction sites suggests intensifying anthropogenic pressure on ecosystems. This process could contribute to further landscape fragmentation and changes in the hydrological regime. Landscape fragmentation negatively affects wildlife migration routes, reduces ecosystem resilience, and complicates natural restoration processes. Furthermore, the expansion of urban areas is often accompanied by road and infrastructure development, which can alter drainage patterns, leading to increased local groundwater and wetland water levels and causing flooding. The projected reduction of water bodies to 17.5% of their 1984 extent by 2054 is an alarming indicator. Lakes and other water bodies provide numerous essential ecosystem services, including maintaining hydrological balance, serving as habitats for diverse flora and fauna, regulating the local climate, and supporting recreation and fisheries. Lake shallowing may result in the loss of these ecosystem functions. Moreover, reduced lake depth and encroachment by aquatic vegetation can cause unpredictable shifts in the carbon balance (Aben et al., 2022 ; Bodmer et al., 2024 ; Oliveira-Junior et al., 2018 ). Many studies have highlighted increased methane emissions from shallow and overgrown aquatic ecosystems (Deemer and Holgerson, 2021 ; Struik et al., 2022 ). Overall, the reduction in moisture levels across the study area increases the risk of forest and peatland fires, which can cause severe damage to biodiversity, soil resources, and carbon storage, leading to significant carbon emissions into the atmosphere (Jacobson et al., 2024 ; Prat-Guitart et al., 2017 ). The projected reduction in water bodies underscores the urgent need for additional conservation and restoration measures to prevent negative ecological consequences. 5. Conclusions The analysis of land cover dynamics revealed a steady decline in the area of peat extraction sites and meadows, accompanied by the expansion of forest ecosystems, indicating the dominance of forest succession on drained lands. Forest ecosystems play a crucial role in carbon sequestration and maintaining hydrological balance, thereby reducing greenhouse gas emissions. However, their recovery may have contrasting effects on secondary waterloggingtemporarily lowering groundwater levels, while in the long term, promoting wetland formation in poorly drained areas. The reduction in lake area, identified through remote sensing data, requires particular attention. Over the past decade (2013–2024), the water surface area of the Tarmanskoe peatland has decreased by more than 40%, indicating active shallowing and overgrowth. This decline has a complex negative impact on the hydrological balance and ecosystem functions of the region. Spectral analysis revealed trends of increasing biomass accumulation and decreasing water saturation in meadows and forests that have developed on former peat extraction sites. This confirms the gradual recovery of vegetation communities and their transition to a more stable state. The CA-Markov model predicts the continued expansion of forest ecosystems, but at a slower pace, suggesting relative stabilization of meadow ecosystems. The further reduction of water bodies highlights the urgent need to preserve hydrological functions and prevent soil degradation. According to the model, by 2054, lake area may shrink to 17.5% of its 1984 extent, posing a range of long-term ecological risks, including an increased risk of wildfires. The observed landscape mosaic and high ecosystem fragmentation result from anthropogenic impact and uneven recovery of natural complexes. While this fosters habitat diversity and supports biodiversity, it also complicates hydrological processes, hinders the sustainable development of individual ecosystems, and heightens the area’s vulnerability to external stressors, including climate change and human activities. These findings underscore the need for active management of disturbed landscapes to prevent soil degradation and maintain ecosystem functions and biodiversity. The restoration of degraded peatlands requires a comprehensive approach, considering hydrological conditions, succession patterns, anthropogenic pressure, and fire risks. Such an approach will help minimize ecological risks, reduce wildfire hazards, prevent soil degradation and flooding risks, and preserve the ecosystem functions of the landscape. Declarations Acknowledgments This work was supported by the Russian Science Foundation under Grant [number 23-77-10012]; the analysis of changes in the area of water bodies within the wetland ecosystem was carried out within the thematic framework of the State Assignment of the Institute of Forest Science of the Russian Academy of Sciences [number 123033000042-6]. Author contributions Vladimir Ivanov was responsible for data curation, preparation of the original manuscript draft, and contributed significantly to the visualization of the results. Ivan Milyaev carried out the formal analysis and contributed to the development of the methodology, preparation of visualizations, and writing of the Methods section. Evgeniya Soldatova contributed to the conceptualization of the study and provided overall supervision throughout the research and manuscript preparation. All authors participated in conducting the field investigation. Data Availability The data are available directly from the cited data sources and, upon request, from the authors. Declarations Competing Interests The Authors declare no conflicts of interest. References Aben RCH, Velthuis M, Kazanjian G, et al (2022) Temperature response of aquatic greenhouse gas emissions differs between dominant plant types. Water Research 226:119251. https://doi.org/10.1016/j.watres.2022.119251 Akhmeteva N, Mikhailova A, Krichevets G, et al (2019) Transformatsiya Antropoghenno Narushennykh Torfianyh Bolot V Novyy Tip Landshafta V Tsentralnykh Rayonakh Evropeyskoy Chasti Rossii [Transformation of Anthropogenically Disturbed Peat Bogs into a New Type of Landscape in the Central Areas of the European Part of Russia]. Trudy Instorfa [Proceedings of Instorf] (20(73)):3–10 (in Russian) Asare A, Thodsen H, Antwi M, et al (2021) Land use and land cover changes in lake Bosumtwi Watershed, Ghana (West Africa). Remote Sensing Applications: Society and Environment 23:100536. https://doi.org/10.1016/j.rsase.2021.100536 Baisheva E, Martynenko V, Shirokikh P, et al (2022) About distribution of drained peatlands in bashkir CIS-urals. Èkobiteh https://doi.org/10.31163/2618-964x-2021-5-1-10-19 Bodmer P, Vroom RJE, Stepina T, et al (2024) Methane dynamics in vegetated habitats in inland waters: quantification, regulation, and global significance. Frontiers in Water. https://doi.org/10.3389/frwa.2023.1332968 Bondar EG (2022) The current state and prospects of the fuel and energy complex of St. Petersburg and the Leningrad region. The economy of the North-West: problems and prospects of development https://doi.org/10.52897/2411-4588-2022-2-71-77 Van den Bossche J, Fleischmann M, McBride J, et al (2024) GeoPandas. https://geopandas.org/en/stable/ Accessed 15 Oct 2024 Bruisch K (2020) Nature Mistaken: Resource-Making, Emotions and the Transformation of Peatlands in the Russian Empire and the Soviet Union. Environment and History 26:359–382. https://doi.org/10.3197/096734018X15254461646567 Burdun I, Bechtold M, Sagris V, et al (2020) Satellite determination of peatland water Table temporal dynamics by localizing representative pixels of a SWIR-based moisture index. Remote Sensing 12(18):2936. https://doi.org/10.3390/rs12182936 Chen Y, Hu X, Zhang Y, et al (2022) Characterizing the Long-Term Landscape Dynamics of a Typical Cloudy Mountainous Area in Northwest Yunnan, China. Sustainability https://doi.org/10.3390/su142013488 Congalton RG, Green K (2019) Assessing the accuracy of remotely sensed data: principles and practices, third edition, 3rd edn. CRC Press, Boca Raton. https://doi.org/10.1201/9780429052729 Deemer BR, Holgerson MA (2021) Drivers of Methane Flux Differ Between Lakes and Reservoirs, Complicating Global Upscaling Efforts. Journal of Geophysical Research: Biogeosciences 126(4):e2019JG005600. https://doi.org/10.1029/2019JG005600 Edvardsson J, Helama S, Rundgren M, et al (2022) The Integrated Use of Dendrochronological Data and Paleoecological Records From Northwest European Peatlands and Lakes for Understanding Long-Term Ecological and Climatic ChangesA Review. Frontiers in Ecology and Evolution. https://doi.org/10.3389/fevo.2022.781882 Erds L, Kröel-Dulay G, Bátori Z, et al (2018) Habitat heterogeneity as a key to high conservation value in forest-grassland mosaics. Biological Conservation. https://doi.org/10.1016/J.BIOCON.2018.07.029 Eva EA, Marzen LJ, Lamba J, et al (2024) Projection of land use and land cover changes based on land change modeler and integrating both land use land cover and climate change on the hydrological response of Big Creek Lake Watershed, South Alabama. Journal of Environmental Management 370:122923. https://doi.org/10.1016/j.jenvman.2024.122923 Filippova N, Zvyagina E, Rudykina E, et al (2023) The diversity of macromycetes in peatlands: nine years of plot-based monitoring and barcoding in the raised bog "Mukhrino", West Siberia. Biodiversity Data Journal. https://doi.org/10.3897/BDJ.11.e105111 Fu B, Zhang L, Xu Z, et al (2015) Ecosystem services in changing land use. Journal of Soils and Sediments 15:833–843.https://doi.org/10.1007/s11368-015-1082-x Gao Bc (1996) NDWIA normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment 58(3):257–266. https://doi.org/10.1016/S0034-4257(96)00067-3 Ghimire B, Rogan J, Galiano VR, et al (2012) An evaluation of bagging, boosting, and random forests for land-cover classification in Cape Cod, Massachusetts, USA. GIScience & Remote Sensing 49(5):623–643. https://doi.org/10.2747/1548-1603.49.5.623 Giprotorfrazvedka (1955) Detailed Reconnaissance Materials of the Tarmaanskoye Peat Deposit. Tech. Rep. 4, Gorkiy Giprotorfrazvedka (1955) Detailed Reconnaissance Materials of the Tarmaanskoye Peat Deposit. Tech. Rep. 15, Gorkiy Gomes E, Inácio M, Bogdzevi K, et al (2021) Future land-use changes and its impacts on terrestrial ecosystem services: A review. The Science of the total environment 781. https://doi.org/10.1016/j.scitotenv.2021.146716 Gomes L, Bianchi F, Cardoso IM, et al (2020) Land use change drives the spatio-temporal variation of ecosystem services and their interactions along an altitudinal gradient in Brazil. Landscape Ecology 35:1571–1586. https://doi.org/10.1007/s10980-020-01037-1 Gorelick N, Hancher M, Dixon M, et al (2017) Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment 202:18–27. https://doi.org/10.1016/j.rse.2017.06.031 Guan D, Gao W, Watari K, et al (2008) Land use change of Kitakyushu based on landscape ecology and Markov model. Journal of Geographical Sciences 18(4):455–468. https://doi.org/10.1007/s11442-008-0455-0 Gxokwe S, Dube T, Mazvimavi D (2022) Leveraging Google Earth Engine platform to characterize and map small seasonal wetlands in the semiarid environments of South Africa. Science of The Total Environment 803:150139. https://doi.org/10.1016/j.scitotenv.2021.150139 Haapalehto TO, Vasander H, Jauhiainen S, et al (2011) The effects of peatland restoration on watertable depth, elemental concentrations, and vegetation: 10 years of changes. Restoration Ecology 19(5):587–598. https://doi.org/https://doi.org/10.1111/j.1526-100X.2010.00704.x Habib W, Connolly J (2023) A national-scale assessment of land use change in peatlands between 1989 and 2020 using Landsat data and Google Earth Enginea case study of Ireland. Regional Environmental Change 23(4):124. https://doi.org/10.1007/s10113-023-02116-0 Haddad N, Brudvig L, Clobert J, et al (2015) Habitat fragmentation and its lasting impact on Earths ecosystems. Science Advances 1. https://doi.org/10.1126/sciadv.1500052 Hanski I (2015) Habitat fragmentation and species richness. Journal of Biogeography 42. https://doi.org/10.1111/jbi.12478 Harris CR, Millman KJ, van der Walt SJ, et al (2020) Array programming with NumPy. Nature 585(7825):357–362. https://doi.org/10.1038/s41586020-2649-2 Hitchman SM, Mather ME, Smith JM, et al (2018) Identifying keystone habitats with a mosaic approach can improve biodiversity conservation in disturbed ecosystems. Global Change Biology https://doi.org/10.1111/gcb.13846 Holben BN (1986) Characteristics of maximum-value composite images from temporal AVHRR data. International Journal of Remote Sensing (7):1417– 1434. https://doi.org/10.1080/01431168608948945 Hugelius G, Loisel J, Chadburn S, et al (2020) Large stocks of peatland carbon and nitrogen are vulnerable to permafrost thaw. Proceedings of the National Academy of Sciences of the United States of America 117:20438– 20446. https://doi.org/10.1073/pnas.1916387117 Humpenöder F, Karstens K, Lotze-Campen H, et al (2020) Peatland protection and restoration are key for climate change mitigation. Environmental Research Letters 15. https://doi.org/10.1088/1748-9326/abae2a Ihlen V, Zanter K (2019) Landsat 8 (L8) Data Users Handbook, 5th edn. U.S. Geological Survey, Sioux Falls Inisheva L, Shaydak L, Babikov B (2021) Hydrological and gas regime of swamps in the conditions of forest reclamation. Environmental Science pp 39–44. https://doi.org/10.32962/0235-2524-2020-6-39-44 Inisheva L, Sergeeva M, Golovchenko A, et al (2023) Carbon Dioxide and Methane Distribution in Peat Deposits of an Oligotrophic Forest Swamp in Western Siberia and Their Emission. Lesovedenie [Forest Science] https://doi.org/10.31857/s0024114823010060 Jacobson TWP, Seager R, Williams AP, et al (2024) An unexpected decline in spring atmospheric humidity in the interior Southwestern United States and implications for forest fires. Journal of Hydrometeorology https://doi.org/10.1175/jhm-d-23-0121.1 Jin S, Sader SA (2005) Comparison of time series tasseled cap wetness and the normalized difference moisture index in detecting forest disturbances. Remote Sensing of Environment 94(3):364–372. https://doi.org/10.1016/j.rse.2004.10.012 Jin Y, Liu X, Chen Y, et al (2018) Land-cover mapping using Random Forest classification and incorporating NDVI time-series and texture: a case study of central Shandong. International Journal of Remote Sensing 39(23):8703–8723. https://doi.org/10.1080/01431161.2018.1490976 Karpov V (2010) Neft’ i gaz v promyshlennoy politike SSSR (Rossii) [Oil and gas in the industrial policy of the USSR (Russia)]. Vestnik Nizhevartovskogo gosudarstvennogo universiteta [Bulletin of Nizhnevartovsk State University] (4):75–88. (In Russian) Kazakov K (2025) Letopis' Pogody i Klimata Tyumeni [Chronicle of Weather and Climate of Tyumen]. http://www.pogodaiklimat.ru/history/28367.html. (In Russian) Accessed 15 Oct 2024 Kendall M (1948) Rank correlation methods. Rank correlation methods, Griffin, Oxford, England Khozyainova NV, Cheshuina IA, Glazunov VA (1999) Flora Tarmanskogo leso-vodo-bolotnogo kompleksa [Flora of the Tarman forest-water-wetland complex]. Vestnik Tyumenskogo gosudarstvennogo universiteta [Bulletin of Tyumen State University] (3):92–98. (In Russian) Koley S, Jeganathan C (2020) Estimation and evaluation of high spatial resolution surface soil moisture using multi-sensor multi-resolution approach. Geoderma 378:114618. https://doi.org/10.1016/j.geoderma.2020.114618 Krasnov O, Zhabin V, Matukhina V, et al (2005) Syryevaya baza torfa Sredney Sibiri i osnovnye napravleniya ego ratsional’nogo ispol’zovaniya [Raw material base of peat in Central Siberia and the main directions of its rational use]. Interekspo Geo-Sibir’ [Interexpo Geo-Siberia] 3(1):21–26. (In Russian) Kumar M, Mahato LL, Suryavanshi S, et al (2024) Future prediction of water balance using the SWAT and CA-Markov modelăusing INMCM5 climate projections: a case study of the Silwani watershed (Jharkhand), India. Environmental Science and Pollution Research 31(41):54311–54324. https://doi.org/10.1007/s11356-023-27547-4 Kurbatova I, Vereshchaka T, Ivanova A (2021) Space monitoring bog landscape transformation under anthropogenic impact conditions. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa [Corrent problems in remote sensing of the Earth from space] 18(4):216– 227. https://doi.org/10.21046/2070-7401-2021-18-4-216-227, number: 4 Kurganova I, Gerenyu VL, Kuzyakov Y (2015) Large-scale carbon sequestration in post-agrogenic ecosystems in Russia and Kazakhstan. Catena 133:461–466. https://doi.org/10.1016/J.CATENA.2015.06.002 Lapytskaya M (2020) Transformations in the Agrarian Sector as a Method of Problem Resolution of the Peasant Land Shortages in Russia (Based on the Research on Agrarian Reforms in Russia in Second Half of XIX - Beginning of the XX Centuries) 20:601–624. https://doi.org/10.17150/23082488.2019.20(4).601-624 Laudon H, Maher Hasselquist E (2023) Applying continuous-cover forestry on drained boreal peatlands; water regulation, biodiversity, climate benefits and remaining uncertainties. Trees, Forests and People 11:100363. https://doi.org/10.1016/j.tfp.2022.100363 Leifeld J, Wüst-Galley C, Page S (2019) Intact and managed peatland soils as a source and sink of GHGs from 1850 to 2100. Nature Climate Change pp 1–3. https://doi.org/10.1038/s41558-019-0615-5 Leng LY, Ahmed O, Jalloh M (2018) Brief review on climate change and tropical peatlands. Geoscience Frontiers. https://doi.org/10.1016/J.GSF.2017.12.018 Liu J, Wilson M, Hu G, et al (2018) How does habitat fragmentation affect the biodiversity and ecosystem functioning relationship? Landscape Ecology 33:341–352. https://doi.org/10.1007/s10980-018-0620-5 Loisel J, Gallego-Sala A, Amesbury M, et al (2020) Expert assessment of future vulnerability of the global peatland carbon sink. Nature Climate Change 11:70–77. https://doi.org/10.1038/s41558-020-00944-0 Luan C, Liu R (2022) A comparative study of various land use and land cover change models to predict ecosystem service value. International Journal of Environmental Research and Public Health 19(24):16484. https://doi.org/10.3390/ijerph192416484 Ludwig C, Walli A, Schleicher C, et al (2019) A highly automated algorithm for wetland detection using multi-temporal optical satellite data. Remote Sensing of Environment https://doi.org/10.1016/J.RSE.2019.01.017 Lvov Y (1995) Torf i formy ego ispol’zovaniya v Sibiri [Peat and its forms of use in Siberia] pp 31–39. Publisher: Natsional’nyy issledovatel’skiy Tomskiy gosudarstvennyy universitet [National Research Tomsk State University] (In Russian) Lysenko Y (2021) Ethno-Economics of the Kazakhs of the Steppe Region in the Modernization Plans of the Russian Empire (second half of the XIX beginning of the XX century). Bylye Gody. https://doi.org/10.13187/bg.2021.2.840 Ma XY, Xu H, Cao ZY, et al (2022) Will climate change cause the global peatland to expand or contract? Evidence from the habitat shift pattern of Sphagnum mosses. Global Change Biology 28(21):6419–6432. https://doi.org/10.1111/gcb.16354 Malek Ž, Douw B, Vliet Jv, et al (2019) Local land-use decisionmaking in a global context. Environmental Research Letters 14. https://doi.org/10.1088/1748-9326/ab309e Mann HB (1945) Nonparametric tests against trend. Econometrica 13(3):245–259. https://doi.org/10.2307/1907187 Maslov M, Maslova O (2020) Temperate peatlands use-management effects on seasonal patterns of soil microbial activity and nitrogen availability. Catena 190. https://doi.org/10.1016/j.catena.2020.104548 McDaniel M, Saha D, Dumont M, et al (2019) The Effect of Land-Use Change on Soil CH4 and N2O Fluxes: A Global Meta-Analysis. Ecosystems 22:1424–1443. https://doi.org/10.1007/s10021-019-00347-z McLaughlin DL, Cohen MJ (2013) Realizing ecosystem services: wetland hydrologic function along a gradient of ecosystem condition. Ecological Applications 23(7):1619–1631. https://doi.org/10.1890/12-1489.1 Mickler R (2021) Carbon emissions from a temperate coastal peatland wildfire: contributions from natural plant communities and organic soils. Carbon Balance and Management 16. https://doi.org/10.1186/s13021-02100189-0 Millard K, Richardson M (2015) On the importance of training data sample selection in random forest image classification: a case study in peatland ecosystem mapping. Remote Sensing 7(7):8489–8515. https://doi.org/10.3390/rs70708489 Minaeva T, Sirin A (2009) A Quick Scan of Peatlands in Central and Eastern Europe. Wageningen, The Netherlands. Mobaied S, Geoffroy J, Machon N (2016) The Importance of Spatiotemporal Heterogeneity for Biodiversity in Forest Heathland Mosaics and Implications for Heathland Conservation. Journal of Environmental Protection 07:1317–1332. https://doi.org/10.4236/JEP.2016.710115 Monier E, Kicklighter D, Sokolov A, et al (2017) A review of and perspectives on global change modeling for Northern Eurasia. Environmental Research Letters 12. https://doi.org/10.1088/1748-9326/aa7aae Moskalenko N, Bulko N, Tolkacheva N, et al (2020) K voprosu o sostoyanii meliorirovannykh zemel’, nakhodyashchikhsya v sostave lesnogo fonda [Regarding the condition of improved lands that are part of the forest fund]. Vestnik Grodnyenskogo Gosudarstvennogo Universiteta Imeni Yanki Kupaly Seriya 5 Ekonomika Sotsiologiya Biologiya [Bulletin of the Grodno State University named after Yanka Kupala Series 5 Economics Sociology Biology] 10(1):125–132. (In Russian) Nekrich A, Lyuri D (2019) Izmeneniya dinamiki agrarnykh ugodiy Rossii v 19902014 gg. [Changes in the dynamics of agrarian lands in Russia in 19902014]. Izvestiya Rossiyskoy Akademii Nauk Seriya geograficheskaya [Proceedings of the Russian Academy of Sciences Geographical Series] 0(3):64–77. (In Russian) Nguyen H, Hölzel N, Völker A, et al (2018) Patterns and Determinants of Post-Soviet Cropland Abandonment in the Western Siberian Grain Belt. Remote Sens 10. https://doi.org/10.3390/rs10121973 Ogorodnov E (1971) Atlas Tyumenskoi oblasti [Atlas of the Tyumen Region], vol 1. Nauka [Science], Moscow. (In Russian) Oliveira-Junior ES, Tang Y, van den Berg SJP, et al (2018) The impact of water hyacinth ( Eichhornia crassipes ) on greenhouse gas emission and nutrient mobilization depends on rooting and plant coverage. Aquatic Botany 145:1–9. https://doi.org/10.1016/j.aquabot.2017.11.005 Pashnina E, Simakova T (2017) Analiz Ispol’zovaniya Melioriruyemykh Zemel’ Tarmanskogo Bolotnogo Massiva Tyumenskoy Oblasti [Analysis of the use of improved lands of the Tarman Wetland Complex of the Tyumen Region]. Gosudarstvennyy agrarnyy universitet Severnogo Zauralsya [State Agrarian University of Northern Trans-Urals], pp 116–119 (In Russian) Phan TN, Kuch V, Lehnert LW (2020) Land cover classification using Google Earth Engine and random forest classifierthe role of image composition. Remote Sensing 12(15):2411. https://doi.org/10.3390/rs12152411 Prat-Guitart N, Belcher C, Thompson D, et al (2017) Finescale distribution of moisture in the surface of a degraded blanket fen and its effects on the potential spread of smouldering fire. Ecohydrology 10. https://doi.org/10.1002/eco.1898 Provost GL, Badenhausser I, Bagousse-Pinguet YL, et al (2020) Landuse history impacts functional diversity across multiple trophic groups. Proceedings of the National Academy of Sciences 117:1573–1579. https://doi.org/10.1073/pnas.1910023117 Rada NE, Liefert W, Liefert O (2020) Evaluating Agricultural Productivity and Policy in Russia. Journal of Agricultural Economics https://doi.org/10.1111/1477-9552.12338 Rappaport DI, Tambosi L, Metzger J (2015) A landscape triage approach: combining spatial and temporal dynamics to prioritize restoration and conservation. Journal of Applied Ecology 52:590–601. https://doi.org/10.1111/1365-2664.12405 Redon M, Luque S, Gosselin F, et al (2014) Is generalisation of uneven-aged management in mountain forests the key to improve biodiversity conservation within forest landscape mosaics? Annals of Forest Science 71(7):751– 760. https://doi.org/10.1007/s13595-014-0371-7 Ritson JP, Alderson DM, Robinson CH, et al (2021) Towards a microbial process-based understanding of the resilience of peatland ecosystem service provisioning A research agenda. Science of The Total Environment 759:143467. https://doi.org/10.1016/j.scitotenv.2020.143467 Rouse JW, Haas RH, Schell JA, et al (1974) Monitoring vegetation systems in the Great Plains with ERTS 351(1):309–317. Sali M, Piaser E, Boschetti M, et al (2021) A burned area mapping algorithm for Sentinel-2 data based on approximate reasoning and region growing. Remote Sensing 13(11):2214. https://doi.org/10.3390/rs13112214, number: 11 Publisher: Multidisciplinary Digital Publishing Institute Selmy SAH, Kucher DE, Mozgeris G, et al (2023) Detecting, analyzing, and predicting land use/land cover (LULC) changes in arid regions using Landsat images, CA-Markov Hybrid Model, and GIS techniques. Remote Sensing 15(23):5522. https://doi.org/10.3390/rs15235522 Shafizadeh-Moghadam H, Khazaei M, Alavipanah SK, et al (2021) Google Earth Engine for large-scale land use and land cover mapping: an object-based classification approach using spectral, textural and topographical factors. GIScience & Remote Sensing 58(6):914– 928. https://doi.org/10.1080/15481603.2021.1947623 Shcherbakova IK (2021) Agrarian reforms of Russia at the turn of the XIXXX centuries in the coverage of economists of the XX century. Vestnik Universiteta https://doi.org/10.26425/1816-4277-2021-5-141-144 Sheludkov A, Kamp J, Müller D (2020) Decreasing labor intensity in agriculture and the accessibility of major cities shape the rural population decline in postsocialist Russia. Eurasian Geography and Economics 62:481–506. https://doi.org/10.1080/15387216.2020.1822751 Simakova T, Simakov A, Ivanova A (2023) Monitoring melioryruemykh zemel’ s ispol’zovaniem landshaftno-ekologicheskogo podkhoda [Monitoring of improved lands using a landscape-ecological approach]. Vestnik Voronezhskogo gosudarstvennogo agrarnogo universiteta [Bulletin of the Voronezh State Agrarian University] 16(3):78. (In Russian) Singh S, Gupta R, Kumari M, et al (2015) Nontarget effects of chemical pesticides and biological pesticide on rhizospheric microbial community structure and function in Vigna radiata. Environmental Science and Pollution Research 22(15):11290–11300. https://doi.org/10.1007/s11356-015-4341-x Sirin A, Medvedeva M, Korotkov V, et al (2021) Addressing Peatland Rewetting in Russian Federation Climate Reporting. Land 10(11):1200. https://doi.org/10.3390/land10111200 Sirin AA, Medvedeva MA, Makarov DA, et al (2020) Multispectral satellite based monitoring of land cover change and associated fire reduction after large-scale peatland rewetting following the 2010 peat fires in Moscow Region (Russia). Ecological Engineering 158:106044. https://doi.org/10.1016/j.ecoleng.2020.106044 Solodovnikov A (2021) Rol’ zakaznika federal’nogo znacheniya "Tyumensky" v sokhranenii vidovogo raznoobraziya flory i fauny Nizhnetavdinskogo rayona Tyumenskoy oblasti [Role of the federal significance reserve "Tyumensky" in the preservation of species diversity of flora and fauna of the Nizhnetavdinsky district of the Tyumen region]. Zametki Uchenogo [Notes of the Scientist] (3-1):378–385. (In Russian) Struik Q, Oliveira Junior ES, Veraart AJ, et al (2022) Methane emissions through water hyacinth are controlled by plant traits and environmental conditions. Aquatic Botany 183:103574. https://doi.org/10.1016/j.aquabot.2022.103574 Taillardat P, Thompson BS, Garneau M, et al (2020) Climate change mitigation potential of wetlands and the cost-effectiveness of their restoration. Interface Focus 10. https://doi.org/10.1098/rsfs.2019.0129 Tassi A, Gigante D, Modica G, et al (2021) Pixel- vs. object-based Landsat 8 data classification in Google Earth Engine using random forest: the case study of Maiella National Park. Remote Sensing 13(12):2299. https://doi.org/10.3390/rs13122299, URL https://www.mdpi.com/20724292/13/12/2299 Teluguntla P, Thenkabail PS, Oliphant A, et al (2018) A 30-m Landsatderived cropland extent product of Australia and China using random forest machine learning algorithm on Google Earth Engine cloud computing platform. ISPRS Journal of Photogrammetry and Remote Sensing 144:325–340. https://doi.org/10.1016/j.isprsjprs.2018.07.017 Turetsky M, Donahue W, Benscoter B (2011) Experimental drying intensifies burning and carbon losses in a northern peatland. Nature communications 2. https://doi.org/10.1038/ncomms1523 Ulanov A, Smirnova AV, Ulanov N (2023) Forest Stands Formation on Exhausted Peat Bog in the North-East of the European Part of Russia. Lesovedenie [Forest Science] https://doi.org/10.31857/s0024114823040137 de Waard F, Connolly J, Barthelmes A, et al (2024) Remote sensing of peatland degradation in temperate and boreal climate zones A review of the potentials, gaps, and challenges. Ecological Indicators 166:112437. https://doi.org/10.1016/j.ecolind.2024.112437 Yang Q, Li X, Shi X (2008) Cellular automata for simulating land use changes based on support vector machines. Computers & Geosciences 34(6):592– 602. https://doi.org/10.1016/j.cageo.2007.08.003 Zhao B, Zhuang Q (2023) Peatlands and their carbon dynamics in northern high latitudes from 1990 to 2300: a process-based biogeochemistry model analysis. Biogeosciences https://doi.org/10.5194/bg-20-251-2023 Åhlén I, Thorslund J, Hambäck P, et al (2022) Wetland position in the landscape: Impact on water storage and flood buffering. Ecohydrology 15(7):e2458. https://doi.org/10.1002/eco.2458 Additional Declarations The authors declare no competing interests. Supplementary Files AppendixA.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7012873","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":478598174,"identity":"1ae34338-c4bc-4c3b-b90f-3c8e2f2d422e","order_by":0,"name":"Vladimir Ivanov","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwUlEQVRIiWNgGAWjYBACNghlw8NHqpY0HjZSLTvMQLwWPonkZw8+tp2XYWM/e4C5oqaOwXxGAgGHSaSZG85su83DxpOXwHjm2GEGmRuEtPAcMJPmOQPUwpBjwNjAdoBBQoKgluPfgFrO8bDxvwFq+VdHhBb2HqAtFQd42CSAtjS2MROlpUxyRkUyUMsbg4ONfYd5JHge4Nci38y+TeKDgZ09P3+O4cOGb3VyEuwEbEEBB4CYhwT1o2AUjIJRMApwAQAHtjLkd0KOBQAAAABJRU5ErkJggg==","orcid":"","institution":"Institute of Forest Science, Russian Academy of Sciences","correspondingAuthor":true,"prefix":"","firstName":"Vladimir","middleName":"","lastName":"Ivanov","suffix":""},{"id":478598175,"identity":"ae008e0f-66da-4774-be56-d5fd6df1a97f","order_by":1,"name":"Evgeniya Soldatova","email":"","orcid":"","institution":"Institute of Forest Science, Russian Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Evgeniya","middleName":"","lastName":"Soldatova","suffix":""},{"id":478598176,"identity":"ad3ff264-c719-4207-80b1-d1b583c0d50c","order_by":2,"name":"Milyaev Ivan","email":"","orcid":"","institution":"University of Tyumen","correspondingAuthor":false,"prefix":"","firstName":"Milyaev","middleName":"","lastName":"Ivan","suffix":""}],"badges":[],"createdAt":"2025-06-30 16:48:10","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7012873/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7012873/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85849061,"identity":"a8f5e1de-d942-40ea-b557-79eb8af9914f","added_by":"auto","created_at":"2025-07-02 10:07:20","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":600176,"visible":true,"origin":"","legend":"\u003cp\u003eStudy Area Location\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/d12fcb563093a5f56cb51768.png"},{"id":85850501,"identity":"1a9feb5d-c003-43e6-9184-fb73e937d601","added_by":"auto","created_at":"2025-07-02 10:15:20","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":183757,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of the data processing\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/995e42f124b9816a18241349.png"},{"id":85849060,"identity":"f59f2ebb-f9bd-470b-bd4a-3747ae8497d6","added_by":"auto","created_at":"2025-07-02 10:07:20","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":54703,"visible":true,"origin":"","legend":"\u003cp\u003eDynamics of LULC of Tarmanskoye fen from 1984 to 2024 years\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/2b6ae41247f8ed6627f4fa42.png"},{"id":85849065,"identity":"7c28baa8-17d1-4d30-bf44-9c566c40bd1f","added_by":"auto","created_at":"2025-07-02 10:07:20","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":381438,"visible":true,"origin":"","legend":"\u003cp\u003eLand Cover Dynamics Map\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/f1a1b881c41230dbc53f27a2.png"},{"id":85850504,"identity":"91493b6c-e219-49cd-8e07-7b3667a72563","added_by":"auto","created_at":"2025-07-02 10:15:20","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":183127,"visible":true,"origin":"","legend":"\u003cp\u003eFrequency of Land Cover Class Changes in Individual Polygons\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/4235525559fa5cb88b7e8c3c.png"},{"id":85851168,"identity":"373fca0c-bb4b-4657-ac49-ca65f5aaaf74","added_by":"auto","created_at":"2025-07-02 10:23:20","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":275256,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of Polygons by Percentage of Area Covered\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/1e5925fbbf90ce282ab99a8d.png"},{"id":85849070,"identity":"130ec3d5-f868-49eb-8716-ff236c145c31","added_by":"auto","created_at":"2025-07-02 10:07:20","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":69077,"visible":true,"origin":"","legend":"\u003cp\u003eLand Cover Dynamics of Drained Peatland Areas\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/7afcf8e878c1786c12069b24.png"},{"id":85849077,"identity":"6bfc3de7-527a-4d8a-b8b2-06a104846c02","added_by":"auto","created_at":"2025-07-02 10:07:20","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":85009,"visible":true,"origin":"","legend":"\u003cp\u003eAnnual Frequency and Area of Successional Transitions\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/daaf2f23efa913732c1c4f5f.png"},{"id":85850508,"identity":"0c2ebf99-ff69-4b1b-8140-0e250abf5896","added_by":"auto","created_at":"2025-07-02 10:15:20","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":219152,"visible":true,"origin":"","legend":"\u003cp\u003eChanges in Spectral Characteristics of the Studied Ecosystems\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/5d31855a0c9085e0a1e3bbd3.png"},{"id":85850509,"identity":"9cfa954b-9a65-4e37-a3f2-0f733fda0b56","added_by":"auto","created_at":"2025-07-02 10:15:20","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":97666,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted Future Landscape Conditions for 2034, 2044, and 2054. (Percentages are cumulative relative to 2024)\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/f8e673d323ec03b3be531f25.png"},{"id":85853185,"identity":"09e134e1-1c78-43c4-9a40-c7553f9c84e3","added_by":"auto","created_at":"2025-07-02 10:39:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3051703,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/9b4b030a-e8b2-4de1-95d2-e3753c888e63.pdf"},{"id":85851166,"identity":"b0dfca70-d481-43f3-9278-019c0fe35be4","added_by":"auto","created_at":"2025-07-02 10:23:20","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1955721,"visible":true,"origin":"","legend":"","description":"","filename":"AppendixA.docx","url":"https://assets-eu.researchsquare.com/files/rs-7012873/v1/c98d260279178ac0d086370a.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eSuccessional pathways after peatland draining: remote sensing and predictive modelling of landscape dynamics\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eWetlands are valuable ecosystems that perform multiple essential ecological functions. Their widespread geographic distribution and diversity make them habitats for a vast number of plant and animal species (Ma et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Ritson et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The geomorphological structure and organic composition of wetland soils enable them to store large volumes of water, preventing erosion and positively influencing the frequency and intensity of floods (\u0026Aring;hl\u0026eacute;n et al., \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; McLaughlin and Cohen, \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). The vegetation, structure, and microbiological activity of wetlands contribute to their ability to retain and absorb pollutants from water. Additionally, their role in climate changerelated processes is noteworthy. Peatlands cover approximately 3% of the Earth\u0026rsquo;s land surface but store one-third of the planets soil carbon and exert an overall cooling effect on the atmosphere, at least at a local level (Taillardat et al., \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Tassi et al., \u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). On the other hand, wetlands also have significant potential as sources of carbon dioxide, methane, nitrous oxide, and dissolved organic compounds (Hugelius et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Leng et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Loisel et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The risk of wetlands transitioning from carbon sinks to emitters of climate-active gases increases with the intensification of anthropogenic factors, such as land-use changes and the growing incidence of landscape fires (Leifeld et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Mickler, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Turetsky et al., \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAn increasing number of studies examining the effects of land-use changes on ecosystems have highlighted that landscape recovery dynamics, plant community biodiversity (Gomes et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Luan and Liu, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Malek et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) and ecosystem services (Gomes et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Luan and Liu, \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), including carbon sequestration and storage (Kurganova et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; McDaniel et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), are significantly influenced by the nature and intensity of previous land use (Fu et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Provost et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Complementing these findings with recent analyses of contemporary and paleoecological data that identify peatlands as crucial determinants of the future global climate, the detailed study of peatland land-cover and land-use dynamics and their impact on ecosystem services has become more relevant than ever (Edvardsson et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhao and Zhuang, \u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The obtained data will provide a foundation for developing policies for the management and conservation of peatlands, aimed at mitigating the effects of climate change in both natural and previously industrially and agriculturally utilized areas (Chen et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Humpen\u0026ouml;der et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Rappaport et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Moreover, given the challenges in accurately assessing the extent of degraded peatlands, current data remain approximate (Sirin et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), while precise estimates of the area and current condition of disturbed lands are either fragmented or entirely absent-this is particularly true for the southern part of Western Siberia.\u003c/p\u003e \u003cp\u003eOver the past 200 years, large-scale land-use changes have occurred in northeastern Eurasia, affecting various ecosystems, including peatlands (Monier et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Malek et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). From the extensive settlement-driven land cultivation in the 19th century to the collectivization and prolonged extensive agricultural practices of the 20th century, followed by the collapse of the Soviet Union and the large-scale abandonment of developed lands in the late 20th and early 21st centuries, vast territories have undergone profound anthropogenic transformation (Lapytskaya, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Lysenko, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Rada et al., \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Shcherbakova, \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Drainage reclamation of wetlands was frequently employed to expand agricultural lands (Baisheva et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Bruisch, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Maslov and Maslova, \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, plowing was not the sole reason for wetland drainage. Since many thermal power plants operated on peat, extensive peatlands were drained for extraction and use as fuel for electricity and heat production (Akhmeteva et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Bruisch, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Moskalenko et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The subsequent development of extraction and transportation technologies in the oil and gas sector rendered natural gas a more economically viable energy source, leading to a gradual decline in peat extraction (Bondar, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Karpov, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Peatland reclamation, primarily for agricultural purposes, was conducted in the Soviet Union until around 1990 but has since ceased (Minaeva and Sirin, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Sirin et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), resulting in most disturbed peatlands undergoing spontaneous succession.\u003c/p\u003e \u003cp\u003eIn Western Siberia, agriculture has historically been highly dependent on state support, leading to a sharp response to socio-political changes (Nekrich and Lyuri, \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Nguyen et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Peat was widely used both as fertilizer and as an energy source, leading to widespread peat extraction sites, most of which were abandoned without proper reclamation. Consequently, large areas of former agricultural lands interwoven with peat extraction sites continue to form fragmented ecosystems under the influence of natural, anthropogenic, and post-agricultural factors, exhibiting diverse successional pathways (Lvov, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Filippova et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Krasnov et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). The situation is further complicated by the expansion of urban peripheries and the emergence of new settlements on lands previously used for agriculture or peat extraction.\u003c/p\u003e \u003cp\u003eThe Tarmanskoe peatland, located west of Tyumen and partially within the city limits, exemplifies a site where all these processes manifest with varying intensity. According to detailed surveys, the exploration of the Tarmanskoe peat deposit was the largest in the history of Soviet peat industry development. Drainage of the Tarmanskoe wetland began in the 1960s and continued until the 1970s, primarily for peat extraction to supply CHPP-1. In addition to peat mining, drained lands were used for arable farming, haymaking, and livestock grazing. Currently, most of the reclaimed area is no longer managed and is undergoing various types of succession, primarily afforestation and shrub encroachment, with localized secondary waterlogging.\u003c/p\u003e \u003cp\u003eConsidering the environmental conditions and the complex, dynamic history of industrial and agricultural land use, we propose two hypotheses regarding existing and future ecosystem development trends. First, ecosystems on former peat extraction sites will remain fragmented due to the combined effects of natural and anthropogenic factors, including hydrological changes and climate impacts. Second, the predominant successional pathway on previously drained lands will be forest encroachment, with the possibility of shifting toward wetland formation depending on hydrological conditions and human intervention. Thus, this study aims to identify the main trends and patterns of ecosystem dynamics in the Tarmanskoe peatland, shaped by anthropogenic transformation and natural succession, and to model potential landscape changes over the next 30 years using the hybrid CA-Markov model.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study Area\u003c/h2\u003e \u003cp\u003eThe Tarmanskoe peatland is located within the West Siberian Plain, on the southern periphery of the subtaiga zone, on the second abovefloodplain terrace of the left bank of the Tura River (Khozyainova et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). The terrain in this area is predominantly flat, with weakly expressed features of fluvial erosion and accumulation, which contributed to the development of wetland conditions. The Tarmanskoe peatland extends from west to east for 130 km, with a width varying between 10 and 50 km.\u003c/p\u003e \u003cp\u003eThe region is characterized by an extensive distribution of lakes across its territory, with the highest concentration in the central part of the wetland, which has influenced peatland moisture levels and facilitated peat accumulation. The lakes within the Tarmanskoe complex are predominantly shallow and overgrown, with an average depth of approximately 1.5 meters. Their basins often have indistinct and irregularly shaped shorelines, which are frequently swampy.\u003c/p\u003e \u003cp\u003eIn addition to lakes, the landscape features numerous drylands distributed throughout the peatland (Giprotorfrazvedka, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1955\u003c/span\u003e). These drylands are covered by mixed coniferous and small-leaved forests, primarily composed of birch and pine, or used for agricultural purposes, mainly hayfields and pastures, with arable land being less common.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe Tarmanskoe peatland is underlain by heterogeneous, predominantly loamy alluvial deposits of the Upper Pleistocene Sartansk horizon (Ogorodnov, \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e1971\u003c/span\u003e). The clayey composition of the underlying surface is considered one of the most significant factors contributing to the formation of the Tarmanskoe peatland (Giprotorfrazvedka, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1955\u003c/span\u003e). Water supply is primarily sourced from atmospheric precipitation and runoff from the catchment area, as well as unconfined groundwater.\u003c/p\u003e \u003cp\u003eFrom a landscape perspective, the undisturbed portion of the peatland is predominantly occupied by flat, level, and slightly hummocky sedge-hypnum fens with floating mats, as well as sedge-buckbean fens, with an average peat deposit thickness of approximately 2 meters, classifying them as shallow-peat fens. The climate of the region is continental, with an average annual temperature of +\u0026thinsp;1.5 ͦ C. The warmest month, July, has a mean temperature of +\u0026thinsp;18.5 ͦ C, with an annual maximum reaching\u0026thinsp;+\u0026thinsp;37 ͦ C. The coldest month, January, has a mean temperature of -16.4 ͦ C, with an annual minimum of -40 ͦ C and an absolute minimum of -50 ͦ C. The total annual precipitation is approximately 460 mm, with the majority occurring during the warm season (Kazakov, 2024).\u003c/p\u003e \u003cp\u003eSince the collapse of the Soviet Union, the drainage network has not been maintained, leading to the partial failure of some channels, frequent secondary waterlogging, and the flooding of residential areas and agricultural lands. Peat extraction is currently conducted on a limited scale, primarily for agricultural purposes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Methods\u003c/h2\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1. Data Sources, Processing, and Accuracy Assessment\u003c/h2\u003e \u003cp\u003eFor remote sensing data analysis, we utilized multispectral satellite imagery from the Landsat TM, ETM+, and OLI missions, provided by the U.S. Geological Survey (USGS). These sensors measure Earth\u0026rsquo;s reflected radiation in the visible and infrared spectral ranges with a spatial resolution of 30 meters (Ihlen and Zanter, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Landsat mission data have been widely used in numerous studies on land use and land cover change (LULCC) (de Waard et al., \u003cspan citationid=\"CR112\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kurbatova et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sirin et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor time series analysis, we selected atmospherically corrected, low-cloud Landsat Collection 2 Level 2 Surface Reflectance (SR) images for the longest possible observation period. Image processing was conducted using Google Earth Engine (GEE) (Gorelick et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGoogle Earth Engine is a cloud-based geospatial analysis platform that leverages Google\u0026rsquo;s computational power to address critical global issues, including deforestation, droughts, natural disasters, epidemics, food security, water resource management, climate monitoring, and environmental protection (Gorelick et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2. Data Selection and Classification Methodology\u003c/h2\u003e \u003cp\u003eDespite the overall availability of cloud-free and low-cloud satellite images, for certain years, high-quality images were unavailable, preventing a uniform temporal resolution between observation periods. To address this, image selection involved spatial filtering, cloud masking, and, when multiple images were available, their merging into unified median mosaics (Habib and Connolly, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Phan et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). A total of 67 images from the period 1984\u0026ndash;2024 (captured between May and September) were obtained. Given the absence of satellite imagery for certain years and variations in the quality of available cloud-free images, we selected images with a predominantly threeyear interval, which is considered optimal for time-series analysis (Habib and Connolly, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe then conducted object-based classification for two selected images, corresponding to 1984 and 2024, as the initial and final observation points. The first stage involved generating a set of training points for supervised classification through expert assessment and the interpretation of false-color composites (SWIR-NIR-Red), which provide detailed visual information on land cover (Asare et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sali et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). High-resolution Google Earth satellite imagery and Soviet-era topographic maps were used as auxiliary materials for compiling the training and validation datasets. To refine classification accuracy and landscape characterization, we incorporated detailed multispectral and LiDAR data obtained using Unmanned Aerial Vehicle (UAV) surveys conducted in summer 2024. UAV surveys were conducted for five representative landscapes, which include all the land cover classes referenced in this study. The models of the drone and cameras used in this study are described in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSpecifications of Technical Equipment Used in the UAV Survey\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel Name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEquipment Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePurpose of Use\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDJI Matrice 300 RTK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUAV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePlatform for transporting and stabilizing remote sensing payloads\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDJI Zenmuse L1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLiDAR sensor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAcquisition of elevation data and generation of a Digital Elevation Model (DEM)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMicaSense RedEdge MX\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMultispectral camera\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAcquisition of multispectral imagery and generation of orthomosaics and vegetation indices (e.g., NDVI, NDMI)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eA total of eight land cover classes were identified, with three additional classes (settlements, mixed forests on former peat extraction sites, and driedup lake areas) manually delineated at later stages. The dataset was randomly split into training (70%) and validation (30%) subsets. For model training, we used the spectral bands blue, green, red, near-infrared (NIR), and both short-wave infrared (SWIR) bands, following previous methodologies (Habib and Connolly, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe second stage involved applying image segmentation using the Simple Non-Iterative Clustering (SNIC) algorithm, integrated into Google Earth Engine. This technique partitions the image into superpixels-groups of pixels with similar characteristics-enhancing landscape contour detection (Gxokwe et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Shafizadeh-Moghadam et al., \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Tassi et al., \u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor land cover classification, we employed the widely used random forest (RF) algorithm-a robust ensemble learning method known for its resistance to overfitting and high classification accuracy (Jin et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Millard and Richardson, \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Phan et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Teluguntla et al., \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The model utilized 100 decision trees, a number determined based on prior research recommendations (Ghimire et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Phan et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe classification results underwent accuracy assessment using confusion matrix calculations and Kappa coefficient analysis (Congalton and Green, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.2.3. Vector Data Processing and Mapping\u003c/h2\u003e \u003cp\u003eThe classified segments were merged into a single layer using the intersect tool in ArcGIS Pro 3.0 (Esri), followed by visual inspection and manual correction of boundaries and land cover classifications assigned by the algorithm. Based on these layers, additional vector layers were manually created to represent the study area\u0026rsquo;s condition at different time intervals (2\u0026ndash;3 years apart) by visually interpreting satellite imagery. A total of 15 vector layers were produced, each containing 10\u0026ndash;11 land cover classes, as the exposed peat class was present only in the first half of the study period. A list of land cover classes and their corresponding descriptions is provided in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescription of LULC Classes\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand use and land cover classes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDescription\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWater bodies\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWater surface of lakes, ponds and reservoirs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShallow areas of lakes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePeriodically flooded areas of lake basins\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFen\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWaterlogged areas accumulating peat, primarily occupied by herbaceous vegetation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeadow\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDrained areas with diverse meadow vegetation, sometimes including shrubs\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArable land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAreas used for the cultivation of agricultural crops\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eActive peat extraction sites\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePeat extraction sites that are operational as of the date of the imagery\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWaterlogged meadows and hayfields on peat extraction sites\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFormer extraction sites covered with meadow vegetation with varying drainage\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixed forest on peat extraction sites\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFormer extraction sites primarily covered with broadleaf and mixed forest vegetation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmall-leavedandmixed\u003c/p\u003e \u003cp\u003eforests\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAreas with broadleaf and mixed (subtaiga) vegetation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePine forests\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAreas with predominantly coniferous (pine) forest vegetation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResidential areas\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSuburban and cottage development\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe integration of topographic maps from 1976\u0026ndash;1981 allowed for more precise delineation of the drained areas.\u003c/p\u003e \u003cp\u003eBy analyzing multi-temporal land cover data, a landscape dynamics map of the Tarmanskoe peatland was created, providing a detailed representation of the changes that occurred, including their nature and spatial distribution from 1984 to 2024. Similar approaches have been employed in previous studies (Ludwig et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The accuracy of the classification is presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eClassification accuracy assessment for land cover maps.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraining Overall Accuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTest Accuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKappa\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1984\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.82\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.2.4. Spectral Data Analysis\u003c/h2\u003e \u003cp\u003eSpectral channel dynamics of Landsat satellite imagery were analyzed over the period (June-August) for four land cover classes:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eMixed forests on former peat extraction sites\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eWaterlogged and meadow grasslands on former peat extraction sites\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eSmall-leaved and mixed forests\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eWetlands\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThese classes were chosen because the first two represent anthropogenically modified peatlands, while the last two cover areas minimally affected or unaffected by human activity. For mixed forests and waterlogged meadows on former peat extraction sites, areas with observed peat extraction in 1984 were selected.\u003c/p\u003e \u003cp\u003eSpectral values from six bands (Blue, Green, Red, NIR, SWIR1, SWIR2) were extracted and three spectral indices computed (computational formulas are provided in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e):\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eNormalized Difference Vegetation Index (NDVI);\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eNormalized Difference Water Index (NDWI);\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eNormalized Difference Moisture Index (NDMI).\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSpectral indices calculating formulas\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpectral Index\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFormula\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReference\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNDVI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eNIR\u0026thinsp;\u0026minus;\u0026thinsp;Red\u003c/span\u003e\u003c/p\u003e \u003cp\u003eNIR\u0026thinsp;+\u0026thinsp;Red\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Rouse et al., \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e1974\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNDWI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eGreen\u0026thinsp;\u0026minus;\u0026thinsp;NIR\u003c/span\u003e\u003c/p\u003e \u003cp\u003eGreen\u0026thinsp;+\u0026thinsp;NIR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Gao, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1996\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNDMI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eNIR\u0026thinsp;\u0026minus;\u0026thinsp;SWIR1\u003c/span\u003e\u003c/p\u003e \u003cp\u003eNIR\u0026thinsp;+\u0026thinsp;SWIR1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Jin and Sader, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.2.5. Time Series Analysis and Statistical Validation\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe extracted time series data for spectral bands and vegetation indices were analyzed and compared. To assess the statistical significance of trends over specific time periods, we applied the non-parametric Mann-Kendall test (Kendall, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e1948\u003c/span\u003e; Mann, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e1945\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.2.6. CA-Markov Analysis for Land Use Change Modelling\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eTo predict future land use changes for 2034, 2044, and 2054, we employed a hybrid CA-Markov model. This method calculates a transition probability matrix using a spatial neighborhood influence algorithm and is widely used for land use and land cover (LULC) forecasting based on past trends and inherent spatial patterns (Guan et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Selmy et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn recent years, the application of the hybrid CA-Markov model for landscape change prediction has significantly increased (Eva et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kumar et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Selmy et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe model was trained using 15 land cover maps (produced at 2\u0026ndash;3 year intervals). The simulation was conducted in Python, utilizing the GeoPandas, Matplotlib, and NumPy libraries (Harris et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Van den Bossche et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e2.2.7. Model Validation\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eModel accuracy was evaluated by comparing predicted land cover changes with observed changes during the validation period. The results of accuracy assessment are presented in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. We used standard Kappa statistics and overall classification accuracy to assess model performance. The Kappa coefficient helps distinguish quantitative errors and spatial misclassification errors between two categorical maps (Selmy et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Singh et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eKappa coefficients for land cover maps at 10-year intervals.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKappa coefficient\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1993\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.915\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.711\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.768\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.877\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Limitations and Uncertainties of the Methodology\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eDespite the methodology employed in this study being widespread and well-established, several limitations should be acknowledged. First, the classification accuracy is inherently influenced by the quality and temporal consistency of the remote sensing datasets. Seasonal differences in vegetation phenology and water levels may affect spectral signatures, especially in dynamic wetland environments, potentially leading to misclassifications. Although field validation was conducted during multiple seasons, some areas remained inaccessible due to waterlogging, limiting ground-truth observations. The absence of historical ground-truth data presents a challenge for validating past imagery, thereby introducing uncertainty in the assessment of satellite-detected changes.\u003c/p\u003e \u003cp\u003eThe algorithms used for classification are sensitive to the quality and representativeness of the training data. Given the heterogeneity of post-extraction peatland surfaces and the ongoing secondary succession processes, the training samples may not fully capture all transitional states, which can reduce classification reliability in ecotonal zones.\u003c/p\u003e \u003cp\u003eLastly, the long-term analysis spanning four decades\u0026rsquo; years relies in part on archival satellite imagery, which varies in spatial, spectral, and radiometric quality. These discrepancies can introduce temporal inconsistencies, despite preprocessing efforts such as atmospheric correction and co-registration.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.1. General Dynamics of Natural Complexes\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e illustrates the overall dynamics of land cover and land use over the study period. Active peat extraction areas have shown a steady decline in extent since the earliest time interval. Drained lands without evidence of active peat extraction have been observed throughout the entire study period. The area of croplands has gradually decreased, with some fluctuations, although at no point they were entirely absent. Residential zones have exhibited continuous expansion, partially replacing drained and natural grassland ecosystems and, less frequently, forests.\u003c/p\u003e \u003cp\u003eWater bodies, primarily lakes, experienced minor fluctuations in total area between 1984 and 2013. However, from that point onward, a declining trend emerged, reaching its lowest recorded extent in 2024. Natural wetland areas and pine forests have remained largely unchanged throughout the observation period. Overall, the most substantial land cover changes were recorded between 1984 and 2003.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eAmong the general patterns of land cover dynamics in the Tarmanskoe peatland, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the expansion of mixed-species forest cover on former peat extraction sites is particularly pronounced. This trend is especially noticeable south of the Bolshoye and Sredneye Tarmanskoe lakes in the central part of the study area, as well as in the northeastern and eastern regions, where it occurs in the vicinity of undisturbed forests and wetlands. However, the majority of former peat extraction sites are now occupied by waterlogged and meadow grasslands, covering an area of 126 km\u003csup\u003e2\u003c/sup\u003e (or 59.2% of the total peat extraction area), compared to 73 km\u003csup\u003e2\u003c/sup\u003e of forest (34.5%). Minor areas of flooding, amounting to 0.17%, are observed in the eastern part of the study area.\u003c/p\u003e \u003cp\u003eAs previously noted, the reduction in surface water area has become a significant phenomenon due to lake shallowing and wetland expansion. This process is most evident in the lakes located in the eastern part of the study area, as well as in Sredneye Tarmanskoe Lake, where a substantial portion has dried up. According to calculations, nearly 49% of the total lake area within the study region has undergone shallowing.\u003c/p\u003e \u003cp\u003eBased on the data presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, most of the pasture, hayfields, and croplands (11.2% of the study area or 92 km\u003csup\u003e2\u003c/sup\u003e) are concentrated in the southwestern part of the region, where numerous fragmented patches of secondary small-leaved and mixed forests have developed on former agricultural lands. The northwestern territories are dominated by stable, undisturbed landscapes consisting of wetlands, coniferous, mixed, and small-leaved forests.\u003c/p\u003e \u003cp\u003eSettlements are concentrated in the southern part of the study area, closer to the city of Tyumen and along the highway running parallel to the Tura River floodplain. The majority of these settlements have been established on land previously used for agriculture (49% of total settlement area), peat extraction (18.2%), and forests (11.9%).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe highest frequency of land cover changes occurs in areas subject to economic activity, particularly peat extraction and agriculture (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Pasture, hayfields, and croplands exhibit the greatest variability, as periodic shifts in land use practices are observed. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e serves as an indicator of the stability of landscape conditions within the study area.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe map of polygon distribution within the study area highlights a pronounced landscape mosaic, with a substantial proportion of polygons belonging to classes with minimal areas (up to 0.41% of the total area).\u003c/p\u003e \u003cp\u003eThe largest polygons correspond to natural wetland, forest, and lake areas, primarily concentrated in the northern part of the study site, where no drainage or peat extraction has taken place. These polygons cover between 4% and 18% of the total area. An exception is found in the eastern part of the site, where abandoned, unforested peat extraction sites account for 12% of the total area.\u003c/p\u003e \u003cp\u003eIt is important to note that this area is conditionally homogeneous, as drainage channels were not classified separately. As a result, the map does not reflect the subdivision into checks, which, in reality, segment the area into smaller components.\u003c/p\u003e \u003cp\u003eThe highest degree of land fragmentation is observed in the western and central parts of the study area, where peat extraction was most intensive, and multiple attempts at agriculture were made. Some of these agricultural practices were partially successful and have persisted to the present day. In these areas, polygons of the smallest size classes account for 65.72% and 68.95% of the total area, respectively.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e presents the dynamics of areas subjected to drainage. The largest extent of active peat extraction was recorded in 1984, followed by a steady decline, ultimately leading to the complete cessation of peat extraction by 1997.\u003c/p\u003e \u003cp\u003eOver time, mixed forests progressively replaced meadow vegetation. Until 2011, this transition occurred at a rapid pace, as evident from the data shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e, but thereafter, the trend persisted while slowing considerably. By 2024, the area of meadows on former peat extraction sites reached 15.2%, which is nearly the same as in 1984 when they covered 15.9% of the study area. In contrast, mixed forests expanded significantly, now occupying 9.1% of the total area in 2024, whereas at the beginning of the study period, they were just emerging and covered less than 0.5% of the territory.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e illustrates the frequency and total area of successional changes that occurred over the study period. The highest frequency and extent of changes were recorded between 1987 and 2006, followed by a period of relative stabilization.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Variability of Spectral Characteristics\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eAs expected, land cover classes unaffected by human activity during the study period exhibit relative stability in their spectral channel values (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Only wetlands show a slight downward trend in near-infrared reflectance (NIR) and an even weaker increase in shortwave infrared reflectance (SWIR1, SWIR2).\u003c/p\u003e \u003cp\u003eFor classes resulting from secondary succession on former peat extraction sites, an increase in NIR reflectance is observed until 1998\u0026ndash;2000, after which values stabilize. Waterlogged and meadow grasslands exhibit greater fluctuations in this spectral range compared to mixed forests. Additionally, after 2015, there is a noticeable decline in NIR reflectance and an increase in SWIR reflectance, influencing trends in NDVI and NDMI.\u003c/p\u003e \u003cp\u003eIn mixed forests, all spectral channel values have remained stable since the early 21st century, a trend also reflected across all spectral indices. Between 1984 and 2000, NDVI values in secondary mixed forests increased from 0.15 to 0.40, NDMI rose from \u0026minus;\u0026thinsp;0.10 to 0.24, while NDWI declined from \u0026minus;\u0026thinsp;0.20 to -0.38. In waterlogged and meadow grasslands, NDVI increased from 0.15 to 0.30, NDMI from \u0026minus;\u0026thinsp;0.10 to 0.15, and NDWI decreased from \u0026minus;\u0026thinsp;0.15 to -0.30.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe Mann-Kendall statistical test for trend significance in NDVI and NDMI from 2016 to 2024 for the class of waterlogged and meadow grasslands on former peat extraction sites yields a p-value near the significance threshold (p\u0026thinsp;=\u0026thinsp;0.04) for both indices-NDVI (declining from 0.31 to 0.21) and NDMI (declining from 0.13 to -0.01). Given this result and the short observation period, the downward trend cannot yet be confirmed with high confidence.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Markov Chain Model Analysis\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe CA-Markov model projection reflects expected land cover changes at three future time points: 2034, 2044, and 2054.\u003c/p\u003e \u003cp\u003eThe simulation predicts the continued expansion of forest ecosystems, particularly mixed forests on former peat extraction sites, following the trend identified in the LULC change analysis based on Landsat imagery from 1984 to 2024. Since 1990, forests have increasingly occupied drained wetland areas, and according to the model, this process will continue. By 2054, forests are projected to cover 13.7% of the total study area. However, the model also forecasts increased stability of meadow ecosystems on former peat extraction sites. Over the 27-year period from 1997, when meadow ecosystems on former peat extraction sides reached its maximum area, to 2024, these meadows have significantly declined, from 23\u0026ndash;15.2%, primarily due to forest encroachment. However, the CA-Markov model predicts that further replacement of meadows by forests and other land cover types will be limited to an additional 4.6%, indicating a remarkable slowing trend in meadow loss.\u003c/p\u003e \u003cp\u003eThe model also projects further expansion of residential areas. By 2034, built-up areas are expected to increase by 2.2% of the total study area compared to 2024. This trend is expected to continue in later time periods, reaching a maximum of 14.3% of the total area by 2054, representing a 6,3% increase from 2024.\u003c/p\u003e \u003cp\u003eThe decline in water bodies is forecasted to persist, following the previously observed trend. Starting at 3.5% of the total area in 2024, their extent is projected to shrink continuously, and by 2054, only 31.9% of their 2024 extent will remain-equivalent to 17.5% of their original area in 1984.\u003c/p\u003e \u003cp\u003eThe most stable land cover types, according to the model, are natural wetlands, coniferous forests, and croplands. Their areas are expected to experience only minor fluctuations within a\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5% range over time.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Main Successional Pathways\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe presence of a significant proportion of waterlogged and meadow grasslands on former peat extraction sites between 1984 and 1990 suggests a prolonged period of site abandonment or difficulties in organizing peat extraction on some drained lands. The first occurrence of forest stands on these sites only in 1993 further indicates the absence or extremely low intensity of afforestation efforts. Artificial reforestation is a common method for rehabilitating disturbed peatlands (Laudon and Maher Hasselquist, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, the long-term effectiveness of large-scale afforestation remains debatable, as without proper maintenance of the drainage network, the gas exchange and hydrological properties of afforested areas tend to revert to their natural state, leading to waterlogging and methane emissions (Inisheva et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In contrast, localized afforestation measures, such as tree belts along drainage channels (Inisheva et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and the creation of forest-meadow-wetland agro-landscapes, are considered more sustainable and ecologically safe approaches (Ulanov et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe sporadic and unsystematic appearance of croplands on former peat extraction sites reflects two key factors: the degradation of the drainage network, which led to waterlogging and secondary wetland formation, and the overall decline in agriculture and rural depopulation in Russia at the turn of the 20th-21st centuries, driven by socio-political changes (Sheludkov et al., \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOver the past 40 years, the dominant successional pathways in the study area have been \u0026rsquo;peat extraction \u0026rarr; waterlogged meadows and hayfields\u0026rsquo; and \u0026rsquo;peat extraction \u0026rarr; waterlogged meadows and hayfields \u0026rarr; mixed forests\u0026rsquo;. The first pathway was predominant for most of the study period, likely due to poor drainage conditions and soil properties. However, from 1997 to 2007, the transition from meadows to forests on former peat extraction sites accelerated significantly. After this period, forest expansion continued until 2024, but at a noticeably slower rate. The onset of rapid mixed forest growth on these sites coincided with the cessation of large-scale industrial peat extraction. This correlation is likely not coincidental, as the discontinuation of direct anthropogenic influences, such as heavy machinery operations, created favourable conditions for rapid forest encroachment on well-drained areas with suitable soil properties. This process aligns with natural succession patterns characteristic of the regional climatic conditions. The expansion of forests is a typical process in recovering peatlands (Habib and Connolly, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Sirin et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and the spread of woody and shrub vegetation in the study area has been previously noted (Pashnina and Simakova, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Simakova et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Dynamics of Water Surface Area\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eA particularly interesting aspect is the dynamics of lakes within the Tarmanskoe peatland. An analysis of water bodies area revealed a significant reduction since 2013, with a more than 40% decrease over the following decade due to progressive shallowing and overgrowth. All lakes in the study area exhibit floating-mat shoreline vegetation expansion (Solodovnikov, \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In the Tarmanskoe peatland, this process was already observed during peat deposit surveys in the 1950s (Giprotorfrazvedka, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1955\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDrainage reclamation also influenced the hydrological regime and water levels, causing some lakes-such as Lake Svetloe in the southern part of the wetland-to completely disappear, as confirmed by satellite imagery analysis. Observations at Lake Srednee Tarmanskoe, where monitoring by the Tyumen Fen Station began in 1960, indicate that active water level decline started in 1971, following the drainage of adjacent wetland areas. This trend continued for about a decade (Materials from Fen Station Observations, 1974, Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e). However, after a few years, the lake\u0026rsquo;s water level nearly returned to its original state and remained stable until the late 1980s, when peat extraction sites were abandoned. Similar or even more pronounced trends of water surface reduction following drainage activities have been observed in other human-modified wetland complexes, though they tend to slow down over time (Kurbatova et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHowever, in the case of the Tarmanskoe peatland, lake area reduction has intensified again in recent decades. Analysis of satellite imagery from 1984 to 2024 shows that the process of shallowing and overgrowth became particularly evident in the 2000s, more than a decade after industrial peat extraction ceased, despite the near-total degradation of the drainage system. Notably, water bodies in parts of the wetland that were never subjected to intensive human impact have also experienced overgrowth and shallowing, albeit to a lesser extent.\u003c/p\u003e \u003cp\u003eThus, the causes of lake area reduction appear to be complex, involving both natural and anthropogenic factors. Most likely, the natural aging of water bodies was compounded by drainage reclamation, which led to lower groundwater and wetland water levels, as well as climate change, manifested in rising mean annual temperatures (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e). The stabilization of the hydrological regime a few years after drainage began, along with a partial rebound in groundwater and wetland water levels following the abandonment of peat extraction sites and drainage system degradation, likely slowed the shallowing process. However, the continued rise in air temperature has reaccelerated this trend (Figs.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e13\u003c/span\u003e, \u003cspan refid=\"Fig14\" class=\"InternalRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.3. Landscape Mosaic and Fragmentation\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eSmall-scale, mosaic-like landscapes are primarily concentrated in areas previously subjected to peat extraction, indicating a lasting impact of human activity on natural ecosystems, even after anthropogenic pressure has ceased. As these areas undergo recovery, they exhibit a diversity of ecosystems, with some sections becoming forested, while others remain open landscapes with varying degrees of water saturation.\u003c/p\u003e \u003cp\u003eForested areas that have developed on former peat extraction sites display different stages of successional processes. Analysis of high-resolution Esri Imagery and UAV surveys revealed a heterogeneous structure of forest communities formed on drained lands. These forests often consist of a mosaic of tree stands with varying canopy density, age, and species composition, interspersed with shrub-dominated patches, as well as open grassland and wetland ecosystems. This structural heterogeneity affects their ecological state, biodiversity, and spatial organization, creating a range of microhabitats that, in turn, support overall biological diversity (Erds et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Mobaied et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Redon et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHowever, such mosaic landscapes have both positive and negative consequences. On the one hand, ecosystem diversity can enhance species richness and resilience to external disturbances, providing varied ecological niches for flora and fauna (Hitchman et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). On the other hand, high landscape fragmentation can disrupt migration routes, alter ecosystem processes, and create challenges for land management (Haddad et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Hanski, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSome former peat extraction sites have been converted for residential, industrial, and agricultural use. However, most drainage systems are no longer maintained, leading to frequent flooding. Drainage reduction has also contributed to rising groundwater levels, which, combined with new construction and road development, has further complicated land use. In addition to the direct risks of property damage, such flooding introduces various contaminants that primarily affect soil and water quality.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Spectral Information\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eNatural ecosystems unaffected by anthropogenic disturbance exhibit stable spectral channel and index values. In contrast, ecosystems that have emerged through secondary succession on former peat extraction sites show a gradual decline in reflectance in SWIR channels and a corresponding increase in NIR reflectance, influencing the dynamics of spectral indices.\u003c/p\u003e \u003cp\u003eIt is well established that SWIR bands correlate with water reflectance and moisture content (Burdun et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Koley and Jeganathan, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), while NIR reflectance is associated with chlorophyll content and vegetation biomass (Holben, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1986\u003c/span\u003e). NDVI and NDMI are linked to chlorophyll and water content in plants, with NDMI particularly sensitive to water stress, while NDWI reflects the amount of liquid water in plant canopies (Jin and Sader, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThus, increasing NDVI and NDMI values during secondary successionas former peat extraction sites become overgrown with forest and meadow vegetation-indicates rising vegetation biomass, while a declining NDWI reflects reduced water saturation and increased drainage. However, the recent trend of rising NDWI in meadow ecosystems suggests increasing water saturation, likely due to rising groundwater levels, which may negatively impact vegetation. This stress effect is reflected in declining NDVI and NDMI values.\u003c/p\u003e \u003cp\u003ePeatland recovery may be gradually shifting plant communities toward helophytic peatland vegetation (Haapalehto et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). The observed spectral index dynamics, combined with previous studies on the Tarmanskoe peatland (Pashnina and Simakova, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), may serve as indirect evidence of rising groundwater levels, resulting in secondary wetland formation and the development of meadow-wetland ecosystems. This raises the question of whether the decline in lake area since 2013 is linked to the increasing water saturation of abandoned peat extraction sites, a trend that has been detectable in spectral data since 2015\u0026ndash;2016.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e4.5. CA-Markov Modelling Results\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe results of the CA-Markov model predicting future changes in the land cover structure of the Tarmanskoe peatland provide important insights into potential landscape dynamics over the next 30 years. The data confirm existing trends and reveal new possible successional pathways, which is particularly relevant in the context of global challenges related to managing disturbed peatlands and mitigating the impacts of climate change.\u003c/p\u003e \u003cp\u003eThe forecast for the period up to 2054 indicates a continued expansion of forest ecosystems, particularly mixed and small-leaved forests on former peat extraction sites. This process aligns with observations from the past 40 years and confirms the hypothesis that forest succession dominates on drained peatlands. Forest ecosystems play a critical role in carbon sequestration and the stabilization of hydrological conditions, which is vital for reducing greenhouse gas emissions. However, the model also projects a slowing of forest expansion by 2044 and 2054, suggesting that the landscape may be reaching a certain level of stabilization.\u003c/p\u003e \u003cp\u003eWaterlogged meadow and hayfields on former peat extraction sites will continue to cover significant areas despite their gradual replacement by forest vegetation. This reflects the second key successional pathway, which involves the preservation of resilient meadow communities. Over the 30-year forecast period (2024\u0026ndash;2054), the model predicts a 3.7% reduction in meadow area, compared to a 7.8% decline during the 40-year observation period from 1984 to 2024, largely due to forest encroachment. This indicates increasing ecosystem stability. Meadows on former peat extraction sites remain ecologically significant, contributing to moisture retention, soil erosion prevention, and biodiversity preservation. However, without proper management, these areas could be at risk of secondary waterlogging or desiccation, depending on changes in drainage or climatic conditions. This would lead to the loss of ecosystem functions or increased fire hazards. The forecast highlights the need for a comprehensive management strategy for these territories.\u003c/p\u003e \u003cp\u003eOf particular concern is the projected increase in urbanized areas. The growth of settlements on former agricultural lands and peat extraction sites suggests intensifying anthropogenic pressure on ecosystems. This process could contribute to further landscape fragmentation and changes in the hydrological regime. Landscape fragmentation negatively affects wildlife migration routes, reduces ecosystem resilience, and complicates natural restoration processes. Furthermore, the expansion of urban areas is often accompanied by road and infrastructure development, which can alter drainage patterns, leading to increased local groundwater and wetland water levels and causing flooding.\u003c/p\u003e \u003cp\u003eThe projected reduction of water bodies to 17.5% of their 1984 extent by 2054 is an alarming indicator. Lakes and other water bodies provide numerous essential ecosystem services, including maintaining hydrological balance, serving as habitats for diverse flora and fauna, regulating the local climate, and supporting recreation and fisheries. Lake shallowing may result in the loss of these ecosystem functions. Moreover, reduced lake depth and encroachment by aquatic vegetation can cause unpredictable shifts in the carbon balance (Aben et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Bodmer et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Oliveira-Junior et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Many studies have highlighted increased methane emissions from shallow and overgrown aquatic ecosystems (Deemer and Holgerson, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Struik et al., \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOverall, the reduction in moisture levels across the study area increases the risk of forest and peatland fires, which can cause severe damage to biodiversity, soil resources, and carbon storage, leading to significant carbon emissions into the atmosphere (Jacobson et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Prat-Guitart et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The projected reduction in water bodies underscores the urgent need for additional conservation and restoration measures to prevent negative ecological consequences.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe analysis of land cover dynamics revealed a steady decline in the area of peat extraction sites and meadows, accompanied by the expansion of forest ecosystems, indicating the dominance of forest succession on drained lands. Forest ecosystems play a crucial role in carbon sequestration and maintaining hydrological balance, thereby reducing greenhouse gas emissions. However, their recovery may have contrasting effects on secondary waterloggingtemporarily lowering groundwater levels, while in the long term, promoting wetland formation in poorly drained areas.\u003c/p\u003e \u003cp\u003eThe reduction in lake area, identified through remote sensing data, requires particular attention. Over the past decade (2013\u0026ndash;2024), the water surface area of the Tarmanskoe peatland has decreased by more than 40%, indicating active shallowing and overgrowth. This decline has a complex negative impact on the hydrological balance and ecosystem functions of the region.\u003c/p\u003e \u003cp\u003eSpectral analysis revealed trends of increasing biomass accumulation and decreasing water saturation in meadows and forests that have developed on former peat extraction sites. This confirms the gradual recovery of vegetation communities and their transition to a more stable state.\u003c/p\u003e \u003cp\u003eThe CA-Markov model predicts the continued expansion of forest ecosystems, but at a slower pace, suggesting relative stabilization of meadow ecosystems. The further reduction of water bodies highlights the urgent need to preserve hydrological functions and prevent soil degradation. According to the model, by 2054, lake area may shrink to 17.5% of its 1984 extent, posing a range of long-term ecological risks, including an increased risk of wildfires.\u003c/p\u003e \u003cp\u003eThe observed landscape mosaic and high ecosystem fragmentation result from anthropogenic impact and uneven recovery of natural complexes. While this fosters habitat diversity and supports biodiversity, it also complicates hydrological processes, hinders the sustainable development of individual ecosystems, and heightens the area\u0026rsquo;s vulnerability to external stressors, including climate change and human activities.\u003c/p\u003e \u003cp\u003eThese findings underscore the need for active management of disturbed landscapes to prevent soil degradation and maintain ecosystem functions and biodiversity. The restoration of degraded peatlands requires a comprehensive approach, considering hydrological conditions, succession patterns, anthropogenic pressure, and fire risks. Such an approach will help minimize ecological risks, reduce wildfire hazards, prevent soil degradation and flooding risks, and preserve the ecosystem functions of the landscape.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Russian Science Foundation under Grant [number 23-77-10012]; the analysis of changes in the area of water bodies within the wetland ecosystem was carried out within the thematic framework of the State Assignment of the Institute of Forest Science of the Russian Academy of Sciences [number 123033000042-6].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eVladimir Ivanov was responsible for data curation, preparation of the original manuscript draft, and contributed significantly to the visualization of the results.\u003c/p\u003e\n\u003cp\u003eIvan Milyaev carried out the formal analysis and contributed to the development of the methodology, preparation of visualizations, and writing of the Methods section.\u003c/p\u003e\n\u003cp\u003eEvgeniya Soldatova contributed to the conceptualization of the study and provided overall supervision throughout the research and manuscript preparation.\u003c/p\u003e\n\u003cp\u003eAll authors participated in conducting the field investigation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe data are available directly from the cited data sources and, upon request, from the authors.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclarations Competing Interests\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Authors declare no conflicts of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAben RCH, Velthuis M, Kazanjian G, et al (2022) Temperature response of aquatic greenhouse gas emissions differs between dominant plant types. Water Research 226:119251. https://doi.org/10.1016/j.watres.2022.119251\u003c/li\u003e\n \u003cli\u003eAkhmeteva N, Mikhailova A, Krichevets G, et al (2019) Transformatsiya Antropoghenno Narushennykh Torfianyh Bolot V Novyy Tip Landshafta V Tsentralnykh Rayonakh Evropeyskoy Chasti Rossii [Transformation of Anthropogenically Disturbed Peat Bogs into a New Type of Landscape in the Central Areas of the European Part of Russia]. Trudy Instorfa [Proceedings of Instorf] (20(73)):3\u0026ndash;10 (in Russian)\u003c/li\u003e\n \u003cli\u003eAsare A, Thodsen H, Antwi M, et al (2021) Land use and land cover changes in lake Bosumtwi Watershed, Ghana (West Africa). Remote Sensing Applications: Society and Environment 23:100536. https://doi.org/10.1016/j.rsase.2021.100536\u003c/li\u003e\n \u003cli\u003eBaisheva E, Martynenko V, Shirokikh P, et al (2022) About distribution of drained peatlands in bashkir CIS-urals. \u0026Egrave;kobiteh https://doi.org/10.31163/2618-964x-2021-5-1-10-19\u003c/li\u003e\n \u003cli\u003eBodmer P, Vroom RJE, Stepina T, et al (2024) Methane dynamics in vegetated habitats in inland waters: quantification, regulation, and global significance. Frontiers in Water. https://doi.org/10.3389/frwa.2023.1332968\u003c/li\u003e\n \u003cli\u003eBondar EG (2022) The current state and prospects of the fuel and energy complex of St. Petersburg and the Leningrad region. The economy of the North-West: problems and prospects of development https://doi.org/10.52897/2411-4588-2022-2-71-77\u003c/li\u003e\n \u003cli\u003eVan den Bossche J, Fleischmann M, McBride J, et al (2024) GeoPandas. https://geopandas.org/en/stable/ Accessed 15 Oct 2024\u003c/li\u003e\n \u003cli\u003eBruisch K (2020) Nature Mistaken: Resource-Making, Emotions and the Transformation of Peatlands in the Russian Empire and the Soviet Union. Environment and History 26:359\u0026ndash;382. https://doi.org/10.3197/096734018X15254461646567\u003c/li\u003e\n \u003cli\u003eBurdun I, Bechtold M, Sagris V, et al (2020) Satellite determination of peatland water Table temporal dynamics by localizing representative pixels of a SWIR-based moisture index. Remote Sensing 12(18):2936. https://doi.org/10.3390/rs12182936\u003c/li\u003e\n \u003cli\u003eChen Y, Hu X, Zhang Y, et al (2022) Characterizing the Long-Term Landscape Dynamics of a Typical Cloudy Mountainous Area in Northwest Yunnan, China. Sustainability https://doi.org/10.3390/su142013488\u003c/li\u003e\n \u003cli\u003eCongalton RG, Green K (2019) Assessing the accuracy of remotely sensed data: principles and practices, third edition, 3rd edn. CRC Press, Boca Raton. https://doi.org/10.1201/9780429052729\u003c/li\u003e\n \u003cli\u003eDeemer BR, Holgerson MA (2021) Drivers of Methane Flux Differ Between Lakes and Reservoirs, Complicating Global Upscaling Efforts. Journal of Geophysical Research: Biogeosciences 126(4):e2019JG005600. https://doi.org/10.1029/2019JG005600\u003c/li\u003e\n \u003cli\u003eEdvardsson J, Helama S, Rundgren M, et al (2022) The Integrated Use of Dendrochronological Data and Paleoecological Records From Northwest European Peatlands and Lakes for Understanding Long-Term Ecological and Climatic ChangesA Review. Frontiers in Ecology and Evolution. https://doi.org/10.3389/fevo.2022.781882\u003c/li\u003e\n \u003cli\u003eErds L, Kr\u0026ouml;el-Dulay G, B\u0026aacute;tori Z, et al (2018) Habitat heterogeneity as a key to high conservation value in forest-grassland mosaics. Biological Conservation. https://doi.org/10.1016/J.BIOCON.2018.07.029\u003c/li\u003e\n \u003cli\u003eEva EA, Marzen LJ, Lamba J, et al (2024) Projection of land use and land cover changes based on land change modeler and integrating both land use land cover and climate change on the hydrological response of Big Creek Lake Watershed, South Alabama. Journal of Environmental Management 370:122923. https://doi.org/10.1016/j.jenvman.2024.122923\u003c/li\u003e\n \u003cli\u003eFilippova N, Zvyagina E, Rudykina E, et al (2023) The diversity of macromycetes in peatlands: nine years of plot-based monitoring and barcoding in the raised bog \u0026quot;Mukhrino\u0026quot;, West Siberia. Biodiversity Data Journal. https://doi.org/10.3897/BDJ.11.e105111\u003c/li\u003e\n \u003cli\u003eFu B, Zhang L, Xu Z, et al (2015) Ecosystem services in changing land use. Journal of Soils and Sediments 15:833\u0026ndash;843.https://doi.org/10.1007/s11368-015-1082-x\u003c/li\u003e\n \u003cli\u003eGao Bc (1996) NDWIA normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment 58(3):257\u0026ndash;266. https://doi.org/10.1016/S0034-4257(96)00067-3\u003c/li\u003e\n \u003cli\u003eGhimire B, Rogan J, Galiano VR, et al (2012) An evaluation of bagging, boosting, and random forests for land-cover classification in Cape Cod, Massachusetts, USA. GIScience \u0026amp; Remote Sensing 49(5):623\u0026ndash;643. https://doi.org/10.2747/1548-1603.49.5.623\u003c/li\u003e\n \u003cli\u003eGiprotorfrazvedka (1955) Detailed Reconnaissance Materials of the Tarmaanskoye Peat Deposit. Tech. Rep. 4, Gorkiy\u003c/li\u003e\n \u003cli\u003eGiprotorfrazvedka (1955) Detailed Reconnaissance Materials of the Tarmaanskoye Peat Deposit. Tech. Rep. 15, Gorkiy\u003c/li\u003e\n \u003cli\u003eGomes E, In\u0026aacute;cio M, Bogdzevi K, et al (2021) Future land-use changes and its impacts on terrestrial ecosystem services: A review. The Science of the total environment 781. https://doi.org/10.1016/j.scitotenv.2021.146716\u003c/li\u003e\n \u003cli\u003eGomes L, Bianchi F, Cardoso IM, et al (2020) Land use change drives the spatio-temporal variation of ecosystem services and their interactions along an altitudinal gradient in Brazil. Landscape Ecology 35:1571\u0026ndash;1586. https://doi.org/10.1007/s10980-020-01037-1\u003c/li\u003e\n \u003cli\u003eGorelick N, Hancher M, Dixon M, et al (2017) Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment 202:18\u0026ndash;27. https://doi.org/10.1016/j.rse.2017.06.031\u003c/li\u003e\n \u003cli\u003eGuan D, Gao W, Watari K, et al (2008) Land use change of Kitakyushu based on landscape ecology and Markov model. Journal of Geographical Sciences 18(4):455\u0026ndash;468. https://doi.org/10.1007/s11442-008-0455-0\u003c/li\u003e\n \u003cli\u003eGxokwe S, Dube T, Mazvimavi D (2022) Leveraging Google Earth Engine platform to characterize and map small seasonal wetlands in the semiarid environments of South Africa. Science of The Total Environment 803:150139. https://doi.org/10.1016/j.scitotenv.2021.150139\u003c/li\u003e\n \u003cli\u003eHaapalehto TO, Vasander H, Jauhiainen S, et al (2011) The effects of peatland restoration on watertable depth, elemental concentrations, and vegetation: 10 years of changes. Restoration Ecology 19(5):587\u0026ndash;598. https://doi.org/https://doi.org/10.1111/j.1526-100X.2010.00704.x\u003c/li\u003e\n \u003cli\u003eHabib W, Connolly J (2023) A national-scale assessment of land use change in peatlands between 1989 and 2020 using Landsat data and Google Earth Enginea case study of Ireland. Regional Environmental Change 23(4):124. https://doi.org/10.1007/s10113-023-02116-0\u003c/li\u003e\n \u003cli\u003eHaddad N, Brudvig L, Clobert J, et al (2015) Habitat fragmentation and its lasting impact on Earths ecosystems. Science Advances 1. https://doi.org/10.1126/sciadv.1500052\u003c/li\u003e\n \u003cli\u003eHanski I (2015) Habitat fragmentation and species richness. Journal of Biogeography 42. https://doi.org/10.1111/jbi.12478\u003c/li\u003e\n \u003cli\u003eHarris CR, Millman KJ, van der Walt SJ, et al (2020) Array programming with NumPy. Nature 585(7825):357\u0026ndash;362. https://doi.org/10.1038/s41586020-2649-2\u003c/li\u003e\n \u003cli\u003eHitchman SM, Mather ME, Smith JM, et al (2018) Identifying keystone habitats with a mosaic approach can improve biodiversity conservation in disturbed ecosystems. Global Change Biology https://doi.org/10.1111/gcb.13846\u003c/li\u003e\n \u003cli\u003eHolben BN (1986) Characteristics of maximum-value composite images from temporal AVHRR data. International Journal of Remote Sensing (7):1417\u0026ndash; 1434. https://doi.org/10.1080/01431168608948945\u003c/li\u003e\n \u003cli\u003eHugelius G, Loisel J, Chadburn S, et al (2020) Large stocks of peatland carbon and nitrogen are vulnerable to permafrost thaw. Proceedings of the National Academy of Sciences of the United States of America 117:20438\u0026ndash; 20446. https://doi.org/10.1073/pnas.1916387117\u003c/li\u003e\n \u003cli\u003eHumpen\u0026ouml;der F, Karstens K, Lotze-Campen H, et al (2020) Peatland protection and restoration are key for climate change mitigation. Environmental Research Letters 15. https://doi.org/10.1088/1748-9326/abae2a\u003c/li\u003e\n \u003cli\u003eIhlen V, Zanter K (2019) Landsat 8 (L8) Data Users Handbook, 5th edn. U.S. Geological Survey, Sioux Falls\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eInisheva L, Shaydak L, Babikov B (2021) Hydrological and gas regime of swamps in the conditions of forest reclamation. Environmental Science pp 39\u0026ndash;44. https://doi.org/10.32962/0235-2524-2020-6-39-44\u003c/li\u003e\n \u003cli\u003eInisheva L, Sergeeva M, Golovchenko A, et al (2023) Carbon Dioxide and Methane Distribution in Peat Deposits of an Oligotrophic Forest Swamp in Western Siberia and Their Emission. Lesovedenie [Forest Science] https://doi.org/10.31857/s0024114823010060\u003c/li\u003e\n \u003cli\u003eJacobson TWP, Seager R, Williams AP, et al (2024) An unexpected decline in spring atmospheric humidity in the interior Southwestern United States and implications for forest fires. Journal of Hydrometeorology https://doi.org/10.1175/jhm-d-23-0121.1\u003c/li\u003e\n \u003cli\u003eJin S, Sader SA (2005) Comparison of time series tasseled cap wetness and the normalized difference moisture index in detecting forest disturbances. Remote Sensing of Environment 94(3):364\u0026ndash;372. https://doi.org/10.1016/j.rse.2004.10.012\u003c/li\u003e\n \u003cli\u003eJin Y, Liu X, Chen Y, et al (2018) Land-cover mapping using Random Forest classification and incorporating NDVI time-series and texture: a case study of central Shandong. International Journal of Remote Sensing 39(23):8703\u0026ndash;8723. https://doi.org/10.1080/01431161.2018.1490976\u003c/li\u003e\n \u003cli\u003eKarpov V (2010) Neft\u0026rsquo; i gaz v promyshlennoy politike SSSR (Rossii) [Oil and gas in the industrial policy of the USSR (Russia)]. Vestnik Nizhevartovskogo gosudarstvennogo universiteta [Bulletin of Nizhnevartovsk State University] (4):75\u0026ndash;88. (In Russian)\u003c/li\u003e\n \u003cli\u003eKazakov K (2025) Letopis\u0026apos; Pogody i Klimata Tyumeni [Chronicle of Weather and Climate of Tyumen]. http://www.pogodaiklimat.ru/history/28367.html. (In Russian) Accessed 15 Oct 2024\u003c/li\u003e\n \u003cli\u003eKendall M (1948) Rank correlation methods. Rank correlation methods, Griffin, Oxford, England\u003c/li\u003e\n \u003cli\u003eKhozyainova NV, Cheshuina IA, Glazunov VA (1999) Flora Tarmanskogo leso-vodo-bolotnogo kompleksa [Flora of the Tarman forest-water-wetland complex]. Vestnik Tyumenskogo gosudarstvennogo universiteta [Bulletin of Tyumen State University] (3):92\u0026ndash;98. (In Russian)\u003c/li\u003e\n \u003cli\u003eKoley S, Jeganathan C (2020) Estimation and evaluation of high spatial resolution surface soil moisture using multi-sensor multi-resolution approach. Geoderma 378:114618. https://doi.org/10.1016/j.geoderma.2020.114618\u003c/li\u003e\n \u003cli\u003eKrasnov O, Zhabin V, Matukhina V, et al (2005) Syryevaya baza torfa Sredney Sibiri i osnovnye napravleniya ego ratsional\u0026rsquo;nogo ispol\u0026rsquo;zovaniya [Raw material base of peat in Central Siberia and the main directions of its rational use]. Interekspo Geo-Sibir\u0026rsquo; [Interexpo Geo-Siberia] 3(1):21\u0026ndash;26. (In Russian)\u003c/li\u003e\n \u003cli\u003eKumar M, Mahato LL, Suryavanshi S, et al (2024) Future prediction of water balance using the SWAT and CA-Markov modelăusing INMCM5 climate projections: a case study of the Silwani watershed (Jharkhand), India. Environmental Science and Pollution Research 31(41):54311\u0026ndash;54324. https://doi.org/10.1007/s11356-023-27547-4\u003c/li\u003e\n \u003cli\u003eKurbatova I, Vereshchaka T, Ivanova A (2021) Space monitoring bog landscape transformation under anthropogenic impact conditions. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa [Corrent problems in remote sensing of the Earth from space] 18(4):216\u0026ndash; 227. https://doi.org/10.21046/2070-7401-2021-18-4-216-227, number: 4\u003c/li\u003e\n \u003cli\u003eKurganova I, Gerenyu VL, Kuzyakov Y (2015) Large-scale carbon sequestration in post-agrogenic ecosystems in Russia and Kazakhstan. Catena 133:461\u0026ndash;466. https://doi.org/10.1016/J.CATENA.2015.06.002\u003c/li\u003e\n \u003cli\u003eLapytskaya M (2020) Transformations in the Agrarian Sector as a Method of Problem Resolution of the Peasant Land Shortages in Russia (Based on the Research on Agrarian Reforms in Russia in Second Half of XIX - Beginning of the XX Centuries) 20:601\u0026ndash;624. https://doi.org/10.17150/23082488.2019.20(4).601-624\u003c/li\u003e\n \u003cli\u003eLaudon H, Maher Hasselquist E (2023) Applying continuous-cover forestry on drained boreal peatlands; water regulation, biodiversity, climate benefits and remaining uncertainties. Trees, Forests and People 11:100363. https://doi.org/10.1016/j.tfp.2022.100363\u003c/li\u003e\n \u003cli\u003eLeifeld J, W\u0026uuml;st-Galley C, Page S (2019) Intact and managed peatland soils as a source and sink of GHGs from 1850 to 2100. Nature Climate Change pp 1\u0026ndash;3. https://doi.org/10.1038/s41558-019-0615-5\u003c/li\u003e\n \u003cli\u003eLeng LY, Ahmed O, Jalloh M (2018) Brief review on climate change and tropical peatlands. Geoscience Frontiers. https://doi.org/10.1016/J.GSF.2017.12.018\u003c/li\u003e\n \u003cli\u003eLiu J, Wilson M, Hu G, et al (2018) How does habitat fragmentation affect the biodiversity and ecosystem functioning relationship? Landscape Ecology 33:341\u0026ndash;352. https://doi.org/10.1007/s10980-018-0620-5\u003c/li\u003e\n \u003cli\u003eLoisel J, Gallego-Sala A, Amesbury M, et al (2020) Expert assessment of future vulnerability of the global peatland carbon sink. Nature Climate Change 11:70\u0026ndash;77. https://doi.org/10.1038/s41558-020-00944-0\u003c/li\u003e\n \u003cli\u003eLuan C, Liu R (2022) A comparative study of various land use and land cover change models to predict ecosystem service value. International Journal of Environmental Research and Public Health 19(24):16484. https://doi.org/10.3390/ijerph192416484\u003c/li\u003e\n \u003cli\u003eLudwig C, Walli A, Schleicher C, et al (2019) A highly automated algorithm for wetland detection using multi-temporal optical satellite data. Remote Sensing of Environment https://doi.org/10.1016/J.RSE.2019.01.017\u003c/li\u003e\n \u003cli\u003eLvov Y (1995) Torf i formy ego ispol\u0026rsquo;zovaniya v Sibiri [Peat and its forms of use in Siberia] pp 31\u0026ndash;39. Publisher: Natsional\u0026rsquo;nyy issledovatel\u0026rsquo;skiy Tomskiy gosudarstvennyy universitet [National Research Tomsk State University] (In Russian)\u003c/li\u003e\n \u003cli\u003eLysenko Y (2021) Ethno-Economics of the Kazakhs of the Steppe Region in the Modernization Plans of the Russian Empire (second half of the XIX beginning of the XX century). Bylye Gody. https://doi.org/10.13187/bg.2021.2.840\u003c/li\u003e\n \u003cli\u003eMa XY, Xu H, Cao ZY, et al (2022) Will climate change cause the global peatland to expand or contract? Evidence from the habitat shift pattern of Sphagnum mosses. Global Change Biology 28(21):6419\u0026ndash;6432. https://doi.org/10.1111/gcb.16354\u003c/li\u003e\n \u003cli\u003eMalek Ž, Douw B, Vliet Jv, et al (2019) Local land-use decisionmaking in a global context. Environmental Research Letters 14. https://doi.org/10.1088/1748-9326/ab309e\u003c/li\u003e\n \u003cli\u003eMann HB (1945) Nonparametric tests against trend. Econometrica 13(3):245\u0026ndash;259. https://doi.org/10.2307/1907187\u003c/li\u003e\n \u003cli\u003eMaslov M, Maslova O (2020) Temperate peatlands use-management effects on seasonal patterns of soil microbial activity and nitrogen availability. Catena 190. https://doi.org/10.1016/j.catena.2020.104548\u003c/li\u003e\n \u003cli\u003eMcDaniel M, Saha D, Dumont M, et al (2019) The Effect of Land-Use Change on Soil CH4 and N2O Fluxes: A Global Meta-Analysis. Ecosystems 22:1424\u0026ndash;1443. https://doi.org/10.1007/s10021-019-00347-z\u003c/li\u003e\n \u003cli\u003eMcLaughlin DL, Cohen MJ (2013) Realizing ecosystem services: wetland hydrologic function along a gradient of ecosystem condition. Ecological Applications 23(7):1619\u0026ndash;1631. https://doi.org/10.1890/12-1489.1\u003c/li\u003e\n \u003cli\u003eMickler R (2021) Carbon emissions from a temperate coastal peatland wildfire: contributions from natural plant communities and organic soils. Carbon Balance and Management 16. https://doi.org/10.1186/s13021-02100189-0\u003c/li\u003e\n \u003cli\u003eMillard K, Richardson M (2015) On the importance of training data sample selection in random forest image classification: a case study in peatland ecosystem mapping. Remote Sensing 7(7):8489\u0026ndash;8515. https://doi.org/10.3390/rs70708489\u003c/li\u003e\n \u003cli\u003eMinaeva T, Sirin A (2009) A Quick Scan of Peatlands in Central and Eastern Europe. Wageningen, The Netherlands.\u003c/li\u003e\n \u003cli\u003eMobaied S, Geoffroy J, Machon N (2016) The Importance of Spatiotemporal Heterogeneity for Biodiversity in Forest Heathland Mosaics and Implications for Heathland Conservation. Journal of Environmental Protection 07:1317\u0026ndash;1332. https://doi.org/10.4236/JEP.2016.710115\u003c/li\u003e\n \u003cli\u003eMonier E, Kicklighter D, Sokolov A, et al (2017) A review of and perspectives on global change modeling for Northern Eurasia. Environmental Research Letters 12. https://doi.org/10.1088/1748-9326/aa7aae\u003c/li\u003e\n \u003cli\u003eMoskalenko N, Bulko N, Tolkacheva N, et al (2020) K voprosu o sostoyanii meliorirovannykh zemel\u0026rsquo;, nakhodyashchikhsya v sostave lesnogo fonda [Regarding the condition of improved lands that are part of the forest fund]. Vestnik Grodnyenskogo Gosudarstvennogo Universiteta Imeni Yanki Kupaly Seriya 5 Ekonomika Sotsiologiya Biologiya [Bulletin of the Grodno State University named after Yanka Kupala Series 5 Economics Sociology Biology] 10(1):125\u0026ndash;132. (In Russian)\u003c/li\u003e\n \u003cli\u003eNekrich A, Lyuri D (2019) Izmeneniya dinamiki agrarnykh ugodiy Rossii v 19902014 gg. [Changes in the dynamics of agrarian lands in Russia in 19902014]. Izvestiya Rossiyskoy Akademii Nauk Seriya geograficheskaya [Proceedings of the Russian Academy of Sciences Geographical Series] 0(3):64\u0026ndash;77. (In Russian)\u003c/li\u003e\n \u003cli\u003eNguyen H, H\u0026ouml;lzel N, V\u0026ouml;lker A, et al (2018) Patterns and Determinants of Post-Soviet Cropland Abandonment in the Western Siberian Grain Belt. Remote Sens 10. https://doi.org/10.3390/rs10121973\u003c/li\u003e\n \u003cli\u003eOgorodnov E (1971) Atlas Tyumenskoi oblasti [Atlas of the Tyumen Region], vol 1. Nauka [Science], Moscow. (In Russian)\u003c/li\u003e\n \u003cli\u003eOliveira-Junior ES, Tang Y, van den Berg SJP, et al (2018) The impact of water hyacinth (\u003cem\u003eEichhornia crassipes\u003c/em\u003e) on greenhouse gas emission and nutrient mobilization depends on rooting and plant coverage. Aquatic Botany 145:1\u0026ndash;9. https://doi.org/10.1016/j.aquabot.2017.11.005\u003c/li\u003e\n \u003cli\u003ePashnina E, Simakova T (2017) Analiz Ispol\u0026rsquo;zovaniya Melioriruyemykh Zemel\u0026rsquo; Tarmanskogo Bolotnogo Massiva Tyumenskoy Oblasti [Analysis of the use of improved lands of the Tarman Wetland Complex of the Tyumen Region]. Gosudarstvennyy agrarnyy universitet Severnogo Zauralsya [State Agrarian University of Northern Trans-Urals], pp 116\u0026ndash;119 (In Russian)\u003c/li\u003e\n \u003cli\u003ePhan TN, Kuch V, Lehnert LW (2020) Land cover classification using Google Earth Engine and random forest classifierthe role of image composition. Remote Sensing 12(15):2411. https://doi.org/10.3390/rs12152411\u003c/li\u003e\n \u003cli\u003ePrat-Guitart N, Belcher C, Thompson D, et al (2017) Finescale distribution of moisture in the surface of a degraded blanket fen and its effects on the potential spread of smouldering fire. Ecohydrology 10. https://doi.org/10.1002/eco.1898\u003c/li\u003e\n \u003cli\u003eProvost GL, Badenhausser I, Bagousse-Pinguet YL, et al (2020) Landuse history impacts functional diversity across multiple trophic groups. Proceedings of the National Academy of Sciences 117:1573\u0026ndash;1579. https://doi.org/10.1073/pnas.1910023117\u003c/li\u003e\n \u003cli\u003eRada NE, Liefert W, Liefert O (2020) Evaluating Agricultural Productivity and Policy in Russia. Journal of Agricultural Economics https://doi.org/10.1111/1477-9552.12338\u003c/li\u003e\n \u003cli\u003eRappaport DI, Tambosi L, Metzger J (2015) A landscape triage approach: combining spatial and temporal dynamics to prioritize restoration and conservation. Journal of Applied Ecology 52:590\u0026ndash;601. https://doi.org/10.1111/1365-2664.12405\u003c/li\u003e\n \u003cli\u003eRedon M, Luque S, Gosselin F, et al (2014) Is generalisation of uneven-aged management in mountain forests the key to improve biodiversity conservation within forest landscape mosaics? Annals of Forest Science 71(7):751\u0026ndash; 760. https://doi.org/10.1007/s13595-014-0371-7\u003c/li\u003e\n \u003cli\u003eRitson JP, Alderson DM, Robinson CH, et al (2021) Towards a microbial process-based understanding of the resilience of peatland ecosystem service provisioning A research agenda. Science of The Total Environment 759:143467. https://doi.org/10.1016/j.scitotenv.2020.143467\u003c/li\u003e\n \u003cli\u003eRouse JW, Haas RH, Schell JA, et al (1974) Monitoring vegetation systems in the Great Plains with ERTS 351(1):309\u0026ndash;317.\u003c/li\u003e\n \u003cli\u003eSali M, Piaser E, Boschetti M, et al (2021) A burned area mapping algorithm for Sentinel-2 data based on approximate reasoning and region growing. Remote Sensing 13(11):2214. https://doi.org/10.3390/rs13112214, number: 11 Publisher: Multidisciplinary Digital Publishing Institute\u003c/li\u003e\n \u003cli\u003eSelmy SAH, Kucher DE, Mozgeris G, et al (2023) Detecting, analyzing, and predicting land use/land cover (LULC) changes in arid regions using Landsat images, CA-Markov Hybrid Model, and GIS techniques. Remote Sensing 15(23):5522. https://doi.org/10.3390/rs15235522\u003c/li\u003e\n \u003cli\u003eShafizadeh-Moghadam H, Khazaei M, Alavipanah SK, et al (2021) Google Earth Engine for large-scale land use and land cover mapping: an object-based classification approach using spectral, textural and topographical factors. GIScience \u0026amp; Remote Sensing 58(6):914\u0026ndash; 928. https://doi.org/10.1080/15481603.2021.1947623\u003c/li\u003e\n \u003cli\u003eShcherbakova IK (2021) Agrarian reforms of Russia at the turn of the XIXXX centuries in the coverage of economists of the XX century. Vestnik Universiteta https://doi.org/10.26425/1816-4277-2021-5-141-144\u003c/li\u003e\n \u003cli\u003eSheludkov A, Kamp J, M\u0026uuml;ller D (2020) Decreasing labor intensity in agriculture and the accessibility of major cities shape the rural population decline in postsocialist Russia. Eurasian Geography and Economics 62:481\u0026ndash;506. https://doi.org/10.1080/15387216.2020.1822751\u003c/li\u003e\n \u003cli\u003eSimakova T, Simakov A, Ivanova A (2023) Monitoring melioryruemykh zemel\u0026rsquo; s ispol\u0026rsquo;zovaniem landshaftno-ekologicheskogo podkhoda [Monitoring of improved lands using a landscape-ecological approach]. Vestnik Voronezhskogo gosudarstvennogo agrarnogo universiteta [Bulletin of the Voronezh State Agrarian University] 16(3):78. (In Russian)\u003c/li\u003e\n \u003cli\u003eSingh S, Gupta R, Kumari M, et al (2015) Nontarget effects of chemical pesticides and biological pesticide on rhizospheric microbial community structure and function in Vigna radiata. Environmental Science and Pollution Research 22(15):11290\u0026ndash;11300. https://doi.org/10.1007/s11356-015-4341-x\u003c/li\u003e\n \u003cli\u003eSirin A, Medvedeva M, Korotkov V, et al (2021) Addressing Peatland Rewetting in Russian Federation Climate Reporting. Land 10(11):1200. https://doi.org/10.3390/land10111200\u003c/li\u003e\n \u003cli\u003eSirin AA, Medvedeva MA, Makarov DA, et al (2020) Multispectral satellite based monitoring of land cover change and associated fire reduction after large-scale peatland rewetting following the 2010 peat fires in Moscow Region (Russia). Ecological Engineering 158:106044. https://doi.org/10.1016/j.ecoleng.2020.106044\u003c/li\u003e\n \u003cli\u003eSolodovnikov A (2021) Rol\u0026rsquo; zakaznika federal\u0026rsquo;nogo znacheniya \u0026quot;Tyumensky\u0026quot; v sokhranenii vidovogo raznoobraziya flory i fauny Nizhnetavdinskogo rayona Tyumenskoy oblasti [Role of the federal significance reserve \u0026quot;Tyumensky\u0026quot; in the preservation of species diversity of flora and fauna of the Nizhnetavdinsky district of the Tyumen region]. Zametki Uchenogo [Notes of the Scientist] (3-1):378\u0026ndash;385. (In Russian)\u003c/li\u003e\n \u003cli\u003eStruik Q, Oliveira Junior ES, Veraart AJ, et al (2022) Methane emissions through water hyacinth are controlled by plant traits and environmental conditions. Aquatic Botany 183:103574. https://doi.org/10.1016/j.aquabot.2022.103574\u003c/li\u003e\n \u003cli\u003eTaillardat P, Thompson BS, Garneau M, et al (2020) Climate change mitigation potential of wetlands and the cost-effectiveness of their restoration. Interface Focus 10. https://doi.org/10.1098/rsfs.2019.0129\u003c/li\u003e\n \u003cli\u003eTassi A, Gigante D, Modica G, et al (2021) Pixel- vs. object-based Landsat 8 data classification in Google Earth Engine using random forest: the case study of Maiella National Park. Remote Sensing 13(12):2299. https://doi.org/10.3390/rs13122299, URL https://www.mdpi.com/20724292/13/12/2299\u003c/li\u003e\n \u003cli\u003eTeluguntla P, Thenkabail PS, Oliphant A, et al (2018) A 30-m Landsatderived cropland extent product of Australia and China using random forest machine learning algorithm on Google Earth Engine cloud computing platform. ISPRS Journal of Photogrammetry and Remote Sensing 144:325\u0026ndash;340. https://doi.org/10.1016/j.isprsjprs.2018.07.017\u003c/li\u003e\n \u003cli\u003eTuretsky M, Donahue W, Benscoter B (2011) Experimental drying intensifies burning and carbon losses in a northern peatland. Nature communications 2. https://doi.org/10.1038/ncomms1523\u003c/li\u003e\n \u003cli\u003eUlanov A, Smirnova AV, Ulanov N (2023) Forest Stands Formation on Exhausted Peat Bog in the North-East of the European Part of Russia. Lesovedenie [Forest Science] https://doi.org/10.31857/s0024114823040137\u003c/li\u003e\n \u003cli\u003ede Waard F, Connolly J, Barthelmes A, et al (2024) Remote sensing of peatland degradation in temperate and boreal climate zones A review of the potentials, gaps, and challenges. Ecological Indicators 166:112437. https://doi.org/10.1016/j.ecolind.2024.112437\u003c/li\u003e\n \u003cli\u003eYang Q, Li X, Shi X (2008) Cellular automata for simulating land use changes based on support vector machines. Computers \u0026amp; Geosciences 34(6):592\u0026ndash; 602. https://doi.org/10.1016/j.cageo.2007.08.003\u003c/li\u003e\n \u003cli\u003eZhao B, Zhuang Q (2023) Peatlands and their carbon dynamics in northern high latitudes from 1990 to 2300: a process-based biogeochemistry model analysis. Biogeosciences https://doi.org/10.5194/bg-20-251-2023\u003c/li\u003e\n \u003cli\u003e\u0026Aring;hl\u0026eacute;n I, Thorslund J, Hamb\u0026auml;ck P, et al (2022) Wetland position in the landscape: Impact on water storage and flood buffering. Ecohydrology 15(7):e2458. https://doi.org/10.1002/eco.2458\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"68aa9d96-1b70-4ed7-8577-b73f57d6b4cc","identifier":"10.13039/501100006769","name":"Russian Science Foundation","awardNumber":"23-77-10012","order_by":0}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Peatland, Land-Cover Change, Remote Sensing, CA-Markov Model, Succession, Peat Extraction","lastPublishedDoi":"10.21203/rs.3.rs-7012873/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7012873/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePeatlands are key contributors to carbon storage and hydrological regulation but their role and ecosystem functions and services have been substantially altered by anthropogenic interference, primarily through drainage and peat extraction. This study focuses on the Tarmanskoe peatland in Western Siberia, where large areas were drained for peat extraction and agricultural use from the 1960s to 1970s. Using Landsat satellite imagery from 1984 to 2024 - complemented by high-resolution Unmanned Aerial Vehicle (UAV) data - we applied object-based classification (Random Forest) to assess historical land-cover changes. We then employed a hybrid CA-Markov (Cellular Automata-Marcov) model to project future landscape transformations over the next three decades (2034\u0026ndash;2054). Results indicate that formerly drained peatlands followed two main successional pathways: an initial phase of meadow formation with varying levels of waterlogging, followed by a gradual expansion of mixed forests. By 2024, about half of the drained peatland areas transitioned from meadows to forest cover, suggesting a dominant trend toward forest succession. Simultaneously, lakes in the region underwent significant water losses - nearly a 50% reduction in total area since 2013 - driven by natural aging processes, drainage-induced lowering of water levels, and rising mean annual temperatures. The CA-Markov projections reveal a continued, albeit slower, increase in forested areas and a further reduction in water bodies, reaching only 17.4% of their 1984 extent by 2054. These findings underscore the lasting ecological impacts of drainage and peat extraction, as evidenced by spatially heterogeneous successional processes and widespread fragmentation of ecosystems. They also highlight emerging risks, including further water-level declines, increased fire hazard, and ongoing landscape fragmentation. From a conservation perspective, proactive management and the restoration of hydrological functions in disturbed peatlands may help mitigate long-term ecological and climate-related vulnerabilities.\u003c/p\u003e","manuscriptTitle":"Successional pathways after peatland draining: remote sensing and predictive modelling of landscape dynamics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-02 10:07:15","doi":"10.21203/rs.3.rs-7012873/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c03abddc-5ffc-43a2-9add-41bd448eec0d","owner":[],"postedDate":"July 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-07-02T10:07:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-02 10:07:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7012873","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7012873","identity":"rs-7012873","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}