GIS-Based Soil Erosion Susceptibility Mapping in Serbia Using the EPM Model and Satellite Remote Sensing: A National-Scale Prediction

GIS-Based Soil Erosion Susceptibility Mapping in Serbia Using the EPM Model and Satellite Remote Sensing: A National-Scale Prediction | 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 GIS-Based Soil Erosion Susceptibility Mapping in Serbia Using the EPM Model and Satellite Remote Sensing: A National-Scale Prediction Uroš Durlević, Tanja Srejić, Sanja Manojlović, Marko V. Milošević, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9569900/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 Soil erosion represents a complex geomorphological and geological process that negatively affects the environment, the quality of natural resources, and the safety of the population. In this study, a mechanical soil erosion map for the territory of Serbia was produced using the Erosion Potential Model (EPM), combined with remote sensing data (Sentinel-2) and Geographic Information Systems (GIS). The analysis used contemporary geospatial data on lithology, land use, and terrain slope, with a 30 m spatial resolution. The average erosion intensity at the national level is 0.239, corresponding to the weak erosion class. More than half of the territory (53.8%) is affected by very weak erosion, while 24.2% is characterized by weak erosion intensity. Additionally, 12.7% of Serbia falls within the medium erosion intensity class. The results further indicate that 7.6% of the territory is affected by intensive erosion, while 0.7% is exposed to excessive erosion. Approximately 1% of the territory is exposed to sediment accumulation. Multivariate analysis of geographical conditions showed that the highest values of the erosion coefficient (Z) were determined by land use (r = 0.826). In contrast, the lowest values were associated with terrain slope (r = −0.805). In addition to the national-scale assessment, spatial differentiation of the results was performed at the local (municipal) level. The most susceptible municipalities were identified and analyzed, as well as those characterized by specific natural and anthropogenic conditions. Municipalities were subsequently grouped into six clusters based on six indicators using Agglomerative Hierarchical Clustering (AHC), according to their susceptibility to mechanical erosion. The results were validated using the ROC-AUC metric, which showed perfect predictive power (100%). This study significantly contributes to decision-making at both national and local levels by providing a scientific basis for developing strategies for sustainable forest management and soil conservation, highlighting the important role of forests in reducing soil erosion and maintaining ecosystem stability. Geomorphology Geographic Information Systems Geosciences Sentinel-2 Land use Agricultural land Erosion potential model (EPM) Agglomerative Hierarchical Clustering (AHC) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 1.Introduction Soil is considered a non-renewable resource, as soil erosion rates are significantly higher than the rate of its natural formation, and its loss and degradation cannot be compensated within the timescale of human life. Soil erosion is a natural geomorphological process that occurs continuously on Earth’s surface (Lovrić et al., 2018). However, in terms of the area affected, it has been recognized as a global problem (Durlević et al., 2019 ). Soil erosion results from multiple interacting natural and anthropogenic processes. Its intensity is largely controlled by the combined effects of relief characteristics, lithology, climatic conditions, and land-use practices (Panagos et al., 2015 ; Srejić et al., 2023 ; Malušević et al., 2025 ). Erosion is often a consequence of long-term deforestation and improper land use, leading to numerous adverse environmental impacts and significant economic losses (Milevski et al., 2024 ). Modeling results indicate that approximately 25% of land in the European Union experiences erosion rates exceeding the recommended sustainable threshold (2 t/ha/year). In comparison, more than 6% of agricultural land is affected by severe erosion (11 t/ha/year) (Panagos et al., 2020 ). Furthermore, soil erosion poses a serious threat to sustainable agriculture, with studies indicating a global decline in crop yields of approximately 0.4% per year due to erosion (Altobelli et al., 2020 ). In this context, soil erosion, as the primary driver of land degradation, represents a major challenge to ecosystem sustainability (Wang et al., 2025 ). Soil erosion is one of the most significant environmental challenges for sustainable natural resource management and the quality of life of populations across Europe (Prăvălie et al., 2024 ). Human activities and land-use changes, including intensive agriculture, deforestation, urbanization, and industrialization, have been identified as the primary drivers of accelerated erosion (Borrelli et al., 2020 ). Projections up to 2070 indicate that future erosion intensity will largely depend on the combined effects of climate change and land-use change, with different socio-economic scenarios and agricultural practices significantly modulating these processes (Borrelli et al., 2020 ). Contemporary soil management practices in agriculture have considerable potential to mitigate the negative effects of land degradation (Kaffas et al., 2018). Under the influence of climate change, precipitation erosivity forecasts indicate increases of approximately 30–66% in soil loss by 2070, highlighting the critical role of climate in the future dynamics of this process (Panagos et al., 2022 ; Životić et al., 2022; Nicosia et al., 2024 ). At the same time, the United Nations Convention to Combat Desertification (UNCCD) warns that, without urgent conservation measures, land degradation will continue to threaten the quality of life of billions of people worldwide (United Nations Convention to Combat Desertification, 2022 ). The scientific community has been addressing the challenges posed by soil erosion for several decades, at global, regional, and local scales, through the application of various types of empirical models for estimating soil erosion rates (Goodrich et al., 2012 ; Alewell et al, 2019 ). The development of the first soil erosion map of the Republic of Serbia began in 1966 and was published in 1983 (Lazarević et al., 1983). The Erosion Potential Method (EPM), developed by Gavrilović ( 1972 ) and later modified by Lazarević ( 1985 ) and Tošić & Dragićević (2012), remains one of the most frequently applied models in regional studies. Previous studies have confirmed the scientific validity of the EPM model (De Vente et al., 2005). Over the past two decades, the development of geospatial databases and the application of Geographic Information Systems (GIS) technology have significantly enhanced the implementation of empirical models, including the EPM model (Spalević et al., 2020; Polovina et al., 2024 ). In general, monitoring changes in soil erosion intensity is conducted on a regular basis. Addressing soil erosion requires continuous monitoring of processes, forms, and parameters across spatial and temporal scales, as changes in these factors directly influence the spatial patterns and intensity of erosion (Durlević et al., 2025 ). Assessing erosion rates and identifying the most susceptible areas are essential for selecting appropriate soil management and protection strategies (Papageorgiou et al, 2025 ; Rao et al., 2026 ). Given that the evaluation of soil erosion intensity represents the fundamental basis for planning prevention and protection measures, as well as for determining the type and extent of anti-erosion works, the main objective of this study is to produce a national-scale map of mechanical soil erosion intensity. The map is generated using contemporary high-resolution spatial and satellite data to identify the most susceptible municipalities and analyze their geospatial distribution through cluster classification based on dominant indicators. The spatial differentiation of results at the regional and local levels will enable more precise land management planning. The findings of this study will have significant practical applications, particularly for local governments, forest management services, and local communities, contributing to more efficient management of land and natural resources. 2. Materials and Methods 2.1. Study area Serbia is positioned in the northern temperate light-heat zone between 41°51' and 46°11' north latitude. It is a continental country whose territory is 84 km from the Adriatic, 202 km from the Aegean, and 405 km from the Black Sea. In terms of relief, three macromorphological units stand out in Serbia: plains, hilly areas, and mountainous regions (Ćalić et al., 2017 ). The plains (Pannonian and Moesian) are flat areas formed by block subsidence and sediment accumulation, with active fluvial and aeolian processes (dominated by Quaternary sediments). The Pannonian Plain covers northern Serbia, while the Moesian Plain extends to the far northeast. The hilly areas (peri-Pannonian and peri-Moesian) represent transitional zones between plains and mountains, with rougher relief and dominant slope processes and erosion (Oro et al., 2021 ). Lithologically, they consist of Neogene sediments (gravels, sands, clays, marls). The mountainous regions include the Dinarides, the mountains of the Vardar zone, the Serbo-Macedonian massif, and the Carpatho-Balkanides. Often, tectonic depressions (basins) with flat bottoms are found between mountains, formed by block subsidence and sediment accumulation (Carević et al., 2021 ). Together, these three units form the basic morphostructural division of Serbia. The territory lies within an elevation range between 28 m and 2,660 m (Velika Rudoka) (Durlević et al., 2024 ). Depending on altitude, annual precipitation in Serbia ranges from 500–600 mm to over 1100 mm. June precipitation averages between 60 and 140 mm, while February precipitation ranges from 30 to 100 mm (Stanišić et al., 2021 ). Above 1,800 meters, the mean annual air temperature is around 3°C. In southern Serbia and Belgrade, it exceeds 12°C (Pantić et al., 2024 ). A statistically significant increase in temperature has been recorded in most of the country, with only a slight decrease observed in the far southeast. These results indicate a clear warming trend across almost the entire territory of Serbia (Milovanović et al., 2017 ). The most widespread climate type in Serbia is Cfbq, which gradually changes into Cfbr in parts of eastern Serbia. The least represented is Efcr, present only on high mountains such as Prokletije and Šar Mountains (Stojković et al., 2023 ). The Cfaq climate occurs in lowland areas of central and northern Serbia (Vojvodina), at elevations up to about 120–140 meters. In the lowlands of central Serbia, the Dfbq climate prevails. Mountainous areas above 1200 meters in southwestern, southern, and southeastern Serbia are characterized by the Dfcq climate. At the highest elevations, around 2250 meters in Prokletije and Šar Mountains, the Dfcr climate is present (Milovanović et al., 2018 ). Under the influence of numerous climatic, geological, geomorphological, biological, and anthropogenic factors, Serbia’s hydrographic network has been formed. The average river network density is 323 m/km² (Urošev et al., 2017 ). In areas where annual precipitation exceeds 1000 mm (Prokletije, Šar Mountains), river network density goes beyond 2000 m/km². Conversely, the eastern and northeastern parts of the country receive less than 600 mm of precipitation, resulting in sparser, often intermittent watercourses (Tošić et al., 2025a ). This is especially true for karst areas (about 10% of the territory). Hydrological regimes differ: in the Black Sea basin, the continental regime dominates with maximum flow in spring, while in the Adriatic basin, a modified maritime regime is expressed, with peak precipitation in late autumn and early winter (Urošev et al., 2017 ). The largest amounts of water are formed on magmatic and metamorphic rocks. Altitude and slope also influence the number of watercourses, while water quality improves with elevation due to reduced anthropogenic pressure. In terms of vegetation zones, northern Serbia belongs to the steppe zone, while the central and southern parts are predominantly covered by deciduous forests (oak and beech) (Maruna et al., 2019 ; Kovačević et al., 2020 ). In the far south, within smaller areas, Mediterranean forests are also present. 2.2. Data Set and Methodology To calculate erosion intensity and related natural and anthropogenic factors, numerous spatial datasets were used. The vector state border of Serbia used for the preparation of Fig. 1 was obtained from the GeoSerbia geoportal (GeoSrbija, 2026 ). The Digital Elevation Model (DEM) with a 30 m spatial resolution was generated from the Copernicus Global Digital Elevation Model platform (European Space Agency, 2024 ). Larger cities and major rivers were digitized using data from OpenStreetMap (OpenStreetMap Team, 2026 ). For obtaining land-use information, spatial data from two European geospatial databases were used. The base layer was derived from Sentinel-2 imagery from 2024 obtained from the Environmental Systems Research Institute (ESRI) database, while data from the Copernicus Land Monitoring Service were integrated to extract open forests and closed (dense) forests (Copernicus Land Monitoring Service, 2023 ; Environmental Systems Research Institute, 2024 ). Using this approach, eight land-use classes were obtained: water bodies, open forests, closed (dense) forests, seasonally flooded areas, agricultural plots, settlements, bare land, and rangeland (grasslands and pastures). The land-use dataset was resampled from 10 m to 30 m to ensure consistency with the other input parameters used in the analysis. The resampling was performed using the nearest-neighbor method, which preserves the original categorical values of land-use classes and prevents the creation of mixed or interpolated class values that may occur with other resampling techniques. Data on geology, i.e., rock types in Serbia, were obtained by digitizing the lithological content from the Geological Map at a scale of 1:300,000 (Ministry of Mining and Energy of the Republic of Serbia, 2026 ). The rocks were classified according to their resistance to mechanical soil erosion. Subsequently, the data were converted from vector to raster format in QGIS v3.40.9, with a pixel resolution of 30 m (QGIS Development Team, 2026 ). The terrain slope was derived from the Copernicus Global Digital Elevation Model geoportal, with a spatial resolution of 30 m (European Space Agency, 2024 ). For the coefficient of erosion type and extent, land-use and terrain slope data were integrated to provide a more objective representation. Both parameters were calibrated into a single layer, using data from 2024. The spatial resolution of all input parameters was standardized to 30 m in QGIS software. For the purposes of this study, the intensity of the erosion process was calculated using the Erosion Potential Model (EPM), also known as the Gavrilović method (Gavrilović, 1972 ). Results of relevant research in the field of erosion modeling have shown that, among 11 analyzed semi-quantitative methods, the EPM was rated as the most quantitative (De Vente & Poesen, 2005 ). The chosen research approach is aligned with the key geographical (spatial) dimension of the problem, as the application of the erosion model provides the most reliable results under the conditions for which it was developed. Validation of the model based on sediment yield data from over 100 watersheds worldwide has shown that the best agreement of results is observed in continental and temperate regions (Bezak et al., 2024 ). Furthermore, its application in combination with other models has provided a satisfactory level of accuracy and reliability of results (Efthimiou et al., 2016 ; Raza et al., 2021 ; Trendafilov et al., 2024 ). Intensive application of the model at the local, national, and regional levels in recent decades has confirmed its scientific and practical value (Gocić et al., 2020 ; Aleksova et al., 2023 ; Stefanidis et al., 2025 ). The simple structure of the model and its low input data requirements make the EPM suitable for regional analyses and planning of soil conservation measures (Dildabek et al., 2025 ; Smanov et al., 2025 ; Akhmetova, 2025 ). The use of modern remote sensing methods and good integration with GIS tools has allowed the previous shortcomings of the model to be overcome (Dominici et al., 2020 ; Ćurić et al., 2022 ; Durlević et al., 2025 ). This has significantly reduced the subjectivity in the assessment of individual parameters (Oshunsanya et al., 2025 ). The erosion coefficient (Z) is calculated as follows: Z = Y · X · (φ + \(\:\sqrt{I}\) ) , where: Y—coefficient of soil resistance; X - soil protection coefficient; φ - erosion and stream network development coefficient, and I - average slope (%) (Fig. 2 ). The classification of values for the soil resistance coefficient (Y) and the vegetation protection coefficient (X) is presented in Table 1 . Table 1 Classification of Soil Resistance and Vegetation Protection Coefficients (Y and X). Coefficient of Soil Resistance Y value Fine sediments and soils without erosion resistance 0.80–1.00 Sediments, moraines, clay and other rock with little resistance 0.60–0.80 Weak rock, schistose, stabilized 0.50–0.60 Rock with moderate erosion resistance 0.30–0.50 Hard rock, erosion resistant 0.15–0.30 Coefficient of soil protection X Value Areas without vegetal cover 0.80–1.00 Damaged pasture and cultivated land 0.60–0.80 Damaged forest and bushes, pasture 0.40–0.60 Coniferous forest with little grove, scarce bushes, bushy prairie 0.25–0.40 Thin forest with grove 0.20–0.25 Mixed and dense forest 0.15–0.20 The classification of values for the erosion type and extent coefficient (φ) is presented in Table 2 . The average terrain slope (√I) is expressed as a percentage in decimal form. Table 2 Classification of Soil Resistance and Vegetation Protection Coefficients (Y and X). Coefficient of erosion type and extent φ value Areas are strongly dissected by ravines, interfluves, and furrows, developed in eluviated and diluviated deposits or highly weathered, low-resistance rocks; such terrains are unsuitable for use without targeted anti-erosion measures. 0.81–1.00 Firm surfaces with concealed linear erosion on slopes exceeding 10°; degraded forests and pastures characterized by rills and small divides. 0.61–0.80 Cultivated areas on slopes of 5–10°; degraded forests and pastures with reduced vegetation cover; barren terrains on resistant, impermeable rocks; active incision of watercourses into the terrain. 0.31–0.60 Latent erosion processes; well-developed forests with dense ground cover occupying ridge zones; stable meadows and pastures on slopes up to 5°; extensive lowland areas with slopes below 5° 0.11–0.30 Relatively flat terrain (< 2°) where the weakest forms of erosion occur, typically in alluvial plains. 0.10 The EPM erosion categorization is shown in Table 3 . According to the research objectives set in this paper, the intensity of erosion will be considered using the Z coefficient (Lazarević, 1985 ). Previous research has shown that the most important control factor in the intensity of soil erosion is land use change represented in the EPM model as the X coefficient (Manojlović et al., 2018 ). Table 3 Classification of Soil Erosivity Categories and Erosion Coefficient (Z). Soil erosivity category Strength of Erosion Processes Erosion Coefficient (Z) SA Sediment accumulation 0 V 2 Very weak 0.01–0.10 V 1 Very weak 0.11–0.20 IV 2 Weak 0.21–0.30 IV 1 Weak 0.31–0.40 III 2 Medium 0.41–0.55 III 1 Medium 0.56–0.70 II 2 Intensive 0.71–0.85 II 1 Intensive 0.86–1.00 I 3 Excessive 1.01–1.20 I 2 Excessive 1.21–1.40 I 1 Excessive 1.41–1.50 In the first segment of the research, the territory of Serbia was differentiated into zones with varying degrees of soil erosion susceptibility. However, to achieve more detailed spatial differentiation and identify spatial patterns of soil erosion, a cluster analysis method was applied. The municipality was selected as the basic spatial unit. Of the 197 municipalities in Serbia, 181 were included in the cluster classification. In contrast, the urban municipalities of Belgrade (10) and Niš (5), as well as the City of Novi Sad, were excluded from the analysis. The selected municipalities were characterized using a set of six indicators. The mean erosion coefficient (Z av ) was chosen as the primary indicator of soil erosion. The second indicator was the coefficient of variation of the erosion process (CV z ), which enabled the assessment of the variability of erosion intensity within each municipality. In this study, the coefficient of variation proved to be a highly significant statistical measure of landscape spatial homogeneity or heterogeneity. The influence of land use was quantified using the proportions of forest (F) and agricultural land (AgL). Morphometric characteristics of each municipality were defined by the mean slope angle (S) and the mean elevation (E). Cluster analysis is a statistical method for obtaining optimal groupings based on the similarity or distance between sample characteristics (Xu & Tian, 2015 ). Its primary objective is to achieve maximum similarity within clusters while ensuring high heterogeneity between clusters, as well as to identify hidden correlations within the data (Andrade et al., 2008 ; Chen et al., 2016 ). Cluster analysis has proven to be an appropriate method for differentiating soil erosion intensity across various spatial units in Serbia (Srejić et al., 2025 ). In this study, Agglomerative Hierarchical Clustering (AHC) was applied. The AHC algorithm begins by treating each data point as an individual cluster and progressively merging the most similar pairs into larger clusters. Clusters are merged based on the similarity of their centroids, determined by their proximity in the feature space. Thus, clusters with the highest similarity according to the selected linkage criterion are combined. Similarity between data points is quantified using distance measures, and in this study, the Euclidean distance was employed (Raux et al., 2011 ; Yang et al., 2022 ). Different linkage criteria influence the shape and size of clusters, thereby significantly affecting the resulting dendrogram. In this research, Ward’s linkage method was used, which merges clusters by minimizing the increase in total within-cluster variance at each step. As a result, this method produces clusters that are approximately equal in size, highly compact, and characterized by a relatively uniform distribution of data within each cluster (Wani et al., 2024). 2.3. Soil erosion inventory and validation method For validation purposes, a soil erosion inventory consisting of 102 erosion and 109 non-erosion samples was compiled through the integration of historical records, comparative analysis with recent conditions, and the interpretation of GIS-based datasets and satellite imagery (Fig. 3 ). Locations classified as erosion-prone correspond to clearly exposed (bare) surfaces occurring on steep slopes and underlain by highly erodible geological materials. In contrast, non-erosion locations represent stable environments, predominantly characterized by dense forest cover on gentle or flat terrain, developed over resistant geological substrates. The inventory captures representative examples of both erosion-prone and stable conditions across the study area and serves as a reference dataset for the quantitative validation of the EPM-derived erosion coefficient. All procedures and approaches employed in this research are outlined in the flow chart provided in Fig. 4 . 3. Results 3.1. National-Scale Spatial Mapping of Erosion Intensity On the territory of Serbia, very weak erosion prevails (53.8%). It is found in mountainous and hilly areas covered with forest vegetation. In mountainous regions, under conditions of extensive agriculture with meadows and pastures and thin soil layers, slightly higher erosion occurs (Z = 0.11–0.20). In lowland areas such as the Pannonian and Moesian plains and the alluvial plains of major rivers, erosion is weak, occurring only in narrow zones around tributary streams where the banks descend toward the riverbed. Beyond these zones, erosion shifts to moderate or intensive. Medium erosion (Z = 0.41–0.70) is characteristic of hilly areas built of clastic sediments, where surfaces are predominantly covered with orchards. This intensity of erosion is also present in the zone of the Pannonian Plain due to the dune relief (Deliblato Sands) (Fig. 5 ). In the hilly zone, on slopes below 20°, under conditions of cultivated land (arable fields and orchards), intensive and excessive erosion is present. These areas are located on the edges of basins and in the immediate surroundings of larger urban centers. Altogether, they cover 8.3% of the territory, of which 7.6% falls into the intensive class and 0.7% into the excessive class (Table 4 ). Table 4 Distribution of Erosion Intensity Coefficient (Z) in the Republic of Serbia. Strength of Erosion Processes Erosion Coefficient (Z) Total area (km 2 ) Share in total area (%) Sediment accumulation 0 853.1 1.0 Very weak 0.01–0.10 24,215 27.4 Very weak 0.11–0.20 23,391.5 26.4 Weak 0.21–0.30 16,968.5 19.2 Weak 0.31–0.40 4,464.3 5.0 Medium 0.41–0.55 7,516 8.5 Medium 0.56–0.70 3,729.6 4.2 Intensive 0.71–0.85 3,711.7 4.2 Intensive 0.86–1.00 2,976.9 3.4 Excessive 1.01–1.20 281.2 0.3 Excessive 1.21–1.40 251.2 0.3 Excessive 1.41–1.50 79 0.1 3.2. Spatial differentiation of municipalities based on erosion indicators The results of the Agglomerative Hierarchical Clustering (AHC) indicate that municipalities in Serbia are grouped into six distinct clusters. The spatial distribution of these municipalities is presented in Fig. 6 , while the main characteristics of each cluster are summarized in Table 5 . A high proportion of between-cluster variance (74.2%) relative to within-cluster variance (25.8%) suggests strong homogeneity within clusters and clear separation between them. This reflects the underlying structural characteristics of the analyzed dataset. The clustering of municipalities was further validated using the Kruskal–Wallis test (Tölgyesi et al., 2014 ; Pinto et al., 2019 ), which confirmed that the differences among the identified clusters are statistically significant for all analyzed indicators (p < 0.001). Table 5 Controlling variables used in the AHC. Variables – Abbreviation (units) Clusters I II III IV V VI Coefficient eropsion Zav (-) 0.173 0.227 0.232 0.285 0.358 0.245 Coefficient of variation CVz (%) 54.2 63.9 75.1 75.5 72.7 41.2 Forest cover F (%) 82.0 62.9 60.2 36.7 33.1 6.7 Agricultural land AgL (%) 2.2 5.5 20.4 47.4 52.5 83.4 Altitude A (m) 803.5 900.8 513.9 209.7 357.0 88.1 Slope S (°) 15.5 13.8 9.7 4.8 5.5 0.9 Cluster I predominantly comprises the peripheral and border areas of Serbia and includes 22 municipalities (Fig. 6 ). This spatial unit encompasses the highest parts of the Dinaric Mountains in the west, the Šar Mountains in the south, the Serbo–Macedonian Massif in the southeast, and sections of the Carpathian–Balkan Mountains in the east. It covers approximately 13% of Serbia’s territory, i.e., 11,654 km². This cluster is characterized by the lowest erosion intensity (Z av = 0.173). The lowest erosion coefficient was identified in the municipality of Vranjska Banja (Z av = 0.145), while the highest was recorded in Štrpce (Z av = 0.216) (Fig. 7 a). The municipality of Štrpce is distinguished by the highest mean elevation (A = 1,336 m.a.s.l.) and the steepest average slope (S = 18.9°). However, despite a high proportion of forest cover (F = 68%), the elevated erosion coefficient is associated with a substantial share of meadows and pastures (29%). In terms of the uniformity of soil erosion intensity, this is one of the most homogeneous clusters (CV z = 54%) (Fig. 7 b). Within this predominantly mountainous region, with a mean elevation of A = 804 m.a.s.l., one-third of the municipalities have an average elevation exceeding 1,000 m.a.s.l. (Fig. 7 e). All municipalities exhibit an average slope greater than 10°, with a cluster mean of S = 15.5° (Fig. 7 f). These morphometric characteristics have resulted in very limited agricultural land use. Agricultural land is nearly absent (AgL = 2%) (Fig. 7 d), whereas forests cover as much as 82% of the cluster area (Fig. 7 c). Cluster II is slightly larger than the previous one and comprises 26 municipalities located in the western, southwestern, and southeastern mountainous regions of Serbia (Fig. 6 ), including parts of the Dinaric Mountains, the Šar Mountains, and the Serbo–Macedonian Massif. This cluster covers 14,017 km², corresponding to 16% of Serbia’s territory. It is characterized by a weak erosion category, with an average erosion coefficient, Z av , of 0.227. The lowest erosion coefficient was recorded in Arilje (Z av = 0.181), while the highest was observed in Gora (Z av = 0.345) (Fig. 7 a). The variability of the erosion process is somewhat higher than in the previous cluster (CV z = 64%). The highest homogeneity of the erosion coefficient is found in Gora (CV z = 51%), while the lowest is observed in Dečani (CV z = 77%) (Fig. 7 b). A high proportion of forest cover (F = 63%) and a very low share of agricultural land (AgL = 5.5%) are the main land-use characteristics of this spatial unit (Fig. 7 c,d). Municipalities within this cluster exhibit the highest average elevation (A = 900 m.a.s.l.), ranging from A = 609 m.a.s.l. in Svrljig to A = 1,707 m.a.s.l. in Gora (Fig. 7 e). The mean slope angle is S = 14° (Fig. 7 f). Cluster III is the most numerous and spatially extensive cluster, comprising 45 municipalities and covering 28% of Serbia’s territory (24,882 km²). It is most widely distributed in Eastern Serbia, within the West Morava and Kosovo and Metohija basins (Fig. 6 ). In terms of soil erosion susceptibility, this spatial unit does not differ significantly from the previous cluster, with an average erosion coefficient of Z av = 0.232. The municipalities of Golubac and Krupanj exhibit the lowest erosion coefficients (Z av = 0.191), while Glogovac records the highest value (Z av = 0.284) (Fig. 7 a). The coefficient of variation (CV z = 75%) indicates greater spatial variability in the erosion process than in the previous two clusters (Fig. 7 b). A substantial portion of this cluster is covered by forests (F = 60%) (Fig. 7 c). However, at lower elevations (A = 514 m.a.s.l.) and on gentle slopes (S 30%) and steeper slopes are predisposed to the highest erosion rates within this cluster. Cluster IV covers 12,287 km², accounting for 14% of Serbia’s total area, and includes 37 municipalities. The largest continuous area of this cluster is located south of the Sava and Danube rivers, as well as within the immediate basin of the Velika Morava. Smaller areas are found on the southern slopes of Fruška Gora. At the same time, the southernmost extent reaches the valleys of the Južna Morava and Nišava rivers, including the peri-urban zone of Niš (Fig. 6 ). The average erosion coefficient is Z av = 0.285. The municipalities of Smederevo (Z av = 0.330), Lajkovac (Z av = 0.317), and Batočina (Z av = 0.315) exhibit the highest values, whereas Kostolac (Z av = 0.229) and Šid (Z av = 0.242) have the lowest erosion intensity (Fig. 7 a). The average coefficient of variation is CV z = 73%. High values in certain municipalities (CV z > 80%) indicate not only internal heterogeneity of erosion processes but also diverse geographical conditions (Fig. 7 b). This specific combination of factors has resulted in a higher level of erosion susceptibility. Nearly half of the available land in this cluster is used for agriculture (AgL = 47%) (Fig. 7 d), while forest cover is considerably lower compared to previous clusters (F = 37%) (Fig. 7 c). Lower elevations (A = 210 m.a.s.l.) and gentle slopes (S < 5°) are the main characteristics of this predominantly agrarian landscape (Fig. 7 e,f). Cluster V covers 4,015 km², accounting for approximately 5% of Serbia’s territory. It is the smallest cluster in terms of both spatial extent and the number of municipalities, comprising 12 municipalities distributed across several areas. The western area includes the municipalities of Vladimirci, Ub, and Koceljeva. The central part consists of Smederevska Palanka, Mladenovac, Topola, and Rača, located within the immediate basin of the Velika Morava. The southernmost municipalities belong to the Kosovo and Metohija basins (Orahovac, Klina, Srbica, Vučitrn, and Obilić) (Fig. 6 ). The distinguishing feature of this cluster is the highest susceptibility to soil erosion (Z av = 0.358). The highest erosion intensity is recorded in the municipalities of Vladimirci (Z av = 0.413) and Smederevska Palanka (Z av = 0.400) (Fig. 7 a), placing them in the moderate erosion category. The average coefficient of variation of the erosion process is CV z = 73%, which is comparable to Clusters III and IV (Fig. 7 b). Forests cover approximately one-third of the cluster area, slightly less than in the previous cluster, with the largest shares in Koceljeva (F = 47%) and Srbica (F = 46%) (Fig. 7 c). Agricultural land is highly represented, covering nearly half of the cluster (AgL = 52%), and in some municipalities, it exceeds 70% (e.g., Smederevska Palanka) (Fig. 7 d). A key characteristic of this spatial unit is the combination of moderately higher elevation (A = 357 m.a.s.l.) and relatively steeper slopes (S = 5.5°) (Fig. 7 e,f) on land predominantly used for agricultural production. This synergy of geographical factors increases the terrain's susceptibility to erosion. Cluster VI covers 19,311 km², representing approximately one-fifth of Serbia’s territory. It forms the most homogeneous continuous area in northern Serbia and includes 39 municipalities (Fig. 6 ). This spatial unit falls into the weak erosion category, with an average erosion coefficient Z av = 0.245 (Fig. 7 a). The lowest coefficient of variation (CV z = 41%) indicates minimal spatial variability of the erosion process (Fig. 7 b), reflecting the high uniformity of geographical conditions in this intensively used agricultural region. Despite a very high proportion of agricultural land (AgL = 83%) and a low share of forest cover (F = 7%), this cluster is not predisposed to higher erosion rates (Fig. 7 c,d). The key controlling factor is terrain morphometry, particularly low elevation (A < 100 m.a.s.l.) and very gentle slopes (S < 1°) (Fig. 7 e,f). However, even slight increases in these parameters lead to higher erosion intensity. For example, the municipalities of Alibunar and Vršac do not differ significantly in land use from other municipalities in this cluster, yet they exhibit the highest erosion intensity (Z av = 0.275) due to somewhat higher elevations and steeper slopes (A > 100 m.a.s.l.; S ≈ 2°). 3.3. Validation Results To evaluate the performance of the Erosion Potential Model (EPM), a validation procedure was conducted using the soil erosion inventory (Fig. 3 ), consisting of 211 georeferenced locations, including 102 erosion and 109 non-erosion samples. The validation was performed by extracting the EPM-derived erosion coefficient (Z) values at the sample locations and assessing the ability of the model to discriminate between erosion-prone and stable environments. A binary classification was applied using a threshold value, which was determined through systematic evaluation and identified at Z = 0.11. The obtained results indicate a complete separation between erosion and non-erosion samples, with all evaluation metrics reaching their maximum values (ROC–AUC = 1.00; accuracy = 1.00; precision = 1.00; recall = 1.00; F 1 -score = 1.00). The confusion matrix confirms that all non-erosion samples (n = 109) and all erosion samples (n = 102) were correctly classified, with no false positives or false negatives observed. The distribution of Z values further supports this finding. Non-erosion samples are consistently associated with low Z values, while erosion samples exhibit significantly higher values, with no overlap between the two groups (Fig. 8 and Fig. 9 ). A perfect confusion matrix and ROC curve occur because the data are clearly separated, with no “borderline” cases between erosion and stable zones. The model and the validation data are based on the same criteria (slope, vegetation, lithology), which leads to complete agreement. The ROC curve (Fig. 10 ) demonstrates the strong discriminative structure of the Z coefficient, confirming its effectiveness in separating clearly defined erosion-prone and stable conditions. Although ROC analysis and confusion-matrix-based evaluation are commonly used in predictive classification settings, in the present study these metrics should be interpreted primarily as measures of agreement between the EPM-derived erosion coefficient and the expert-defined erosion inventory. Because the validation samples were selected using geomorphological and land-cover criteria closely related to the EPM input factors, the observed perfect separation reflects strong model–inventory consistency rather than fully independent predictive generalization. Therefore, the validation results confirm that the EPM model provides a robust and internally consistent representation of soil erosion susceptibility under clearly defined environmental conditions. 4. Discussion 4.1. Comparative Analysis with Global Soil Erosion Studies As already mentioned, among the analyzed semi-quantitative methods applied worldwide, EPM was rated as the most quantitative (De Vente & Poesen, 2005 ). Also, comparative analysis of sediment yield estimations using different empirical soil erosion models showed that the EPM model is the most appropriate concerning the fulfilment of various tasks such as: estimation of total transport material, assessment of average pattern of erosion risk, identification of high-risk areas, identification of hot spots, detailed erosion and deposition patterns, effects of conservation measures (Blinkov & Kostadinov, 2010 ). The EPM model can be used to estimate all soil erosion types, meeting the needs of various scientific fields and sectors (agriculture, forestry, water resources, and watershed management). Also, the advantage of the EPM model is its applicability across all scales (Efthimiou et al., 2017 ). Finally, qualitative analysis showed that the EPM model can perform best for mountainous areas allowing for the identification on the map and visualization of the areas most prone to erosion, as well as the predicted amount of soil loss and its spatial distribution. A comparative analysis with global, regional, and local soil erosion studies showed that the results of this study are in line with previous research. In a study of the applicability of the EPM model and modified EPM (mEPM) to estimate gross and net erosion rates at the global scale, the analysis shows that the land use pattern has the greatest impact on the model results, as verified by the X coefficient in model aquation (Bezak et al., 2024 ). Soil loss also varies with vegetation cover and land use management practices, which make both extremely influential factors in erosion formation and overall soil loss in the Mediterranean regions (Kosmas et al., 2000; Stefanidis & Stathis, 2018 ). Analysis of the spatial distribution of land use as a determining factor of the intensity of soil erosion has also been detected in regional studies in Bosnia and Hercegovina, which is clearly visible in the spatial distribution of the soil erosion protection coefficient (Tošić et al., 2025b ). Previous research has established a functional relationship between land use changes and soil erosion intensity at the local level, linking these processes with processes of deagrarization in rural settlements (Srejić et al., 2025 ). Spatial differentiation of torrential watersheds in border regions in Serbia has revealed varying trends in soil erosion, highlighting the importance of accurate assessments of these changes through land use impact indicators (Petrović et al., 2024 ). Also, results from small catchments in Croatia showed a good approximation of erosion intensity and soil loss to land cover change at different annual and seasonal spatial-temporal scales (Dragičević et al., 2018 ). Namely, previous research in Serbian watersheds found that the EPM method is most sensitive to the X coefficients, followed by the φ and Y coefficients, while slope, air temperature, and precipitation have a smaller effect on the EPM output. The correlation matrix of EPM model parameters at the statistical significance level of α = 0.05 showed that the coefficient X has the major impact on the sediment gross erosion and coefficient erosion (W = f (X), r = 0.778, and Z = f (X), r = 0.820, respectively) (Manojlović et al., 2018 ). It can be concluded that coefficient X is the primary factor of the strength of the erosive process. Also, the sensitivity estimates of the USLE models indicated that land cover (C factor) is the most important factor in determining soil loss across different mountain and agricultural systems (Perovic et al., 2016 , Estrada-Carmona et al., 2017 ). Therefore, Bezak et al. ( 2024 ) point out that the land use/land cover parameter is particularly important in both the EPM and USLE models, as the main determinant of changes in soil erosion intensity (Bezak et al., 2024 ). 4.2. Spatial differentiation soil erosion intensity according clustering municipaliy Although the overall erosion intensity classification indicates that erosion processes in Serbia are predominantly very weak and weak, areas affected by moderate and intensive erosion are also present. The spatial distribution of erosion intensity, differentiated through cluster analysis of dominant indicators at the municipal level, offers a new perspective on the spatial analysis of this fundamentally geomorphological process. The results obtained using the EPM method and GIS-based modelling highlight the significant roles of terrain morphometric parameters and anthropogenic pressure in shaping the spatial distribution of erosion risk (Kostadinov et al., 2018 ; Đorđević et al., 2026 ). In this context, the influence of key controlling indicators was examined. In addition to forest and agricultural land, areas covered by meadows and pastures were included in the analysis as transitional forms between these two land-use types. The quantitative relationships among the selected indicators were assessed using Pearson’s correlation coefficient within a correlation matrix (Table 6 ). The results indicate a high level of correlation among the indicators at a significance level of α < 0.0001. Table 6 Correlation matrix (Pearson’s r) of analyzed variables. Variables Z F AgL Gr E S Z 1 -0.767 0.826 -0.435 -0.624 -0.805 F -0.767 1 -0.863 0.396 0.685 0.920 AgL 0.826 -0.863 1 -0.594 -0.729 -0.882 Gr -0.435 0.396 -0.594 1 0.741 0.574 E -0.624 0.685 -0.729 0.741 1 0.861 S -0.805 0.920 -0.882 0.574 0.861 1 The results of the quantitative analysis of controlling factors of soil erosion indicate a strong relationship between elevation and the decrease in erosion intensity in Serbia. The erosion coefficient (Z) shows a high negative correlation with morphometric indicators, namely slope angle S (r = − 0.805) and elevation E (r = − 0.624), while these two variables are strongly positively correlated with each other (r = 0.860). For example, in some border municipalities in eastern Serbia, the correlation for the function Z = f(E) indicates an even stronger dependence of decreasing erosion intensity with increasing elevation (r = − 0.98) (Manojlović et al., 2018 ). The spatial variation of the erosion coefficient Z, differentiated according to relief conditions, shows that areas affected by higher erosion intensity increase from mountainous border municipalities (clusters I and II) toward municipalities predominantly located in hilly zones characterized by weakly dissected and gently undulating terrain (cluster V). Areas with preserved forest cover, particularly in mountainous regions (represented by clusters I and II), exhibit significantly lower Z coefficient values compared to degraded pastures and open cropland. Forest cover shows a strong negative correlation with the erosion coefficient (Z = f(F), r = − 0.767). Forested areas are predominantly located on steeper slopes (r = 0.919) and at higher elevations (r = 0.684). These strong positive relationships between forests and morphometric indicators explain their anti-erosion effect in parts of Serbia that are naturally predisposed to higher erosion categories (greater relief dissection and steeper slopes), yet still record low erosion rates. Slopes exceeding 20° are characteristic of mountainous areas (clusters I and II), which are lithologically composed of hard and compact rocks. Due to subaerial processes, a weathering crust develops on the surface, which is susceptible to intensive washing. However, erosion potential does not increase linearly with slope. On slopes steeper than 40°, the weathering crust is largely removed, resulting in a significantly reduced erosion potential. Grasslands and pastures are also negatively correlated with the erosion coefficient Z (r = − 0.435). These land-use types are predominantly located at higher elevations, as indicated by the positive correlation in the function Gr = f(E) (r = 0.741), and often co-occur with forest cover (r = 0.395). Higher elevation and a greater proportion of forest cover contribute to a reduction in erosion intensity. However, demographic characteristics of border municipalities have also influenced the relatively low erosion rates. Serbia is characterized by highly differentiated rural areas with pronounced processes of regional demographic differentiation (Gajić et al., 2021 ; Grčić et al., 2024 ; Gajić Protić et al., 2024 ). A large number of settlements with weak demographic potential indicates a decline in agricultural activity (Manojlović et al., 2022 ). Under the influence of various socio-economic factors (industrialization, agrarian reform, and rural–urban migration) during the second half of the 20th century, sparsely populated and underdeveloped rural settlements in mountainous peripheral and border regions experienced strong processes of depopulation and deagrarization (Martinović & Ratkaj, 2015 ). The abandonment of agricultural land has consequently contributed to a reduction in erosion intensity in these municipalities. On the other hand, agricultural land (AgL) has the strongest influence on erosion intensity (Z = f(AgL), r = 0.826). Agricultural areas are primarily located at lower elevations (r = − 0.728) and on terrains with lower slope angles (r = − 0.882). This indicates that the most intensively cultivated parts of municipalities in Serbia are also the most susceptible to erosion processes. The spatial differentiation of municipalities at risk of erosion, determined through cluster analysis, reveals a general gradient: peripheral border municipalities (clusters I and II) – municipalities in the transitional zone between hilly and mountainous terrain (cluster III) – municipalities predominantly located in hilly and partly lowland areas (clusters IV and V). In this context, municipalities located in the northern part of central Serbia exhibit the highest erosion intensity rates. These municipalities are geographically predisposed to increased land degradation: clastic rocks cover more than 70% of the territory, dominated by Neogene sediments, combined with a high proportion of arable land and reduced forest cover (Table 7 ). Additionally, these areas possess the highest demographic and agrarian potential, characterized by intensive agricultural production, which results in stronger anthropogenic pressure on land resources and intensification of erosion processes (Srejić et al., 2025 ). Table 7 Geographical characteristics of clusters. Cluster Description I Municipalities whose territory is predominantly located in mountainous zones with moderately (100–300 m/km²) and highly dissected relief (300–800 m/km²). The lithology is composed of magmatic and metamorphic rocks, as well as limestones. On slopes up to 35°, a weathering mantle has developed, with a thickness ranging from a few centimeters to several meters. In terms of land use, forests have a dominant share (> 70%), while arable land accounts for less than 10%. II Municipalities located in mountainous areas with the presence of hilly and lowland terrains. In addition to highly and moderately dissected relief, this cluster also includes areas with weakly dissected relief (30–100 m/km²). Alongside the lithological composition characteristic of Cluster 1, unconsolidated clastic sedimentary rocks, primarily Neogene sediments (clays, sands, and gravels) are also present. III Municipalities situated in the transitional zone between hilly and mountainous regions. Areas with moderate and weak relief dissection are significantly represented. The proportion of clastic sedimentary rocks and agricultural land (arable fields and orchards) increases, while forest cover remains above 50%. IV Municipalities where mountainous terrain is present, but hilly and lowland landscapes dominate, characterized by weakly dissected, slightly undulating (5–30 m/km²), and flat relief (0–5 m/km²). Clastic sedimentary rocks cover more than 70% of the territory, dominated by Neogene sediments with a notable presence of Quaternary deposits. There is an increased share of arable land and a reduced proportion of forest cover (< 50%). V Municipalities whose territories are almost entirely located in hilly regions with predominantly weakly dissected and slightly undulating relief. Clastic sedimentary rocks, mainly Neogene sediments, cover nearly the entire municipal territory. Forest cover is limited, while the share of arable land has significantly increased. VI Municipalities entirely situated in lowland areas (Pannonian Plain), characterized by predominantly flat relief (0–5 m/km²). In terms of lithology, only clastic sediments of Quaternary age are present (sands, loess, and clays). Regarding land use, arable land (predominantly cropland) accounts for more than 60% of the territory, while forests cover less than 10% of the total area. 4.3. The Role of Forests and Agriculture in Land Management for Soil Erosion Mitigation Forests play a key role in preventing soil erosion because they provide natural mechanisms for soil stabilization, increase water retention, and reduce the intensity of runoff and denudation. Several interrelated mechanisms mitigate soil erosion through forest ecosystems. One of the main mechanisms is the increase in cohesion through root systems, which strengthen the soil through a stable network, preventing the displacement of soil masses (Gonzalez-Ollauri & Mickovski, 2020 ). Various forest species, such as Scots pine (Pinus nigra), are among the most successful for afforestation of bare and eroded terrain in Serbia, demonstrating high adaptability to such terrain (Ratknić et al., 2025 ). Natural beech forests (Fagus sylvatica) are biologically very stable. They are associated with very low erosion due to their dense crown cover, thus reducing the direct "bombardment" of raindrops on the soil (Lukić et al., 2022 ). Dense vegetation reduces erosion many times over compared to bare terrain. Species such as hawthorn (Crataegus monogyna) show the highest values ​​of root density and the greatest reduction in erosion compared to alder and medlar (Dalir et al., 2025 ). At the same time, erosion processes in Serbia are also intensifying under the influence of anthropogenic factors and various climatic stresses. Some authors, such as Braunović et al. ( 2025 ), show that several factors have a significant influence and increase the susceptibility to water and wind erosion, such as excessive grazing, combined with deforestation and conversion of forest areas to pastures, which leads to a reduction in plant cover and compaction of the surface soil layer (Braunović et al., 2025 ). For these reasons, the maintenance and restoration of forest areas must be seen as a key stabilizer of the degradation and devastation of soil systems. Intensive agriculture on sloping terrain (especially trench crops) without implementing conservation practices results in the highest levels of soil loss (Bezbradica et al., 2023 ). Historical analyses show that the expansion of arable land in the 20th century was accompanied by a more pronounced development of ravine forms (Ristić et al., 2012 ). Simultaneously, rising temperatures are causing forest drying, especially in pedunculate oak (Quercus petraea). This process leads to physiological weakening of trees and thinning of crown cover, thereby weakening the protective function of forest ecosystems (Fernández et al., 2025 ; Sukmawijaya et al., 2026 ). With the loss of dense vegetation cover, raindrops have a strong, direct effect on the soil surface, increasing aggregate destruction and surface runoff (Ascencio-Sanchez et al., 2025 ). On the other hand, agriculture is a critical factor in exacerbating erosion when practised without conservation measures. Intensive cultivation with trench crops on sloping terrain accelerates the mineralization. Soils with a low humus content are less resistant to the destructive force of raindrops (Deđanski et al., 2024 ). Modern concepts of sustainable agricultural development suggest an interaction between forestry and agriculture, grounded in the principle of complementarity within a common landscape context. Mixed agroforestry systems with a dominance of native species, contour tillage, terracing, and the use of cover crops reduce erosion risk by restoring some of the ecosystem functions of natural vegetation (Golijanin et al., 2022 ). Several studies recommend carrying out group-selective and valley fellings on steep terrain rather than clear-cutting to limit and prevent exposure and erosion risks. 5. Conclusions With the ongoing processes of urbanization, depopulation, and deagrarization in Serbia, a decline in the intensity of mechanical soil erosion has been observed. In this study, geospatial data (lithology, land use, terrain slope, and erosion coefficient) with a 30 m spatial resolution were analyzed to produce a high-resolution national-scale map of mechanical soil erosion. Using the Erosion Potential Model (EPM), the results indicate that the mean erosion coefficient (Z av ) for Serbia is 0.239, reflecting a decrease in erosion intensity compared to 1985. The findings clearly demonstrate that areas most affected by intensive erosion are agricultural lands located on steeper slopes (> 10°) and under specific geological conditions (sand, gravel, organogenic, and clay-rich sediments). Overall, 7.6% of the territory is exposed to intensive erosion, while 0.7% of the study area is affected by excessive erosion. A perfect ROC-AUC score of 100%, obtained based on the inventory of erosion points, indicates strong consistency between the input data and the final results. To enable a more detailed understanding of erosion processes at the local level, spatial differentiation was performed through cluster analysis of the final dataset at the municipal scale. Six clusters were identified based on six indicators: mean erosion coefficient, coefficient of variation of the erosion process, the influence of forest cover, agricultural land, and morphometric characteristics (elevation and slope). The results clearly indicate that morphometric conditions and land use play a key role in controlling erosion dynamics in Serbia. The highest susceptibility to erosion was identified in agriculturally dominated areas with moderate slopes (Cluster V), where the combination of steeper terrain and intensive land use significantly enhances erosion processes. In contrast, the lowland regions of northern Serbia (Cluster VI), although predominantly agricultural, exhibit low erosion potential due to minimal terrain slope. Forest-covered areas proved highly effective at mitigating soil erosion. Future research in Serbia should focus on integrating existing erosion data with high-resolution climatic datasets to improve current results and support the development of a national sediment yield map as a key component of the EPM. More efficient land management could be achieved through detailed analyses at finer spatial scales. In this context, the application of the EPM at the level of rural settlements or torrential catchments is recommended. Such an approach would enable the identification of areas with the highest erosion risk and allow for field-based validation of the empirical model and remote sensing methods. Due to its comprehensive spatial coverage, the results of this study are highly relevant for decision-makers at both local and national levels. In addition to local authorities and governmental institutions, these findings may also support various environmental sectors, including emergency management services, forestry, water management, spatial planning, and nature conservation. An effective approach to land management must be multidisciplinary, taking into account natural conditions alongside contemporary demographic and socio-economic processes. Declarations Credit authorship contribution statement Uroš Durlević: Conceptualization, software, formal analysis, data curation, writing—original draft preparation, project administration, funding acquisition. Tanja Srejić: Conceptualization, validation, writing—original draft preparation. Sanja Manojlović: Methodology, writing—review and editing, supervision. Marko V. Milošević: Software, writing—review and editing. Natalija Batoćanin: Investigation, visualization. Milica Dobrić: Investigation, data curation. Jelena Svetozarević: Validation, writing—original draft preparation. Velibor Ilić: Software, data curation, validation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper. Funding This research was funded by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Contract number 451-03-33/2026-03/200172 and 451-03-33/2026-03/200091). Data Availability Statement To obtain the data from this study, please contact the authors via email. References Akhmetova D (2025) Anthropogenic impact on soil and vegetation in Turkistan Region: Chemical composition and heavy metal contamination. J Geogr Inst Jovan Cvijic SASA 75(1):1–15. https://doi.org/10.2298/IJGI240426013A Aleksova B, Lukić T, Milevski I, Spalević V, Marković SB (2023) Modelling water erosion and mass movements (wet) using GIS-based multi-hazard susceptibility assessment approaches: A case study—Kratovska Reka catchment (North Macedonia). Atmosphere 14:1139. https://doi.org/10.3390/atmos14071139 Alewell C, Borrelli P, Meusburger K, Panagos P (2019) Using the USLE: Chances, challenges and limitations of soil erosion modelling. Int Soil Water Conserv Res 7:203–225. https://doi.org/10.1016/j.iswcr.2019.05.004 Altobelli F, Vargas R, Corti G, Dazzi C, Montanarella L, Monteleone A, Caon L, Piazza MG, Calzolari C, Munafò M, Benedetti A (2020) Improving soil and water conservation and ecosystem services by sustainable soil management practices: From a global to an Italian soil partnership. Ital J Agron 15:1765. https://doi.org/10.4081/ija.2020.1765 Andrade EM, Palácio HAQ, Souza IH, Leão RAO, Guerreiro MJ (2008) Land use effects in groundwater composition of an alluvial aquifer (Trussu River, Brazil) by multivariate techniques. Environ Res 106:170–177. https://doi.org/10.1016/j.envres.2007.10.008 Ascencio-Sanchez M, Padilla-Castro C, Riveros-Lizana C, Hermoza-Espezúa RM, Atalluz-Ganoza D, Solórzano-Acosta R (2025) Impacts of land use on soil erosion: RUSLE analysis in a sub-basin of the Peruvian Amazon (2016–2022). Geosciences 15:15. https://doi.org/10.3390/geosciences15010015 Bezak N, Borrelli P, Mikoš M, Jemec Auflič M, Panagos P (2024) Towards multi-model soil erosion modelling: An evaluation of the erosion potential method (EPM) for global soil erosion assessments. CATENA 234:107596. https://doi.org/10.1016/j.catena.2023.107596 Bezbradica L, Josimović B, Milijić S (2023) Impact of repurposing forest land on erosion and sediment production—Case study: Krupanj Municipality—Serbia. Forests 14:1127. https://doi.org/10.3390/f14061127 Blinkov I, Kostadinov S (2010) Applicability of various erosion risk assessment methods for engineering purposes. In: Proceedings of the BALWOIS 2010 Conference, Ohrid, Republic of Macedonia, 25–29 May 2010 Borrelli P, Robinson DA, Panagos P, Ballabio C et al (2020) Land use and climate change impacts on global soil erosion by water (2015–2070). Proc Natl Acad Sci U S A 117(36):21994–22001. https://doi.org/10.1073/pnas.2001403117 Braunović S, Cvetković J, Jovanović F, Polovina S, Šurjanac N, Stojanović V, Momirović N (2025) Assessment of soil erosion intensity using the erosion potential method: A case study of the Grdelica Gorge, Serbia. Sustain For 91:37–56. https://doi.org/10.5937/SustFor2591037 Ćalić J, Milošević MV, Milivojević M, Gaudenji T (2017) Relief of Serbia. In: Radovanović M (ed) Relief of Serbia. Geographical Institute Jovan Cvijić SASA, Belgrade, Serbia, pp 23–92. (In Serbian) Carević I, Sibinović M, Manojlović S, Batoćanin N, Petrović AS, Srejić T (2021) Geological approach for landfill site selection: A case study of Vršac Municipality, Serbia. Sustainability 13:7810. https://doi.org/10.3390/su13147810 Chen J, Li F, Fan Z, Wang Y (2016) Integrated application of multivariate statistical methods to source apportionment of watercourses in the Liao River Basin, Northeast China. Int J Environ Res Public Health 13:1035. https://doi.org/10.3390/ijerph13101035 Copernicus Land Monitoring Service (2023) High resolution layer: Tree cover and forests. https://land.copernicus.eu/en/products/high-resolution-layer-forests-and-tree-cover . Accessed 7 Mar 2026 Ćurić V, Durlević U, Ristić N, Novković I, Čegar N (2022) GIS application in analysis of threat of forest fires and landslides in the Svrljiški Timok basin (Serbia). Glasn Srp Geogr Drust 102(1):107–130. https://doi.org/10.2298/GSGD2201107C Dalir P, Naghdi R, Jafari S, Tsioras PA (2025) Comparative assessment of woody species for runoff and soil erosion control on forest road slopes in harvested sites of the Hyrcanian forests, Northern Iran. Forests 16:1013. https://doi.org/10.3390/f16061013 De Vente J, Poesen J (2005) Predicting soil erosion and sediment yield at the basin scale: Scale issues and semi-quantitative models. Earth-Sci Rev 71:95–125. https://doi.org/10.1016/j.earscirev.2005.02.002 Deđanski V, Durlević U, Kovjanić A, Lukić T (2024) GIS-based spatial modeling of landslide susceptibility using BWM-LSI: A case study—City of Smederevo (Serbia). Open Geosci 16:20220688. https://doi.org/10.1515/geo-2022-0688 Dildabek D, Aliaskarov D, Kaimuldinova K, Bakanov N, Yegizbayeva A, Shuangjie Z (2025) Projected land use land cover dynamics and tourism suitability assessment in the selected Kazakhstan area. J Geogr Inst Jovan Cvijic SASA 75(3):461–469. https://doi.org/10.2298/IJGI2503461D Dominici R, Larosa S, Viscomi A, Mao L, De Rosa R, Cianflone G (2020) Yield erosion sediment (YES): A PyQGIS plug-in for the sediments production calculation based on the erosion potential method. Geosciences 10:324. https://doi.org/10.3390/geosciences10080324 Đorđević M, Đokić M, Manić M, Vesković J, Dragović R, Smičiklas I, Dragović S, Onjia A (2026) Advanced GIS-based RUSLE modeling for soil erosion estimation in the Toplica River Basin, Serbia. Geosciences 16:83. https://doi.org/10.3390/geosciences16020083 Dragičević N, Karleuša B, Ožanić N (2018) Modification of erosion potential method using climate and land cover parameters. Geomat Nat Hazards Risk 9:1085–1105. https://doi.org/10.1080/19475705.2018.1496483 Durlević U, Ilić V, Valjarević A (2025) Wildfire susceptibility mapping using deep learning and machine learning models based on multi-sensor satellite data fusion: A case study of Serbia. Fire 8:407. https://doi.org/10.3390/fire8100407 Durlević U, Momčilović A, Ćirić V, Dragojević M (2019) GIS application in analysis of erosion intensity in the Vlasina River Basin. Glasn Srp Geogr Drust 99(2):17–36. https://doi.org/10.2298/GSGD1902017D Durlević U, Srejić T, Valjarević A, Aleksova B, Deđanski V, Vujović F, Lukić T (2025) GIS-based spatial modeling of soil erosion and wildfire susceptibility using VIIRS and Sentinel-2 data: A case study of Šar Mountains National Park, Serbia. Forests 16:484. https://doi.org/10.3390/f16030484 Durlević U, Valjarević A, Novković I, Vujović F, Josifov N, Krušić J, Komac B, Djekić T, Singh SK, Jović G et al (2024) Universal snow avalanche modeling index based on SAFI–Flow-R approach in poorly-gauged regions. ISPRS Int J Geo-Inf 13:315. https://doi.org/10.3390/ijgi13090315 Efthimiou N, Lykoudi E, Karavitis C (2017) Comparative analysis of sediment yield estimations using different empirical soil erosion models. Hydrol Sci J 62:2674–2694. https://doi.org/10.1080/02626667.2017.1404068 Efthimiou N, Lykoudi E, Panagoulia D, Karavitis C (2016) Assessment of soil susceptibility to erosion using the EPM and RUSLE models: The case of Venetikos River catchment. Glob Nest J 18:164–179. https://doi.org/10.30955/gnj.001847 Environmental Systems Research Institute (2024) Sentinel-2 10 m land use/land cover time series (2024). Esri Living Atlas. https://livingatlas.arcgis.com/landcoverexplorer/ . Accessed 9 Mar 2026 Estrada-Carmona N, Harper EB, DeClerck F, Fremier AK (2017) Quantifying model uncertainty to improve watershed-level ecosystem service quantification: A global sensitivity analysis of the RUSLE. Int J Biodivers Sci Ecosyst Serv Manag 13:40–50. https://doi.org/10.1080/21513732.2016.1237383 European Space Agency (2024) Copernicus global digital elevation model. Distributed by OpenTopography. https://doi.org/10.5069/G9028PQB . Accessed 2 Mar 2026 Fernández G, Merchán L, Sánchez JÁ (2025) Spatial representation of soil erosion and vegetation affected by a forest fire in the Sierra de Francia (Spain) using RUSLE and NDVI. Land 14:793. https://doi.org/10.3390/land14040793 Gajić Protić A, Krunić N, Protić B (2024) Detecting Serbia’s settlement patterns: A fuzzy logic-based approach to rural–urban area delimitation for spatial planning. Land 13:1981. https://doi.org/10.3390/land13121981 Gajić A, Krunić N, Protić B (2021) Classification of rural areas in Serbia: Framework and implications for spatial planning. Sustainability 13:1596. https://doi.org/10.3390/su13041596 Gavrilović S (1972) Inženjering o bujičnim tokovima i eroziji [Engineering of torrents and erosion]. J Constr (Spec Issue) :1–292 (In Serbian) GeoSrbija (2026) Geoportal of Serbia—Cartographic data viewer. https://a3.geosrbija.rs . Accessed 9 Mar 2026 Gocić M, Dragićević S, Radivojević A, Martić Bursać N, Stričević L, Đorđević M (2020) Changes in soil erosion intensity caused by land use and demographic changes in the Jablanica River Basin, Serbia. Agriculture 10:345. https://doi.org/10.3390/agriculture10080345 Golijanin J, Nikolić G, Valjarević A, Ivanović R, Tunguz V, Bojić S, Grmuša M, Lukić Tanović M, Perić M, Hrelja E, Stankov S (2022) Estimation of potential soil erosion reduction using GIS-based RUSLE under different land cover management models: A case study of Pale Municipality, B&H. Front Environ Sci 10:945789. https://doi.org/10.3389/fenvs.2022.945789 Gonzalez-Ollauri A, Mickovski SB (2020) The effect of willow (Salix sp.) on soil moisture and matric suction at a slope scale. Sustainability 12:9789. https://doi.org/10.3390/su12239789 Goodrich DC, Burns IS, Unkrich CL, Semmens DJ, Guertin DP, Hernandez M, Yatheendradas S, Kennedy JR, Levick LR (2012) KINEROS2/AGWA: Model use, calibration, and validation. Trans ASABE 55:1561–1574. https://doi.org/10.13031/2013.42264 Grčić M, Sibinović M, Ratkaj I (2024) The entropy as a parameter of demographic dynamics: Case study of the population of Serbia. Acta Geogr Slov 64:23–39. https://doi.org/10.3986/AGS.11441 Kaffas K, Hrissanthou V (2018) Soil erosion, streambed deposition and streambed erosion—Assessment at the mountainous terrain. Proc 2:626. https://doi.org/10.3390/proceedings2110626 Kosmas C, Gerontidis St, Marathianou M (2000) The effect of land use change on soils and vegetation over various lithological formations on Lesvos (Greece). CATENA 40:51–68. https://doi.org/10.1016/S0341-8162(99)00064-8 Kostadinov S, Braunović S, Dragićević S, Zlatić M, Dragović N, Rakonjac N (2018) Effects of erosion control works: Case study—Grdelica Gorge, the South Morava River (Serbia). Water 10:1094. https://doi.org/10.3390/w10081094 Kovačević J, Cvijetinović Ž, Lakušić D, Kuzmanović N, Šinžar-Sekulić J, Mitrović M, Stančić N, Brodić N, Mihajlović D (2020) Spatio-temporal classification framework for mapping woody vegetation from multi-temporal Sentinel-2 imagery. Remote Sens 12:2845. https://doi.org/10.3390/rs12172845 Lazarević R (1983) Soil erosion map of SR Serbia 1:500,000–Interpretation. Institute for Forestry and Wood Industry, Belgrade, Serbia. (In Serbian) Lazarević R (1985) A new method for determining the erosion coefficient (Z). Erozija – Stručno-informativni bilten 13:53–61 (In Serbian) Lovrić N, Tošić R (2018) Assessment of soil erosion and sediment yield using erosion potential method: Case study—Vrbas River Basin (B&H). Glasn Srp Geogr Drust 98(1):1–14. https://doi.org/10.2298/GSGD180215002L Lukić S, Baumgertel A, Obradović S, Kadović R, Beloica J, Pantić D, Belanović Simić S (2022) Assessment of land sensitivity to degradation using the MEDALUS model—A case study of Grdelica Gorge and Vranjska Valley (Southeastern Serbia). iForest 15:163–170. https://doi.org/10.3832/ifor3871-015 Malušević I, Ristić R, Radić B, Polovina S, Milčanović V, Nešković P (2025) A historical overview of methods for the estimation of erosion processes on the territory of the Republic of Serbia. Land 14:405. https://doi.org/10.3390/land14020405 Manojlović S, Antić M, Šantić D, Sibinović M, Carević I, Srejić T (2018) Anthropogenic impact on erosion intensity: Case study of rural areas of Pirot and Dimitrovgrad municipalities. Serbia Sustain 10:826. https://doi.org/10.3390/su10030826 Manojlović S, Sibinović M, Srejić T, Novković I, Milošević MV, Gatarić D, Carević I, Batoćanin N (2022) Factors controlling the change of soil erosion intensity in mountain watersheds in Serbia. Front Environ Sci 10:888901. https://doi.org/10.3389/fenvs.2022.888901 Martinović M, Ratkaj I (2015) Sustainable rural development in Serbia: Towards a quantitative typology of rural areas. Carpath J Earth Environ Sci 10:37–48 Maruna M, Crnčević T, Milojević MP (2019) The institutional structure of land use planning for urban forest protection in the post-socialist transition environment: Serbian experiences. Forests 10:560. https://doi.org/10.3390/f10070560 Milevski I, Aleksova B, Lukić T, Dragićević S, Valjarević A (2024) Multi-hazard modeling of erosion and landslide susceptibility at the national scale in the example of North Macedonia. Open Geosci 16:20220718. https://doi.org/10.1515/geo-2022-0718 Milovanović B, Ducić V, Radovanović M, Milivojević M (2017) Climate regionalization of Serbia according to the Köppen climate classification. J Geogr Inst Jovan Cvijic SASA 67(2):103–114. https://doi.org/10.2298/IJGI1702103M Milovanović B, Schuster P, Radovanović M, Ristić Vakanjac V, Schneider C, Milivojević M (2018) Spatial–temporal variability of air temperatures in Serbia in the period 1961–2010. J Geogr Inst Jovan Cvijic SASA 68(2):157–175. https://doi.org/10.2298/IJGI1802157M Ministry of Mining and Energy of the Republic of Serbia (2026) GeoLISS: Geological information system of Serbia—Geological map viewer. https://geoliss.mre.gov.rs/vebkarte/geo300.html . Accessed 8 Mar 2026 Nicosia A, Carollo FG, Di Stefano C, Palmeri V, Pampalone V, Serio MA, Bagarello V, Ferro V (2024) The importance of measuring soil erosion by water at the field scale: A review. Water 16:3427. https://doi.org/10.3390/w16233427 OpenStreetMap Team (2026) OpenStreetMap export. https://www.openstreetmap.org/#map=7/44.226/15.562 . Accessed 13 Feb 2026 Oro V, Stanisavljevic R, Nikolic B, Tabakovic M, Secanski M, Tosi S (2021) Diversity of mycobiota associated with the cereal cyst nematode Heterodera filipjevi originating from some localities of the Pannonian Plain in Serbia. Biology 10:283. https://doi.org/10.3390/biology10040283 Oshunsanya SO, Yu H, Odebode AM, Edem ID, Oluwatuyi TS, Imasuen EE, Odeyinka DE (2025) Comparative assessment of fractional and erosion plot methods for quantifying soil erosion and nutrient loss under vetiver grass technology on two contrasting slopes in rainforest agroecology. Agriculture 15:1762. https://doi.org/10.3390/agriculture15161762 Panagos P, Borrelli P, Poesen J, Ballabio C, Lugato E, Meusburger K, Montanarella L, Alewell C (2015) The new assessment of soil loss by water erosion in Europe. Environ Sci Policy 54:438–447. https://doi.org/10.1016/j.envsci.2015.08.012 Panagos P, Ballabio C, Poesen J, Lugato E, Scarpa S, Montanarella L, Borrelli P (2020) A soil erosion indicator for supporting agricultural, environmental and climate policies in the European Union. Remote Sens 12:1365. https://doi.org/10.3390/rs12091365 Panagos P, Borrelli P, Matthews F, Liakos L, Bezak N, Diodato N, Ballabio C (2022) Global rainfall erosivity projections for 2050 and 2070. J Hydrol 610:127865. https://doi.org/10.1016/j.jhydrol.2022.127865 Pantić M, Maričić T, Milijić S (2024) Visualising the relevance of climate change for spatial planning by the example of Serbia. Appl Sci 14:1530. https://doi.org/10.3390/app14041530 Papageorgiou N, Hadjimitsis D, Danezis C, Lasaponara R (2025) Assessment of soil erosion risk in cultural heritage sites: A bibliometric analysis. Heritage 8:307. https://doi.org/10.3390/heritage8080307 Perovic V, Jaramaz D, Zivotić Lj, Cakmak D, Mrvic V, Milanovic M, Saljnikov E (2016) Design and implementation of WebGIS technologies in evaluation of erosion intensity in the municipality of Niš (Serbia). Environ Earth Sci 75:211. https://doi.org/10.1007/s12665-015-4857-x Petrović AM, Manojlović S, Srejić T, Zlatanović N (2024) Insights into land-use and demographical changes: Runoff and erosion modifications in the highlands of Serbia. Land 13:1342. https://doi.org/10.3390/land13091342 Pinto CC, Calazans GM, Oliveira SC (2019) Assessment of spatial variations in the surface water quality of the Velhas River Basin, Brazil, using multivariate statistical analysis and nonparametric statistics. Environ Monit Assess 191:164. https://doi.org/10.1007/s10661-019-7281-y Polovina S, Radić B, Ristić R, Milčanović V (2024) Application of remote sensing for identifying soil erosion processes on a regional scale: An innovative approach to enhance the erosion potential model. Remote Sens 16:2390. https://doi.org/10.3390/rs16132390 Prăvălie R, Borrelli P, Panagos P, Ballabio C, Lugato E, Chappell A, Miguez-Macho G, Maggi F, Peng J, Niculiță M, Roșca B, Patriche C, Dumitrașcu M, Bandoc G, Niță I-A, Bîrsan M-V (2024) A unifying modelling of multiple land degradation pathways in Europe. Nat Commun 15:3862. https://doi.org/10.1038/s41467-024-48252-x QGIS Development Team (2026) QGIS geographic information system v3.40.09 with GRASS [software]. Open Source Geospatial Foundation Project. http://qgis.osgeo.org . Accessed 30 Jun 2025 Rao AU, Sabhahit N, Ananda LU, Bhandary RP (2026) Role of soil erosion in instability of slopes along coastal Karnataka. Geotechnics 6:21. https://doi.org/10.3390/geotechnics6010021 Ratknić M, Braunović S, Rakonjac Lj, Ratknić T (2025) The afforestation strategy of the Republic of Serbia. Belgrade, Serbia Raux J, Copard Y, Laignel B, Fournier M, Massei N (2011) Classification of worldwide drainage basins through the multivariate analysis of variables controlling their hydrosedimentary response. Glob Planet Change 76:117–127. https://doi.org/10.1016/j.gloplacha.2010.12.005 Raza A, Ahrends H, Habib-Ur-Rahman M, Gaiser T (2021) Modeling approaches to assess soil erosion by water at the field scale with special emphasis on heterogeneity of soils and crops. Land 10:422. https://doi.org/10.3390/land10040422 Ristić R, Kostadinov S, Abolmasov B, Dragićević S, Trivan G, Radić B, Trifunović M, Radosavljević Z (2012) Torrential floods and town and country planning in Serbia. Nat Hazards Earth Syst Sci 12:23–35. https://doi.org/10.5194/nhess-12-23-2012 Smanov Z, Duisenbayev S, Zulpykharov K, Laiskhanov S, Turymtayev Z, Kozhayev Z, Taukebayev O (2025) Soil salinization and its impact on the degradation of agricultural landscapes of the Talas District, Kazakhstan. J Geogr Inst Jovan Cvijic SASA 75(2):233–250. https://doi.org/10.2298/IJGI2502233S Spalevic V, Barovic G, Vujacic D, Curovic M, Behzadfar M, Djurovic N, Dudic B, Billi P (2020) The impact of land use changes on soil erosion in the River Basin of Miocki Potok, Montenegro. Water 12:2973. https://doi.org/10.3390/w12112973 Srejić T, Manojlović S, Sibinović M, Bajat B, Novković I, Milošević MV, Carević I, Todosijević M, Sedlak MG (2023) Agricultural land use changes as a driving force of soil erosion in the Velika Morava River Basin, Serbia. Agriculture 13:778. https://doi.org/10.3390/agriculture13040778 Srejić T, Manojlović S, Sibinović M, Lukić T, Kričković E, Durlević U (2025) Impact of the agri-geographical transformation of rural settlements on the geospatial dynamics of soil erosion intensity in municipalities of Central Serbia. Open Geosci 17:20250857. https://doi.org/10.1515/geo-2025-0857 Stanišić M, Lovrić M, Nedeljković J, Nonić D, Pezdevšek Malovrh Š (2021) Climate change governance in forestry and nature conservation in selected forest regions in Serbia: Stakeholders classification and collaboration. Forests 12:709. https://doi.org/10.3390/f12060709 Stefanidis S, Stathis D (2018) Effect of climate change on soil erosion in a mountainous Mediterranean catchment (Central Pindus, Greece). Water 10:1469. https://doi.org/10.3390/w10101469 Stefanidis SP, Proutsos ND, Tigkas D, Chatzichristaki C (2025) Erosion-based classification of mountainous watersheds in Greece: A geospatial approach. Sustainability 17:8710. https://doi.org/10.3390/su17198710 Stojković S, Marković D, Durlević U (2023) Snow cover estimation using Sentinel-2 high spatial resolution data: A case study—National Park Šar Planina (Serbia). In: Ademović N, Mujčić E, Mulić M, Kevrić J, Akšamija Z (eds) Advanced Technologies, Systems, and Applications VII. Lecture Notes in Networks and Systems, vol 539. Springer, Cham, Switzerland, pp 389–399. https://doi.org/10.1007/978-3-031-17697-5_39 Sukmawijaya A, Akber MA, Wang Z, Azizan FA, Bell M, Abdul Aziz A (2026) Spatial assessment of water balance and soil erosion under land-use change in Chieng Hac, Northern Vietnam. Remote Sens 18:998. https://doi.org/10.3390/rs18070998 Tölgyesi C, Bátori Z, Erdős L (2014) Using statistical tests on relative ecological indicator values to compare vegetation units—Different approaches and weighting methods. Ecol Indic 36:441–446. https://doi.org/10.1016/j.ecolind.2013.09.002 Tošić I, da Silva ASA, Filipović L, Tošić M, Lazić I, Putniković S, Stosic T, Stosic B, Djurdjević V (2025a) Trends of extreme precipitation events in Serbia under global warming. Atmosphere 16:436. https://doi.org/10.3390/atmos16040436 Tošić R, Dragićević S, Lovrić N (2012) Assessment of soil erosion and sediment yield changes using the erosion potential model. Carpath J Earth Environ Sci 7:147–154 Tošić R, Dragićević S, Lovrić N, Langović M (2025b) Mapping soil erosion intensity using erosion potential method (EPM): Case study—Bosnia and Herzegovina. Carpath J Earth Environ Sci 58:413–426. https://doi.org/10.26471/cjees/2025/020/343 Trendafilov B, Minchev I, Trendafilov A, Blinkov I (2024) Comparison of EPM with RUSLE for soil erosion modeling in the Strumica River Basin. Geogr Environ Sustain 17:44–49. https://doi.org/10.24057/2071-9388-2024-0580 United Nations Convention to Combat Desertification (2022) Global land outlook, 2nd edn. Land restoration for recovery and resilience. UNCCD, Bonn, Germany Urošev M, Kovačević-Majkić Š, Štrbac D, Milanović Pešić A, Milijašević D, Jakovljević D, Petrović A (2017) Waters of Serbia. In: Radovanović M (ed) Relief of Serbia. Geographical Institute Jovan Cvijić SASA, Belgrade, Serbia, pp 161–234. (In Serbian) Wang J, Wang X, Yan Y, Wang L, Hu H, Ma B, Zhou H, Liu J, Gan F, Fan Y (2025) Peak soil erosion risk in mixed forests: A critical transition phase driven by Moso bamboo expansion. Agriculture 15:1772. https://doi.org/10.3390/agriculture15161772 Wani AA (2024) Comprehensive analysis of clustering algorithms: Exploring limitations and innovative solutions. PeerJ Comput Sci 10:e2286. https://doi.org/10.7717/peerj-cs.2286 Xu D, Tian Y (2015) A comprehensive survey of clustering algorithms. Ann Data Sci 2:165–193. https://doi.org/10.1007/s40745-015-0040-1 Yang A, Zhang L, Zhang S, Zhan Z, Shi J (2022) Research on the temporal and spatial characteristics, spatial clustering and governance strategies of carbon emissions in cities of Shandong. Front Environ Sci 10:1024122. https://doi.org/10.3389/fenvs.2022.1024122 Životić L, Vuković Vimić A (2022) Soil degradation and climate change in Serbia. Belgrade, Serbia Additional Declarations The authors declare no competing interests. Supplementary Files floatimage1.png Graphical abstract descriptions: The graphical abstract presents the complete workflow and key findings of the national-scale soil erosion susceptibility assessment in Serbia, conducted using the Erosion Potential Model (EPM), Geographic Information Systems (GIS), and Sentinel-2 remote sensing data. The process begins with integrating major geospatial inputs, including the Digital Elevation Model (DEM), the ESRI land-use database, and geological maps, into a unified GIS database. Through GIS processing, the main EPM factors were derived: soil resistance coefficient, soil protection coefficient, erosion type and extent coefficient, and slope coefficient. These parameters were used to calculate the erosion intensity using the Z coefficient and to generate a spatial distribution map of soil erosion across Serbia. The results show that weak and very weak erosion predominate across most of the territory, while intensive and excessive erosion is concentrated in specific municipalities. Validation was performed using soil erosion inventory data, confusion matrix, and ROC-AUC analysis, confirming very high model reliability. Additionally, Agglomerative Hierarchical Clustering (AHC) was applied to classify municipalities into six clusters based on erosion characteristics, forest cover, agricultural land, altitude, and slope. This graphical summary highlights the study's practical importance for sustainable land management, forest protection, and planning to reduce erosion risk. 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-9569900","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":632017278,"identity":"3f817486-eabb-4501-8137-0ce278a7f08d","order_by":0,"name":"Uroš Durlević","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/klEQVRIiWNgGAWjYHACAwaGAgYGNgbmAwwJxGsxAGlhSyBRCwMDjwGR6g8wb5OuMLBJ7GM/8/nFA4Y6eXOJBMYPPxju2ePWwlYmecYgLbGNJ3ebRQLDYcOdMxKYJXsYihMbcGrhMZNsMDhszCbBu80ggeEA44YbCQzSDAwJOP0F1fIfqIXnGVBLnT1QC/NvoBY8DgNrOSAH1ML8IIGBORGohQ1kCyMuh0keZiu2bDBIlmPjSTNjSDA4nLzhzMM2yx6DBJx+4TvevPFmQ4Udj3z74ccff1TU2W44nnz4xo8K3A5jYEYw2SQgEQRyEpFxxPyBOHWjYBSMglEw0gAALHRNCaBNUuAAAAAASUVORK5CYII=","orcid":"","institution":"Geographical Institute “Jovan Cvijić”, Serbian Academy of Sciences and Arts","correspondingAuthor":true,"prefix":"","firstName":"Uroš","middleName":"","lastName":"Durlević","suffix":""},{"id":632017819,"identity":"3ce1df96-b6ef-47ac-bd18-78491d92b6ca","order_by":1,"name":"Tanja Srejić","email":"","orcid":"","institution":"University of Belgrade, Faculty of Geography","correspondingAuthor":false,"prefix":"","firstName":"Tanja","middleName":"","lastName":"Srejić","suffix":""},{"id":632017820,"identity":"d8dd5694-00f6-47b9-98d2-81e7577713e1","order_by":2,"name":"Sanja Manojlović","email":"","orcid":"","institution":"University of Belgrade, Faculty of Geography","correspondingAuthor":false,"prefix":"","firstName":"Sanja","middleName":"","lastName":"Manojlović","suffix":""},{"id":632017821,"identity":"488cae86-23a6-4abf-af90-0e3dcc6d9745","order_by":3,"name":"Marko V. Milošević","email":"","orcid":"","institution":"Geographical Institute “Jovan Cvijić”, Serbian Academy of Sciences and Arts","correspondingAuthor":false,"prefix":"","firstName":"Marko","middleName":"V.","lastName":"Milošević","suffix":""},{"id":632017822,"identity":"6d494b73-b31e-4591-8663-77108d3327c3","order_by":4,"name":"Natalija Batoćanin","email":"","orcid":"","institution":"University of Belgrade, Faculty of Geography","correspondingAuthor":false,"prefix":"","firstName":"Natalija","middleName":"","lastName":"Batoćanin","suffix":""},{"id":632017823,"identity":"845c46ba-07de-43ab-bac1-0f5405a71ca0","order_by":5,"name":"Milica Dobrić","email":"","orcid":"","institution":"University of Belgrade, Faculty of Geography","correspondingAuthor":false,"prefix":"","firstName":"Milica","middleName":"","lastName":"Dobrić","suffix":""},{"id":632017824,"identity":"be8df4a9-34ed-4a91-bc61-e60a293e3bd4","order_by":6,"name":"Jelena Svetozarević","email":"","orcid":"","institution":"Climate, Atmosphere and Water Research Institute, Bulgarian Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Jelena","middleName":"","lastName":"Svetozarević","suffix":""},{"id":632017825,"identity":"35c862e5-4a43-4016-ac4a-6c9dbaf420e4","order_by":7,"name":"Velibor Ilić","email":"","orcid":"","institution":"Institute for Artificial Intelligence Research and Development of Serbia","correspondingAuthor":false,"prefix":"","firstName":"Velibor","middleName":"","lastName":"Ilić","suffix":""}],"badges":[],"createdAt":"2026-04-29 21:12:31","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-9569900/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9569900/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108636080,"identity":"be10d42f-1e02-40c4-9954-45da4ecc1067","added_by":"auto","created_at":"2026-05-06 18:00:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":8351705,"visible":true,"origin":"","legend":"\u003cp\u003eGeographical location of Serbia.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/13b70ffbe30c6657b36df926.png"},{"id":108805502,"identity":"89e894c2-1309-4715-b0b0-9c1faa8cf271","added_by":"auto","created_at":"2026-05-08 15:26:07","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":27970965,"visible":true,"origin":"","legend":"\u003cp\u003eThematic maps of the applied parameters: soil resistance coefficient (a), vegetation protection coefficient (b), erosion type and extent coefficient (c), and terrain slope coefficient (d).\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/3e74a02fcdf8ee58b7451668.jpeg"},{"id":108805676,"identity":"767afa8a-47c5-4124-9c4e-c885386e9ef4","added_by":"auto","created_at":"2026-05-08 15:26:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":148651,"visible":true,"origin":"","legend":"\u003cp\u003eSoil erosion inventory\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/14a7b62f969e477e45aaf2b3.png"},{"id":108804968,"identity":"3f84b455-9dcc-4fcb-9873-4589c489d1f9","added_by":"auto","created_at":"2026-05-08 15:24:23","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":320930,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart with all the procedures and methods used in this research.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/60f3723470c8cd221269f737.png"},{"id":108636085,"identity":"0ffcfc15-d842-4f71-b6c1-a004d37df6a5","added_by":"auto","created_at":"2026-05-06 18:00:27","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":17986143,"visible":true,"origin":"","legend":"\u003cp\u003eMap of Mechanical Soil Erosion in Serbia.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/5c04d9decf3b7cc9a4736082.png"},{"id":108805026,"identity":"1d60881c-f647-40cc-80e7-4ad01685f223","added_by":"auto","created_at":"2026-05-08 15:24:33","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":11048268,"visible":true,"origin":"","legend":"\u003cp\u003eThe spatial differentiation of the municipalities using AHC in Serbia.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/e0b04ba34020d92042c99eba.png"},{"id":108636087,"identity":"b51783aa-5ed0-4422-beaf-97417bb24881","added_by":"auto","created_at":"2026-05-06 18:00:28","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":4154329,"visible":true,"origin":"","legend":"\u003cp\u003eBox plots of selected indicators across six clusters.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/11bb2f25bc725477319914e0.png"},{"id":108805276,"identity":"091e2199-225c-4a9b-a897-770514de9ed4","added_by":"auto","created_at":"2026-05-08 15:25:26","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":36987,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of Z values for erosion and non-erosion samples. The two classes exhibit a complete separation, with no overlap in Z values.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/798d272804bfd9b9dc59aa2b.png"},{"id":108636089,"identity":"cd17944b-3ddc-4f5c-8a1d-cf2c63c6bb32","added_by":"auto","created_at":"2026-05-06 18:00:28","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":37989,"visible":true,"origin":"","legend":"\u003cp\u003eBoxplot of Z values by class, showing a clear distinction between erosion and non-erosion samples and confirming the absence of overlapping values.\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/084a3bfe3c2ce6a268f2e997.png"},{"id":108805277,"identity":"8bdbe958-ca16-4d5f-9e94-bbbd0cac930d","added_by":"auto","created_at":"2026-05-08 15:25:26","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":50179,"visible":true,"origin":"","legend":"\u003cp\u003eROC curve of the EPM model, demonstrating perfect classification performance (AUC = 1.00).\u003c/p\u003e\n\u003cp\u003eTherefore, the validation results confirm that the EPM model provides a robust and internally consistent representation of soil erosion susceptibility under clearly defined environmental conditions.\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/d212f8c0de4fef0d2f8f9c2e.png"},{"id":108812191,"identity":"89cd556a-e7fc-4b52-926e-3c95c8cc0e69","added_by":"auto","created_at":"2026-05-08 16:09:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":70675752,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/8262cb17-9401-47bf-8e4f-c774a5110171.pdf"},{"id":108636081,"identity":"590788cc-008c-4ccb-8ac2-44863328dd41","added_by":"auto","created_at":"2026-05-06 18:00:27","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":503491,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGraphical abstract descriptions\u003c/strong\u003e: The graphical abstract presents the complete workflow and key findings of the national-scale soil erosion susceptibility assessment in Serbia, conducted using the Erosion Potential Model (EPM), Geographic Information Systems (GIS), and Sentinel-2 remote sensing data. The process begins with integrating major geospatial inputs, including the Digital Elevation Model (DEM), the ESRI land-use database, and geological maps, into a unified GIS database. Through GIS processing, the main EPM factors were derived: soil resistance coefficient, soil protection coefficient, erosion type and extent coefficient, and slope coefficient. These parameters were used to calculate the erosion intensity using the Z coefficient and to generate a spatial distribution map of soil erosion across Serbia. The results show that weak and very weak erosion predominate across most of the territory, while intensive and excessive erosion is concentrated in specific municipalities. Validation was performed using soil erosion inventory data, confusion matrix, and ROC-AUC analysis, confirming very high model reliability. Additionally, Agglomerative Hierarchical Clustering (AHC) was applied to classify municipalities into six clusters based on erosion characteristics, forest cover, agricultural land, altitude, and slope. This graphical summary highlights the study's practical importance for sustainable land management, forest protection, and planning to reduce erosion risk.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9569900/v1/d91054f96b31731b27be9550.png"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eGIS-Based Soil Erosion Susceptibility Mapping in Serbia Using the EPM Model and Satellite Remote Sensing: A National-Scale Prediction\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1.Introduction","content":"\u003cp\u003eSoil is considered a non-renewable resource, as soil erosion rates are significantly higher than the rate of its natural formation, and its loss and degradation cannot be compensated within the timescale of human life. Soil erosion is a natural geomorphological process that occurs continuously on Earth\u0026rsquo;s surface (Lovrić et al., 2018). However, in terms of the area affected, it has been recognized as a global problem (Durlević et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Soil erosion results from multiple interacting natural and anthropogenic processes. Its intensity is largely controlled by the combined effects of relief characteristics, lithology, climatic conditions, and land-use practices (Panagos et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Srejić et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Malušević et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Erosion is often a consequence of long-term deforestation and improper land use, leading to numerous adverse environmental impacts and significant economic losses (Milevski et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Modeling results indicate that approximately 25% of land in the European Union experiences erosion rates exceeding the recommended sustainable threshold (2 t/ha/year). In comparison, more than 6% of agricultural land is affected by severe erosion (11 t/ha/year) (Panagos et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Furthermore, soil erosion poses a serious threat to sustainable agriculture, with studies indicating a global decline in crop yields of approximately 0.4% per year due to erosion (Altobelli et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In this context, soil erosion, as the primary driver of land degradation, represents a major challenge to ecosystem sustainability (Wang et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSoil erosion is one of the most significant environmental challenges for sustainable natural resource management and the quality of life of populations across Europe (Prăvălie et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Human activities and land-use changes, including intensive agriculture, deforestation, urbanization, and industrialization, have been identified as the primary drivers of accelerated erosion (Borrelli et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Projections up to 2070 indicate that future erosion intensity will largely depend on the combined effects of climate change and land-use change, with different socio-economic scenarios and agricultural practices significantly modulating these processes (Borrelli et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Contemporary soil management practices in agriculture have considerable potential to mitigate the negative effects of land degradation (Kaffas et al., 2018). Under the influence of climate change, precipitation erosivity forecasts indicate increases of approximately 30\u0026ndash;66% in soil loss by 2070, highlighting the critical role of climate in the future dynamics of this process (Panagos et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Životić et al., 2022; Nicosia et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). At the same time, the United Nations Convention to Combat Desertification (UNCCD) warns that, without urgent conservation measures, land degradation will continue to threaten the quality of life of billions of people worldwide (United Nations Convention to Combat Desertification, \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe scientific community has been addressing the challenges posed by soil erosion for several decades, at global, regional, and local scales, through the application of various types of empirical models for estimating soil erosion rates (Goodrich et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Alewell et al, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The development of the first soil erosion map of the Republic of Serbia began in 1966 and was published in 1983 (Lazarević et al., 1983). The Erosion Potential Method (EPM), developed by Gavrilović (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1972\u003c/span\u003e) and later modified by Lazarević (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e1985\u003c/span\u003e) and Tošić \u0026amp; Dragićević (2012), remains one of the most frequently applied models in regional studies. Previous studies have confirmed the scientific validity of the EPM model (De Vente et al., 2005). Over the past two decades, the development of geospatial databases and the application of Geographic Information Systems (GIS) technology have significantly enhanced the implementation of empirical models, including the EPM model (Spalević et al., 2020; Polovina et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn general, monitoring changes in soil erosion intensity is conducted on a regular basis. Addressing soil erosion requires continuous monitoring of processes, forms, and parameters across spatial and temporal scales, as changes in these factors directly influence the spatial patterns and intensity of erosion (Durlević et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Assessing erosion rates and identifying the most susceptible areas are essential for selecting appropriate soil management and protection strategies (Papageorgiou et al, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Rao et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). Given that the evaluation of soil erosion intensity represents the fundamental basis for planning prevention and protection measures, as well as for determining the type and extent of anti-erosion works, the main objective of this study is to produce a national-scale map of mechanical soil erosion intensity. The map is generated using contemporary high-resolution spatial and satellite data to identify the most susceptible municipalities and analyze their geospatial distribution through cluster classification based on dominant indicators. The spatial differentiation of results at the regional and local levels will enable more precise land management planning. The findings of this study will have significant practical applications, particularly for local governments, forest management services, and local communities, contributing to more efficient management of land and natural resources.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study area\u003c/h2\u003e \u003cp\u003eSerbia is positioned in the northern temperate light-heat zone between 41\u0026deg;51' and 46\u0026deg;11' north latitude. It is a continental country whose territory is 84 km from the Adriatic, 202 km from the Aegean, and 405 km from the Black Sea. In terms of relief, three macromorphological units stand out in Serbia: plains, hilly areas, and mountainous regions (Ćalić et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The plains (Pannonian and Moesian) are flat areas formed by block subsidence and sediment accumulation, with active fluvial and aeolian processes (dominated by Quaternary sediments). The Pannonian Plain covers northern Serbia, while the Moesian Plain extends to the far northeast. The hilly areas (peri-Pannonian and peri-Moesian) represent transitional zones between plains and mountains, with rougher relief and dominant slope processes and erosion (Oro et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Lithologically, they consist of Neogene sediments (gravels, sands, clays, marls). The mountainous regions include the Dinarides, the mountains of the Vardar zone, the Serbo-Macedonian massif, and the Carpatho-Balkanides. Often, tectonic depressions (basins) with flat bottoms are found between mountains, formed by block subsidence and sediment accumulation (Carević et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Together, these three units form the basic morphostructural division of Serbia. The territory lies within an elevation range between 28 m and 2,660 m (Velika Rudoka) (Durlević et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDepending on altitude, annual precipitation in Serbia ranges from 500\u0026ndash;600 mm to over 1100 mm. June precipitation averages between 60 and 140 mm, while February precipitation ranges from 30 to 100 mm (Stanišić et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Above 1,800 meters, the mean annual air temperature is around 3\u0026deg;C. In southern Serbia and Belgrade, it exceeds 12\u0026deg;C (Pantić et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). A statistically significant increase in temperature has been recorded in most of the country, with only a slight decrease observed in the far southeast. These results indicate a clear warming trend across almost the entire territory of Serbia (Milovanović et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe most widespread climate type in Serbia is Cfbq, which gradually changes into Cfbr in parts of eastern Serbia. The least represented is Efcr, present only on high mountains such as Prokletije and Šar Mountains (Stojković et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The Cfaq climate occurs in lowland areas of central and northern Serbia (Vojvodina), at elevations up to about 120\u0026ndash;140 meters. In the lowlands of central Serbia, the Dfbq climate prevails. Mountainous areas above 1200 meters in southwestern, southern, and southeastern Serbia are characterized by the Dfcq climate. At the highest elevations, around 2250 meters in Prokletije and Šar Mountains, the Dfcr climate is present (Milovanović et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Under the influence of numerous climatic, geological, geomorphological, biological, and anthropogenic factors, Serbia\u0026rsquo;s hydrographic network has been formed. The average river network density is 323 m/km\u0026sup2; (Urošev et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn areas where annual precipitation exceeds 1000 mm (Prokletije, Šar Mountains), river network density goes beyond 2000 m/km\u0026sup2;. Conversely, the eastern and northeastern parts of the country receive less than 600 mm of precipitation, resulting in sparser, often intermittent watercourses (Tošić et al., \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e). This is especially true for karst areas (about 10% of the territory). Hydrological regimes differ: in the Black Sea basin, the continental regime dominates with maximum flow in spring, while in the Adriatic basin, a modified maritime regime is expressed, with peak precipitation in late autumn and early winter (Urošev et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The largest amounts of water are formed on magmatic and metamorphic rocks. Altitude and slope also influence the number of watercourses, while water quality improves with elevation due to reduced anthropogenic pressure. In terms of vegetation zones, northern Serbia belongs to the steppe zone, while the central and southern parts are predominantly covered by deciduous forests (oak and beech) (Maruna et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Kovačević et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In the far south, within smaller areas, Mediterranean forests are also present.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Data Set and Methodology\u003c/h2\u003e \u003cp\u003eTo calculate erosion intensity and related natural and anthropogenic factors, numerous spatial datasets were used. The vector state border of Serbia used for the preparation of Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e was obtained from the GeoSerbia geoportal (GeoSrbija, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). The Digital Elevation Model (DEM) with a 30 m spatial resolution was generated from the Copernicus Global Digital Elevation Model platform (European Space Agency, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Larger cities and major rivers were digitized using data from OpenStreetMap (OpenStreetMap Team, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2026\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor obtaining land-use information, spatial data from two European geospatial databases were used. The base layer was derived from Sentinel-2 imagery from 2024 obtained from the Environmental Systems Research Institute (ESRI) database, while data from the Copernicus Land Monitoring Service were integrated to extract open forests and closed (dense) forests (Copernicus Land Monitoring Service, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Environmental Systems Research Institute, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Using this approach, eight land-use classes were obtained: water bodies, open forests, closed (dense) forests, seasonally flooded areas, agricultural plots, settlements, bare land, and rangeland (grasslands and pastures). The land-use dataset was resampled from 10 m to 30 m to ensure consistency with the other input parameters used in the analysis. The resampling was performed using the nearest-neighbor method, which preserves the original categorical values of land-use classes and prevents the creation of mixed or interpolated class values that may occur with other resampling techniques.\u003c/p\u003e \u003cp\u003eData on geology, i.e., rock types in Serbia, were obtained by digitizing the lithological content from the Geological Map at a scale of 1:300,000 (Ministry of Mining and Energy of the Republic of Serbia, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). The rocks were classified according to their resistance to mechanical soil erosion. Subsequently, the data were converted from vector to raster format in QGIS v3.40.9, with a pixel resolution of 30 m (QGIS Development Team, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). The terrain slope was derived from the Copernicus Global Digital Elevation Model geoportal, with a spatial resolution of 30 m (European Space Agency, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). For the coefficient of erosion type and extent, land-use and terrain slope data were integrated to provide a more objective representation. Both parameters were calibrated into a single layer, using data from 2024. The spatial resolution of all input parameters was standardized to 30 m in QGIS software.\u003c/p\u003e \u003cp\u003eFor the purposes of this study, the intensity of the erosion process was calculated using the Erosion Potential Model (EPM), also known as the Gavrilović method (Gavrilović, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1972\u003c/span\u003e). Results of relevant research in the field of erosion modeling have shown that, among 11 analyzed semi-quantitative methods, the EPM was rated as the most quantitative (De Vente \u0026amp; Poesen, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). The chosen research approach is aligned with the key geographical (spatial) dimension of the problem, as the application of the erosion model provides the most reliable results under the conditions for which it was developed. Validation of the model based on sediment yield data from over 100 watersheds worldwide has shown that the best agreement of results is observed in continental and temperate regions (Bezak et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Furthermore, its application in combination with other models has provided a satisfactory level of accuracy and reliability of results (Efthimiou et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Raza et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Trendafilov et al., \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Intensive application of the model at the local, national, and regional levels in recent decades has confirmed its scientific and practical value (Gocić et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Aleksova et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Stefanidis et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The simple structure of the model and its low input data requirements make the EPM suitable for regional analyses and planning of soil conservation measures (Dildabek et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Smanov et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Akhmetova, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The use of modern remote sensing methods and good integration with GIS tools has allowed the previous shortcomings of the model to be overcome (Dominici et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Ćurić et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Durlević et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This has significantly reduced the subjectivity in the assessment of individual parameters (Oshunsanya et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The erosion coefficient (Z) is calculated as follows:\u003c/p\u003e \u003cp\u003e \u003cem\u003eZ\u0026thinsp;=\u0026thinsp;Y \u0026middot; X \u0026middot; (φ +\u003c/em\u003e \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\sqrt{I}\\)\u003c/span\u003e\u003c/span\u003e\u003cem\u003e)\u003c/em\u003e,\u003c/p\u003e \u003cp\u003ewhere: Y\u0026mdash;coefficient of soil resistance; X - soil protection coefficient; φ - erosion and stream network development coefficient, and I - average slope (%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe classification of values for the soil resistance coefficient (Y) and the vegetation protection coefficient (X) is presented 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\u003eClassification of Soil Resistance and Vegetation Protection Coefficients (Y and X).\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\u003eCoefficient of Soil Resistance\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eY value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFine sediments and soils without erosion resistance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.80\u0026ndash;1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSediments, moraines, clay and other rock with little resistance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.60\u0026ndash;0.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeak rock, schistose, stabilized\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.50\u0026ndash;0.60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRock with moderate erosion resistance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.30\u0026ndash;0.50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHard rock, erosion resistant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.15\u0026ndash;0.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoefficient of soil protection\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eX Value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAreas without vegetal cover\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.80\u0026ndash;1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDamaged pasture and cultivated land\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.60\u0026ndash;0.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDamaged forest and bushes, pasture\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.40\u0026ndash;0.60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConiferous forest with little grove, scarce bushes, bushy prairie\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.25\u0026ndash;0.40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThin forest with grove\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.20\u0026ndash;0.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixed and dense forest\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.15\u0026ndash;0.20\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 classification of values for the erosion type and extent coefficient (φ) is presented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The average terrain slope (\u0026radic;I) is expressed as a percentage in decimal form.\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\u003eClassification of Soil Resistance and Vegetation Protection Coefficients (Y and X).\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\u003eCoefficient of erosion type and extent\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eφ value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAreas are strongly dissected by ravines, interfluves, and furrows, developed in eluviated and diluviated deposits or highly weathered, low-resistance rocks; such terrains are unsuitable for use without targeted anti-erosion measures.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.81\u0026ndash;1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFirm surfaces with concealed linear erosion on slopes exceeding 10\u0026deg;; degraded forests and pastures characterized by rills and small divides.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.61\u0026ndash;0.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCultivated areas on slopes of 5\u0026ndash;10\u0026deg;; degraded forests and pastures with reduced vegetation cover; barren terrains on resistant, impermeable rocks; active incision of watercourses into the terrain.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.31\u0026ndash;0.60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLatent erosion processes; well-developed forests with dense ground cover occupying ridge zones; stable meadows and pastures on slopes up to 5\u0026deg;; extensive lowland areas with slopes below 5\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.11\u0026ndash;0.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRelatively flat terrain (\u0026lt;\u0026thinsp;2\u0026deg;) where the weakest forms of erosion occur, typically in alluvial plains.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.10\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 EPM erosion categorization is shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. According to the research objectives set in this paper, the intensity of erosion will be considered using the Z coefficient (Lazarević, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e1985\u003c/span\u003e). Previous research has shown that the most important control factor in the intensity of soil erosion is land use change represented in the EPM model as the X coefficient (Manojlović et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2018\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 of Soil Erosivity Categories and Erosion Coefficient (Z).\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\u003eSoil erosivity category\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStrength of Erosion Processes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eErosion Coefficient (Z)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSediment accumulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eV\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery weak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.01\u0026ndash;0.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eV\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery weak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.11\u0026ndash;0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWeak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.21\u0026ndash;0.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWeak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.31\u0026ndash;0.40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMedium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.41\u0026ndash;0.55\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMedium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.56\u0026ndash;0.70\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIntensive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.71\u0026ndash;0.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIntensive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.86\u0026ndash;1.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003csub\u003e3\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eExcessive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.01\u0026ndash;1.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eExcessive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.21\u0026ndash;1.40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eExcessive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.41\u0026ndash;1.50\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\u003eIn the first segment of the research, the territory of Serbia was differentiated into zones with varying degrees of soil erosion susceptibility. However, to achieve more detailed spatial differentiation and identify spatial patterns of soil erosion, a cluster analysis method was applied. The municipality was selected as the basic spatial unit. Of the 197 municipalities in Serbia, 181 were included in the cluster classification. In contrast, the urban municipalities of Belgrade (10) and Niš (5), as well as the City of Novi Sad, were excluded from the analysis.\u003c/p\u003e \u003cp\u003eThe selected municipalities were characterized using a set of six indicators. The mean erosion coefficient (Z\u003csub\u003eav\u003c/sub\u003e) was chosen as the primary indicator of soil erosion. The second indicator was the coefficient of variation of the erosion process (CV\u003csub\u003ez\u003c/sub\u003e), which enabled the assessment of the variability of erosion intensity within each municipality. In this study, the coefficient of variation proved to be a highly significant statistical measure of landscape spatial homogeneity or heterogeneity. The influence of land use was quantified using the proportions of forest (F) and agricultural land (AgL). Morphometric characteristics of each municipality were defined by the mean slope angle (S) and the mean elevation (E).\u003c/p\u003e \u003cp\u003eCluster analysis is a statistical method for obtaining optimal groupings based on the similarity or distance between sample characteristics (Xu \u0026amp; Tian, \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Its primary objective is to achieve maximum similarity within clusters while ensuring high heterogeneity between clusters, as well as to identify hidden correlations within the data (Andrade et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Chen et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Cluster analysis has proven to be an appropriate method for differentiating soil erosion intensity across various spatial units in Serbia (Srejić et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this study, Agglomerative Hierarchical Clustering (AHC) was applied. The AHC algorithm begins by treating each data point as an individual cluster and progressively merging the most similar pairs into larger clusters. Clusters are merged based on the similarity of their centroids, determined by their proximity in the feature space. Thus, clusters with the highest similarity according to the selected linkage criterion are combined. Similarity between data points is quantified using distance measures, and in this study, the Euclidean distance was employed (Raux et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDifferent linkage criteria influence the shape and size of clusters, thereby significantly affecting the resulting dendrogram. In this research, Ward\u0026rsquo;s linkage method was used, which merges clusters by minimizing the increase in total within-cluster variance at each step. As a result, this method produces clusters that are approximately equal in size, highly compact, and characterized by a relatively uniform distribution of data within each cluster (Wani et al., 2024).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Soil erosion inventory and validation method\u003c/h2\u003e \u003cp\u003eFor validation purposes, a soil erosion inventory consisting of 102 erosion and 109 non-erosion samples was compiled through the integration of historical records, comparative analysis with recent conditions, and the interpretation of GIS-based datasets and satellite imagery (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Locations classified as erosion-prone correspond to clearly exposed (bare) surfaces occurring on steep slopes and underlain by highly erodible geological materials. In contrast, non-erosion locations represent stable environments, predominantly characterized by dense forest cover on gentle or flat terrain, developed over resistant geological substrates.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe inventory captures representative examples of both erosion-prone and stable conditions across the study area and serves as a reference dataset for the quantitative validation of the EPM-derived erosion coefficient. All procedures and approaches employed in this research are outlined in the flow chart provided in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.1. National-Scale Spatial Mapping of Erosion Intensity\u003c/h2\u003e \u003cp\u003eOn the territory of Serbia, very weak erosion prevails (53.8%). It is found in mountainous and hilly areas covered with forest vegetation. In mountainous regions, under conditions of extensive agriculture with meadows and pastures and thin soil layers, slightly higher erosion occurs (Z\u0026thinsp;=\u0026thinsp;0.11\u0026ndash;0.20). In lowland areas such as the Pannonian and Moesian plains and the alluvial plains of major rivers, erosion is weak, occurring only in narrow zones around tributary streams where the banks descend toward the riverbed. Beyond these zones, erosion shifts to moderate or intensive.\u003c/p\u003e \u003cp\u003eMedium erosion (Z\u0026thinsp;=\u0026thinsp;0.41\u0026ndash;0.70) is characteristic of hilly areas built of clastic sediments, where surfaces are predominantly covered with orchards. This intensity of erosion is also present in the zone of the Pannonian Plain due to the dune relief (Deliblato Sands) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn the hilly zone, on slopes below 20\u0026deg;, under conditions of cultivated land (arable fields and orchards), intensive and excessive erosion is present. These areas are located on the edges of basins and in the immediate surroundings of larger urban centers. Altogether, they cover 8.3% of the territory, of which 7.6% falls into the intensive class and 0.7% into the excessive class (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\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\u003eDistribution of Erosion Intensity Coefficient (Z) in the Republic of Serbia.\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=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" 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\u003eStrength of Erosion Processes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eErosion Coefficient (Z)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTotal area (km\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eShare in total area (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSediment accumulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e853.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVery weak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.01\u0026ndash;0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24,215\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e27.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVery weak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.11\u0026ndash;0.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23,391.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e26.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.21\u0026ndash;0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16,968.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.31\u0026ndash;0.40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4,464.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.41\u0026ndash;0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7,516\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.56\u0026ndash;0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3,729.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntensive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.71\u0026ndash;0.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3,711.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntensive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.86\u0026ndash;1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2,976.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExcessive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.01\u0026ndash;1.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e281.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExcessive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.21\u0026ndash;1.40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e251.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExcessive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.41\u0026ndash;1.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.1\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=\"Section2\"\u003e \u003ch2\u003e3.2. Spatial differentiation of municipalities based on erosion indicators\u003c/h2\u003e \u003cp\u003eThe results of the Agglomerative Hierarchical Clustering (AHC) indicate that municipalities in Serbia are grouped into six distinct clusters. The spatial distribution of these municipalities is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, while the main characteristics of each cluster are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA high proportion of between-cluster variance (74.2%) relative to within-cluster variance (25.8%) suggests strong homogeneity within clusters and clear separation between them. This reflects the underlying structural characteristics of the analyzed dataset. The clustering of municipalities was further validated using the Kruskal\u0026ndash;Wallis test (T\u0026ouml;lgyesi et al., \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Pinto et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), which confirmed that the differences among the identified clusters are statistically significant for all analyzed indicators (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\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\u003eControlling variables used in the AHC.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\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 \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eVariables \u0026ndash; Abbreviation (units)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eClusters\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eII\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIII\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eVI\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoefficient eropsion Zav (-)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.173\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.227\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.232\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.358\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.245\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoefficient of variation CVz (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e54.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e63.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e75.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e72.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e41.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForest cover F (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e62.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e60.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e36.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e33.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e6.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgricultural land AgL (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e47.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e52.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e83.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAltitude A (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e803.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e900.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e513.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e209.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e357.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e88.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSlope S (\u0026deg;)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e15.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e13.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.9\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 \u003cb\u003eCluster I\u003c/b\u003e predominantly comprises the peripheral and border areas of Serbia and includes 22 municipalities (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). This spatial unit encompasses the highest parts of the Dinaric Mountains in the west, the Šar Mountains in the south, the Serbo\u0026ndash;Macedonian Massif in the southeast, and sections of the Carpathian\u0026ndash;Balkan Mountains in the east.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIt covers approximately 13% of Serbia\u0026rsquo;s territory, i.e., 11,654 km\u0026sup2;. This cluster is characterized by the lowest erosion intensity (Z\u003csub\u003eav\u003c/sub\u003e = 0.173). The lowest erosion coefficient was identified in the municipality of Vranjska Banja (Z\u003csub\u003eav\u003c/sub\u003e = 0.145), while the highest was recorded in Štrpce (Z\u003csub\u003eav\u003c/sub\u003e = 0.216) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). The municipality of Štrpce is distinguished by the highest mean elevation (A\u0026thinsp;=\u0026thinsp;1,336 m.a.s.l.) and the steepest average slope (S\u0026thinsp;=\u0026thinsp;18.9\u0026deg;). However, despite a high proportion of forest cover (F\u0026thinsp;=\u0026thinsp;68%), the elevated erosion coefficient is associated with a substantial share of meadows and pastures (29%). In terms of the uniformity of soil erosion intensity, this is one of the most homogeneous clusters (CV\u003csub\u003ez\u003c/sub\u003e = 54%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). Within this predominantly mountainous region, with a mean elevation of A\u0026thinsp;=\u0026thinsp;804 m.a.s.l., one-third of the municipalities have an average elevation exceeding 1,000 m.a.s.l. (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee). All municipalities exhibit an average slope greater than 10\u0026deg;, with a cluster mean of S\u0026thinsp;=\u0026thinsp;15.5\u0026deg; (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ef). These morphometric characteristics have resulted in very limited agricultural land use. Agricultural land is nearly absent (AgL\u0026thinsp;=\u0026thinsp;2%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed), whereas forests cover as much as 82% of the cluster area (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003e \u003cb\u003eCluster II\u003c/b\u003e is slightly larger than the previous one and comprises 26 municipalities located in the western, southwestern, and southeastern mountainous regions of Serbia (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), including parts of the Dinaric Mountains, the Šar Mountains, and the Serbo\u0026ndash;Macedonian Massif. This cluster covers 14,017 km\u0026sup2;, corresponding to 16% of Serbia\u0026rsquo;s territory. It is characterized by a weak erosion category, with an average erosion coefficient, Z\u003csub\u003eav\u003c/sub\u003e, of 0.227. The lowest erosion coefficient was recorded in Arilje (Z\u003csub\u003eav\u003c/sub\u003e = 0.181), while the highest was observed in Gora (Z\u003csub\u003eav\u003c/sub\u003e = 0.345) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). The variability of the erosion process is somewhat higher than in the previous cluster (CV\u003csub\u003ez\u003c/sub\u003e = 64%). The highest homogeneity of the erosion coefficient is found in Gora (CV\u003csub\u003ez\u003c/sub\u003e = 51%), while the lowest is observed in Dečani (CV\u003csub\u003ez\u003c/sub\u003e = 77%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). A high proportion of forest cover (F\u0026thinsp;=\u0026thinsp;63%) and a very low share of agricultural land (AgL\u0026thinsp;=\u0026thinsp;5.5%) are the main land-use characteristics of this spatial unit (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec,d). Municipalities within this cluster exhibit the highest average elevation (A\u0026thinsp;=\u0026thinsp;900 m.a.s.l.), ranging from A\u0026thinsp;=\u0026thinsp;609 m.a.s.l. in Svrljig to A\u0026thinsp;=\u0026thinsp;1,707 m.a.s.l. in Gora (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee). The mean slope angle is S\u0026thinsp;=\u0026thinsp;14\u0026deg; (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ef).\u003c/p\u003e \u003cp\u003e \u003cb\u003eCluster III\u003c/b\u003e is the most numerous and spatially extensive cluster, comprising 45 municipalities and covering 28% of Serbia\u0026rsquo;s territory (24,882 km\u0026sup2;). It is most widely distributed in Eastern Serbia, within the West Morava and Kosovo and Metohija basins (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). In terms of soil erosion susceptibility, this spatial unit does not differ significantly from the previous cluster, with an average erosion coefficient of Z\u003csub\u003eav\u003c/sub\u003e = 0.232. The municipalities of Golubac and Krupanj exhibit the lowest erosion coefficients (Z\u003csub\u003eav\u003c/sub\u003e = 0.191), while Glogovac records the highest value (Z\u003csub\u003eav\u003c/sub\u003e = 0.284) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). The coefficient of variation (CV\u003csub\u003ez\u003c/sub\u003e = 75%) indicates greater spatial variability in the erosion process than in the previous two clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). A substantial portion of this cluster is covered by forests (F\u0026thinsp;=\u0026thinsp;60%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec). However, at lower elevations (A\u0026thinsp;=\u0026thinsp;514 m.a.s.l.) and on gentle slopes (S\u0026thinsp;\u0026lt;\u0026thinsp;10\u0026deg;), a higher share of agricultural land (AgL\u0026thinsp;=\u0026thinsp;20%) has resulted (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed). Consequently, municipalities with a high proportion of agricultural land (AgL\u0026thinsp;\u0026gt;\u0026thinsp;30%) and steeper slopes are predisposed to the highest erosion rates within this cluster.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCluster IV\u003c/b\u003e covers 12,287 km\u0026sup2;, accounting for 14% of Serbia\u0026rsquo;s total area, and includes 37 municipalities. The largest continuous area of this cluster is located south of the Sava and Danube rivers, as well as within the immediate basin of the Velika Morava. Smaller areas are found on the southern slopes of Fruška Gora. At the same time, the southernmost extent reaches the valleys of the Južna Morava and Nišava rivers, including the peri-urban zone of Niš (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The average erosion coefficient is Z\u003csub\u003eav\u003c/sub\u003e = 0.285. The municipalities of Smederevo (Z\u003csub\u003eav\u003c/sub\u003e = 0.330), Lajkovac (Z\u003csub\u003eav\u003c/sub\u003e = 0.317), and Batočina (Z\u003csub\u003eav\u003c/sub\u003e = 0.315) exhibit the highest values, whereas Kostolac (Z\u003csub\u003eav\u003c/sub\u003e = 0.229) and Šid (Z\u003csub\u003eav\u003c/sub\u003e = 0.242) have the lowest erosion intensity (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). The average coefficient of variation is CV\u003csub\u003ez\u003c/sub\u003e = 73%. High values in certain municipalities (CV\u003csub\u003ez\u003c/sub\u003e \u0026gt; 80%) indicate not only internal heterogeneity of erosion processes but also diverse geographical conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). This specific combination of factors has resulted in a higher level of erosion susceptibility. Nearly half of the available land in this cluster is used for agriculture (AgL\u0026thinsp;=\u0026thinsp;47%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed), while forest cover is considerably lower compared to previous clusters (F\u0026thinsp;=\u0026thinsp;37%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec). Lower elevations (A\u0026thinsp;=\u0026thinsp;210 m.a.s.l.) and gentle slopes (S\u0026thinsp;\u0026lt;\u0026thinsp;5\u0026deg;) are the main characteristics of this predominantly agrarian landscape (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee,f).\u003c/p\u003e \u003cp\u003e \u003cb\u003eCluster V\u003c/b\u003e covers 4,015 km\u0026sup2;, accounting for approximately 5% of Serbia\u0026rsquo;s territory. It is the smallest cluster in terms of both spatial extent and the number of municipalities, comprising 12 municipalities distributed across several areas. The western area includes the municipalities of Vladimirci, Ub, and Koceljeva. The central part consists of Smederevska Palanka, Mladenovac, Topola, and Rača, located within the immediate basin of the Velika Morava. The southernmost municipalities belong to the Kosovo and Metohija basins (Orahovac, Klina, Srbica, Vučitrn, and Obilić) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe distinguishing feature of this cluster is the highest susceptibility to soil erosion (Z\u003csub\u003eav\u003c/sub\u003e = 0.358). The highest erosion intensity is recorded in the municipalities of Vladimirci (Z\u003csub\u003eav\u003c/sub\u003e = 0.413) and Smederevska Palanka (Z\u003csub\u003eav\u003c/sub\u003e = 0.400) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea), placing them in the moderate erosion category. The average coefficient of variation of the erosion process is CV\u003csub\u003ez\u003c/sub\u003e = 73%, which is comparable to Clusters III and IV (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). Forests cover approximately one-third of the cluster area, slightly less than in the previous cluster, with the largest shares in Koceljeva (F\u0026thinsp;=\u0026thinsp;47%) and Srbica (F\u0026thinsp;=\u0026thinsp;46%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec). Agricultural land is highly represented, covering nearly half of the cluster (AgL\u0026thinsp;=\u0026thinsp;52%), and in some municipalities, it exceeds 70% (e.g., Smederevska Palanka) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed). A key characteristic of this spatial unit is the combination of moderately higher elevation (A\u0026thinsp;=\u0026thinsp;357 m.a.s.l.) and relatively steeper slopes (S\u0026thinsp;=\u0026thinsp;5.5\u0026deg;) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee,f) on land predominantly used for agricultural production. This synergy of geographical factors increases the terrain's susceptibility to erosion.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCluster VI\u003c/b\u003e covers 19,311 km\u0026sup2;, representing approximately one-fifth of Serbia\u0026rsquo;s territory. It forms the most homogeneous continuous area in northern Serbia and includes 39 municipalities (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). This spatial unit falls into the weak erosion category, with an average erosion coefficient Z\u003csub\u003eav\u003c/sub\u003e = 0.245 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). The lowest coefficient of variation (CV\u003csub\u003ez\u003c/sub\u003e = 41%) indicates minimal spatial variability of the erosion process (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb), reflecting the high uniformity of geographical conditions in this intensively used agricultural region. Despite a very high proportion of agricultural land (AgL\u0026thinsp;=\u0026thinsp;83%) and a low share of forest cover (F\u0026thinsp;=\u0026thinsp;7%), this cluster is not predisposed to higher erosion rates (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec,d). The key controlling factor is terrain morphometry, particularly low elevation (A\u0026thinsp;\u0026lt;\u0026thinsp;100 m.a.s.l.) and very gentle slopes (S\u0026thinsp;\u0026lt;\u0026thinsp;1\u0026deg;) (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ee,f). However, even slight increases in these parameters lead to higher erosion intensity. For example, the municipalities of Alibunar and Vršac do not differ significantly in land use from other municipalities in this cluster, yet they exhibit the highest erosion intensity (Z\u003csub\u003eav\u003c/sub\u003e = 0.275) due to somewhat higher elevations and steeper slopes (A\u0026thinsp;\u0026gt;\u0026thinsp;100 m.a.s.l.; S\u0026thinsp;\u0026asymp;\u0026thinsp;2\u0026deg;).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Validation Results\u003c/h2\u003e \u003cp\u003eTo evaluate the performance of the Erosion Potential Model (EPM), a validation procedure was conducted using the soil erosion inventory (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), consisting of 211 georeferenced locations, including 102 erosion and 109 non-erosion samples. The validation was performed by extracting the EPM-derived erosion coefficient (Z) values at the sample locations and assessing the ability of the model to discriminate between erosion-prone and stable environments. A binary classification was applied using a threshold value, which was determined through systematic evaluation and identified at Z\u0026thinsp;=\u0026thinsp;0.11.\u003c/p\u003e \u003cp\u003eThe obtained results indicate a complete separation between erosion and non-erosion samples, with all evaluation metrics reaching their maximum values (ROC\u0026ndash;AUC\u0026thinsp;=\u0026thinsp;1.00; accuracy\u0026thinsp;=\u0026thinsp;1.00; precision\u0026thinsp;=\u0026thinsp;1.00; recall\u0026thinsp;=\u0026thinsp;1.00; F\u003csub\u003e1\u003c/sub\u003e-score\u0026thinsp;=\u0026thinsp;1.00). The confusion matrix confirms that all non-erosion samples (n\u0026thinsp;=\u0026thinsp;109) and all erosion samples (n\u0026thinsp;=\u0026thinsp;102) were correctly classified, with no false positives or false negatives observed. The distribution of Z values further supports this finding. Non-erosion samples are consistently associated with low Z values, while erosion samples exhibit significantly higher values, with no overlap between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA perfect confusion matrix and ROC curve occur because the data are clearly separated, with no \u0026ldquo;borderline\u0026rdquo; cases between erosion and stable zones. The model and the validation data are based on the same criteria (slope, vegetation, lithology), which leads to complete agreement.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe ROC curve (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e) demonstrates the strong discriminative structure of the Z coefficient, confirming its effectiveness in separating clearly defined erosion-prone and stable conditions. Although ROC analysis and confusion-matrix-based evaluation are commonly used in predictive classification settings, in the present study these metrics should be interpreted primarily as measures of agreement between the EPM-derived erosion coefficient and the expert-defined erosion inventory. Because the validation samples were selected using geomorphological and land-cover criteria closely related to the EPM input factors, the observed perfect separation reflects strong model\u0026ndash;inventory consistency rather than fully independent predictive generalization.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTherefore, the validation results confirm that the EPM model provides a robust and internally consistent representation of soil erosion susceptibility under clearly defined environmental conditions.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Comparative Analysis with Global Soil Erosion Studies\u003c/h2\u003e \u003cp\u003eAs already mentioned, among the analyzed semi-quantitative methods applied worldwide, EPM was rated as the most quantitative (De Vente \u0026amp; Poesen, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Also, comparative analysis of sediment yield estimations using different empirical soil erosion models showed that the EPM model is the most appropriate concerning the fulfilment of various tasks such as: estimation of total transport material, assessment of average pattern of erosion risk, identification of high-risk areas, identification of hot spots, detailed erosion and deposition patterns, effects of conservation measures (Blinkov \u0026amp; Kostadinov, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The EPM model can be used to estimate all soil erosion types, meeting the needs of various scientific fields and sectors (agriculture, forestry, water resources, and watershed management). Also, the advantage of the EPM model is its applicability across all scales (Efthimiou et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Finally, qualitative analysis showed that the EPM model can perform best for mountainous areas allowing for the identification on the map and visualization of the areas most prone to erosion, as well as the predicted amount of soil loss and its spatial distribution.\u003c/p\u003e \u003cp\u003eA comparative analysis with global, regional, and local soil erosion studies showed that the results of this study are in line with previous research. In a study of the applicability of the EPM model and modified EPM (mEPM) to estimate gross and net erosion rates at the global scale, the analysis shows that the land use pattern has the greatest impact on the model results, as verified by the X coefficient in model aquation (Bezak et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Soil loss also varies with vegetation cover and land use management practices, which make both extremely influential factors in erosion formation and overall soil loss in the Mediterranean regions (Kosmas et al., 2000; Stefanidis \u0026amp; Stathis, \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Analysis of the spatial distribution of land use as a determining factor of the intensity of soil erosion has also been detected in regional studies in Bosnia and Hercegovina, which is clearly visible in the spatial distribution of the soil erosion protection coefficient (Tošić et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e). Previous research has established a functional relationship between land use changes and soil erosion intensity at the local level, linking these processes with processes of deagrarization in rural settlements (Srejić et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Spatial differentiation of torrential watersheds in border regions in Serbia has revealed varying trends in soil erosion, highlighting the importance of accurate assessments of these changes through land use impact indicators (Petrović et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Also, results from small catchments in Croatia showed a good approximation of erosion intensity and soil loss to land cover change at different annual and seasonal spatial-temporal scales (Dragičević et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNamely, previous research in Serbian watersheds found that the EPM method is most sensitive to the X coefficients, followed by the φ and Y coefficients, while slope, air temperature, and precipitation have a smaller effect on the EPM output. The correlation matrix of EPM model parameters at the statistical significance level of α\u0026thinsp;=\u0026thinsp;0.05 showed that the coefficient X has the major impact on the sediment gross erosion and coefficient erosion (W\u0026thinsp;=\u0026thinsp;\u003cem\u003ef\u003c/em\u003e(X), r\u0026thinsp;=\u0026thinsp;0.778, and Z\u0026thinsp;=\u0026thinsp;\u003cem\u003ef\u003c/em\u003e(X), r\u0026thinsp;=\u0026thinsp;0.820, respectively) (Manojlović et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). It can be concluded that coefficient X is the primary factor of the strength of the erosive process. Also, the sensitivity estimates of the USLE models indicated that land cover (C factor) is the most important factor in determining soil loss across different mountain and agricultural systems (Perovic et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2016\u003c/span\u003e, Estrada-Carmona et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Therefore, Bezak et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) point out that the land use/land cover parameter is particularly important in both the EPM and USLE models, as the main determinant of changes in soil erosion intensity (Bezak et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Spatial differentiation soil erosion intensity according clustering municipaliy\u003c/h2\u003e \u003cp\u003eAlthough the overall erosion intensity classification indicates that erosion processes in Serbia are predominantly very weak and weak, areas affected by moderate and intensive erosion are also present. The spatial distribution of erosion intensity, differentiated through cluster analysis of dominant indicators at the municipal level, offers a new perspective on the spatial analysis of this fundamentally geomorphological process. The results obtained using the EPM method and GIS-based modelling highlight the significant roles of terrain morphometric parameters and anthropogenic pressure in shaping the spatial distribution of erosion risk (Kostadinov et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Đorđević et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2026\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this context, the influence of key controlling indicators was examined. In addition to forest and agricultural land, areas covered by meadows and pastures were included in the analysis as transitional forms between these two land-use types. The quantitative relationships among the selected indicators were assessed using Pearson\u0026rsquo;s correlation coefficient within a correlation matrix (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The results indicate a high level of correlation among the indicators at a significance level of α\u0026thinsp;\u0026lt;\u0026thinsp;0.0001.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCorrelation matrix (Pearson\u0026rsquo;s r) of analyzed variables.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\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 \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eZ\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAgL\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGr\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.767\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.826\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.435\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.624\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-0.805\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.767\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.685\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.920\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAgL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.826\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-0.594\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.729\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-0.882\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGr\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.435\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.396\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.594\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.741\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.574\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.624\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.685\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.729\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.741\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.861\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.805\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.920\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-0.882\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.574\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.861\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1\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 results of the quantitative analysis of controlling factors of soil erosion indicate a strong relationship between elevation and the decrease in erosion intensity in Serbia. The erosion coefficient (Z) shows a high negative correlation with morphometric indicators, namely slope angle S (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.805) and elevation E (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.624), while these two variables are strongly positively correlated with each other (r\u0026thinsp;=\u0026thinsp;0.860). For example, in some border municipalities in eastern Serbia, the correlation for the function Z\u0026thinsp;=\u0026thinsp;f(E) indicates an even stronger dependence of decreasing erosion intensity with increasing elevation (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.98) (Manojlović et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe spatial variation of the erosion coefficient Z, differentiated according to relief conditions, shows that areas affected by higher erosion intensity increase from mountainous border municipalities (clusters I and II) toward municipalities predominantly located in hilly zones characterized by weakly dissected and gently undulating terrain (cluster V). Areas with preserved forest cover, particularly in mountainous regions (represented by clusters I and II), exhibit significantly lower Z coefficient values compared to degraded pastures and open cropland. Forest cover shows a strong negative correlation with the erosion coefficient (Z\u0026thinsp;=\u0026thinsp;f(F), r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.767). Forested areas are predominantly located on steeper slopes (r\u0026thinsp;=\u0026thinsp;0.919) and at higher elevations (r\u0026thinsp;=\u0026thinsp;0.684). These strong positive relationships between forests and morphometric indicators explain their anti-erosion effect in parts of Serbia that are naturally predisposed to higher erosion categories (greater relief dissection and steeper slopes), yet still record low erosion rates. Slopes exceeding 20\u0026deg; are characteristic of mountainous areas (clusters I and II), which are lithologically composed of hard and compact rocks. Due to subaerial processes, a weathering crust develops on the surface, which is susceptible to intensive washing. However, erosion potential does not increase linearly with slope. On slopes steeper than 40\u0026deg;, the weathering crust is largely removed, resulting in a significantly reduced erosion potential.\u003c/p\u003e \u003cp\u003eGrasslands and pastures are also negatively correlated with the erosion coefficient Z (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.435). These land-use types are predominantly located at higher elevations, as indicated by the positive correlation in the function Gr\u0026thinsp;=\u0026thinsp;f(E) (r\u0026thinsp;=\u0026thinsp;0.741), and often co-occur with forest cover (r\u0026thinsp;=\u0026thinsp;0.395). Higher elevation and a greater proportion of forest cover contribute to a reduction in erosion intensity. However, demographic characteristics of border municipalities have also influenced the relatively low erosion rates. Serbia is characterized by highly differentiated rural areas with pronounced processes of regional demographic differentiation (Gajić et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Grčić et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Gajić Protić et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). A large number of settlements with weak demographic potential indicates a decline in agricultural activity (Manojlović et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Under the influence of various socio-economic factors (industrialization, agrarian reform, and rural\u0026ndash;urban migration) during the second half of the 20th century, sparsely populated and underdeveloped rural settlements in mountainous peripheral and border regions experienced strong processes of depopulation and deagrarization (Martinović \u0026amp; Ratkaj, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The abandonment of agricultural land has consequently contributed to a reduction in erosion intensity in these municipalities.\u003c/p\u003e \u003cp\u003eOn the other hand, agricultural land (AgL) has the strongest influence on erosion intensity (Z\u0026thinsp;=\u0026thinsp;f(AgL), r\u0026thinsp;=\u0026thinsp;0.826). Agricultural areas are primarily located at lower elevations (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.728) and on terrains with lower slope angles (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.882). This indicates that the most intensively cultivated parts of municipalities in Serbia are also the most susceptible to erosion processes. The spatial differentiation of municipalities at risk of erosion, determined through cluster analysis, reveals a general gradient: peripheral border municipalities (clusters I and II) \u0026ndash; municipalities in the transitional zone between hilly and mountainous terrain (cluster III) \u0026ndash; municipalities predominantly located in hilly and partly lowland areas (clusters IV and V).\u003c/p\u003e \u003cp\u003eIn this context, municipalities located in the northern part of central Serbia exhibit the highest erosion intensity rates. These municipalities are geographically predisposed to increased land degradation: clastic rocks cover more than 70% of the territory, dominated by Neogene sediments, combined with a high proportion of arable land and reduced forest cover (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Additionally, these areas possess the highest demographic and agrarian potential, characterized by intensive agricultural production, which results in stronger anthropogenic pressure on land resources and intensification of erosion processes (Srejić et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eGeographical characteristics of clusters.\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\u003eCluster\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\u003eI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMunicipalities whose territory is predominantly located in mountainous zones with moderately (100\u0026ndash;300 m/km\u0026sup2;) and highly dissected relief (300\u0026ndash;800 m/km\u0026sup2;). The lithology is composed of magmatic and metamorphic rocks, as well as limestones. On slopes up to 35\u0026deg;, a weathering mantle has developed, with a thickness ranging from a few centimeters to several meters. In terms of land use, forests have a dominant share (\u0026gt;\u0026thinsp;70%), while arable land accounts for less than 10%.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMunicipalities located in mountainous areas with the presence of hilly and lowland terrains. In addition to highly and moderately dissected relief, this cluster also includes areas with weakly dissected relief (30\u0026ndash;100 m/km\u0026sup2;). Alongside the lithological composition characteristic of Cluster 1, unconsolidated clastic sedimentary rocks, primarily Neogene sediments (clays, sands, and gravels) are also present.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMunicipalities situated in the transitional zone between hilly and mountainous regions. Areas with moderate and weak relief dissection are significantly represented. The proportion of clastic sedimentary rocks and agricultural land (arable fields and orchards) increases, while forest cover remains above 50%.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMunicipalities where mountainous terrain is present, but hilly and lowland landscapes dominate, characterized by weakly dissected, slightly undulating (5\u0026ndash;30 m/km\u0026sup2;), and flat relief (0\u0026ndash;5 m/km\u0026sup2;). Clastic sedimentary rocks cover more than 70% of the territory, dominated by Neogene sediments with a notable presence of Quaternary deposits. There is an increased share of arable land and a reduced proportion of forest cover (\u0026lt;\u0026thinsp;50%).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMunicipalities whose territories are almost entirely located in hilly regions with predominantly weakly dissected and slightly undulating relief. Clastic sedimentary rocks, mainly Neogene sediments, cover nearly the entire municipal territory. Forest cover is limited, while the share of arable land has significantly increased.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMunicipalities entirely situated in lowland areas (Pannonian Plain), characterized by predominantly flat relief (0\u0026ndash;5 m/km\u0026sup2;). In terms of lithology, only clastic sediments of Quaternary age are present (sands, loess, and clays). Regarding land use, arable land (predominantly cropland) accounts for more than 60% of the territory, while forests cover less than 10% of the total area.\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=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.3. The Role of Forests and Agriculture in Land Management for Soil Erosion Mitigation\u003c/h2\u003e \u003cp\u003eForests play a key role in preventing soil erosion because they provide natural mechanisms for soil stabilization, increase water retention, and reduce the intensity of runoff and denudation. Several interrelated mechanisms mitigate soil erosion through forest ecosystems. One of the main mechanisms is the increase in cohesion through root systems, which strengthen the soil through a stable network, preventing the displacement of soil masses (Gonzalez-Ollauri \u0026amp; Mickovski, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Various forest species, such as Scots pine (Pinus nigra), are among the most successful for afforestation of bare and eroded terrain in Serbia, demonstrating high adaptability to such terrain (Ratknić et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Natural beech forests (Fagus sylvatica) are biologically very stable. They are associated with very low erosion due to their dense crown cover, thus reducing the direct \"bombardment\" of raindrops on the soil (Lukić et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Dense vegetation reduces erosion many times over compared to bare terrain. Species such as hawthorn (Crataegus monogyna) show the highest values ​​of root density and the greatest reduction in erosion compared to alder and medlar (Dalir et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). At the same time, erosion processes in Serbia are also intensifying under the influence of anthropogenic factors and various climatic stresses. Some authors, such as Braunović et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), show that several factors have a significant influence and increase the susceptibility to water and wind erosion, such as excessive grazing, combined with deforestation and conversion of forest areas to pastures, which leads to a reduction in plant cover and compaction of the surface soil layer (Braunović et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). For these reasons, the maintenance and restoration of forest areas must be seen as a key stabilizer of the degradation and devastation of soil systems.\u003c/p\u003e \u003cp\u003eIntensive agriculture on sloping terrain (especially trench crops) without implementing conservation practices results in the highest levels of soil loss (Bezbradica et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Historical analyses show that the expansion of arable land in the 20th century was accompanied by a more pronounced development of ravine forms (Ristić et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Simultaneously, rising temperatures are causing forest drying, especially in pedunculate oak (Quercus petraea). This process leads to physiological weakening of trees and thinning of crown cover, thereby weakening the protective function of forest ecosystems (Fern\u0026aacute;ndez et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Sukmawijaya et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). With the loss of dense vegetation cover, raindrops have a strong, direct effect on the soil surface, increasing aggregate destruction and surface runoff (Ascencio-Sanchez et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOn the other hand, agriculture is a critical factor in exacerbating erosion when practised without conservation measures. Intensive cultivation with trench crops on sloping terrain accelerates the mineralization. Soils with a low humus content are less resistant to the destructive force of raindrops (Deđanski et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Modern concepts of sustainable agricultural development suggest an interaction between forestry and agriculture, grounded in the principle of complementarity within a common landscape context. Mixed agroforestry systems with a dominance of native species, contour tillage, terracing, and the use of cover crops reduce erosion risk by restoring some of the ecosystem functions of natural vegetation (Golijanin et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Several studies recommend carrying out group-selective and valley fellings on steep terrain rather than clear-cutting to limit and prevent exposure and erosion risks.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eWith the ongoing processes of urbanization, depopulation, and deagrarization in Serbia, a decline in the intensity of mechanical soil erosion has been observed. In this study, geospatial data (lithology, land use, terrain slope, and erosion coefficient) with a 30 m spatial resolution were analyzed to produce a high-resolution national-scale map of mechanical soil erosion. Using the Erosion Potential Model (EPM), the results indicate that the mean erosion coefficient (Z\u003csub\u003eav\u003c/sub\u003e) for Serbia is 0.239, reflecting a decrease in erosion intensity compared to 1985. The findings clearly demonstrate that areas most affected by intensive erosion are agricultural lands located on steeper slopes (\u0026gt;\u0026thinsp;10\u0026deg;) and under specific geological conditions (sand, gravel, organogenic, and clay-rich sediments). Overall, 7.6% of the territory is exposed to intensive erosion, while 0.7% of the study area is affected by excessive erosion. A perfect ROC-AUC score of 100%, obtained based on the inventory of erosion points, indicates strong consistency between the input data and the final results.\u003c/p\u003e \u003cp\u003eTo enable a more detailed understanding of erosion processes at the local level, spatial differentiation was performed through cluster analysis of the final dataset at the municipal scale. Six clusters were identified based on six indicators: mean erosion coefficient, coefficient of variation of the erosion process, the influence of forest cover, agricultural land, and morphometric characteristics (elevation and slope). The results clearly indicate that morphometric conditions and land use play a key role in controlling erosion dynamics in Serbia. The highest susceptibility to erosion was identified in agriculturally dominated areas with moderate slopes (Cluster V), where the combination of steeper terrain and intensive land use significantly enhances erosion processes. In contrast, the lowland regions of northern Serbia (Cluster VI), although predominantly agricultural, exhibit low erosion potential due to minimal terrain slope. Forest-covered areas proved highly effective at mitigating soil erosion.\u003c/p\u003e \u003cp\u003eFuture research in Serbia should focus on integrating existing erosion data with high-resolution climatic datasets to improve current results and support the development of a national sediment yield map as a key component of the EPM. More efficient land management could be achieved through detailed analyses at finer spatial scales. In this context, the application of the EPM at the level of rural settlements or torrential catchments is recommended. Such an approach would enable the identification of areas with the highest erosion risk and allow for field-based validation of the empirical model and remote sensing methods. Due to its comprehensive spatial coverage, the results of this study are highly relevant for decision-makers at both local and national levels. In addition to local authorities and governmental institutions, these findings may also support various environmental sectors, including emergency management services, forestry, water management, spatial planning, and nature conservation. An effective approach to land management must be multidisciplinary, taking into account natural conditions alongside contemporary demographic and socio-economic processes.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e \u003cb\u003eCredit authorship contribution statement\u003c/b\u003e \u003c/p\u003e \u003cp\u003eUroš Durlević: Conceptualization, software, formal analysis, data curation, writing\u0026mdash;original draft preparation, project administration, funding acquisition. Tanja Srejić: Conceptualization, validation, writing\u0026mdash;original draft preparation. Sanja Manojlović: Methodology, writing\u0026mdash;review and editing, supervision. Marko V. Milošević: Software, writing\u0026mdash;review and editing. Natalija Batoćanin: Investigation, visualization. Milica Dobrić: Investigation, data curation. Jelena Svetozarević: Validation, writing\u0026mdash;original draft preparation. Velibor Ilić: Software, data curation, validation.\u003c/p\u003e\u003cp\u003e \u003ch2\u003eDeclaration of competing interest\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis research was funded by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Contract number 451-03-33/2026-03/200172 and 451-03-33/2026-03/200091).\u003c/p\u003e\u003ch2\u003eData Availability Statement\u003c/h2\u003e \u003cp\u003eTo obtain the data from this study, please contact the authors via email.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAkhmetova D (2025) Anthropogenic impact on soil and vegetation in Turkistan Region: Chemical composition and heavy metal contamination. J Geogr Inst Jovan Cvijic SASA 75(1):1\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/IJGI240426013A\u003c/span\u003e\u003cspan address=\"10.2298/IJGI240426013A\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAleksova B, Lukić T, Milevski I, Spalević V, Marković SB (2023) Modelling water erosion and mass movements (wet) using GIS-based multi-hazard susceptibility assessment approaches: A case study\u0026mdash;Kratovska Reka catchment (North Macedonia). Atmosphere 14:1139. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/atmos14071139\u003c/span\u003e\u003cspan address=\"10.3390/atmos14071139\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlewell C, Borrelli P, Meusburger K, Panagos P (2019) Using the USLE: Chances, challenges and limitations of soil erosion modelling. Int Soil Water Conserv Res 7:203\u0026ndash;225. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.iswcr.2019.05.004\u003c/span\u003e\u003cspan address=\"10.1016/j.iswcr.2019.05.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAltobelli F, Vargas R, Corti G, Dazzi C, Montanarella L, Monteleone A, Caon L, Piazza MG, Calzolari C, Munaf\u0026ograve; M, Benedetti A (2020) Improving soil and water conservation and ecosystem services by sustainable soil management practices: From a global to an Italian soil partnership. Ital J Agron 15:1765. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4081/ija.2020.1765\u003c/span\u003e\u003cspan address=\"10.4081/ija.2020.1765\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrade EM, Pal\u0026aacute;cio HAQ, Souza IH, Le\u0026atilde;o RAO, Guerreiro MJ (2008) Land use effects in groundwater composition of an alluvial aquifer (Trussu River, Brazil) by multivariate techniques. Environ Res 106:170\u0026ndash;177. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envres.2007.10.008\u003c/span\u003e\u003cspan address=\"10.1016/j.envres.2007.10.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAscencio-Sanchez M, Padilla-Castro C, Riveros-Lizana C, Hermoza-Espez\u0026uacute;a RM, Atalluz-Ganoza D, Sol\u0026oacute;rzano-Acosta R (2025) Impacts of land use on soil erosion: RUSLE analysis in a sub-basin of the Peruvian Amazon (2016\u0026ndash;2022). Geosciences 15:15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/geosciences15010015\u003c/span\u003e\u003cspan address=\"10.3390/geosciences15010015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBezak N, Borrelli P, Mikoš M, Jemec Auflič M, Panagos P (2024) Towards multi-model soil erosion modelling: An evaluation of the erosion potential method (EPM) for global soil erosion assessments. CATENA 234:107596. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.catena.2023.107596\u003c/span\u003e\u003cspan address=\"10.1016/j.catena.2023.107596\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBezbradica L, Josimović B, Milijić S (2023) Impact of repurposing forest land on erosion and sediment production\u0026mdash;Case study: Krupanj Municipality\u0026mdash;Serbia. Forests 14:1127. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/f14061127\u003c/span\u003e\u003cspan address=\"10.3390/f14061127\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlinkov I, Kostadinov S (2010) Applicability of various erosion risk assessment methods for engineering purposes. In: Proceedings of the BALWOIS 2010 Conference, Ohrid, Republic of Macedonia, 25\u0026ndash;29 May 2010\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBorrelli P, Robinson DA, Panagos P, Ballabio C et al (2020) Land use and climate change impacts on global soil erosion by water (2015\u0026ndash;2070). Proc Natl Acad Sci U S A 117(36):21994\u0026ndash;22001. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.2001403117\u003c/span\u003e\u003cspan address=\"10.1073/pnas.2001403117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBraunović S, Cvetković J, Jovanović F, Polovina S, Šurjanac N, Stojanović V, Momirović N (2025) Assessment of soil erosion intensity using the erosion potential method: A case study of the Grdelica Gorge, Serbia. Sustain For 91:37\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5937/SustFor2591037\u003c/span\u003e\u003cspan address=\"10.5937/SustFor2591037\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eĆalić J, Milošević MV, Milivojević M, Gaudenji T (2017) Relief of Serbia. In: Radovanović M (ed) Relief of Serbia. Geographical Institute Jovan Cvijić SASA, Belgrade, Serbia, pp 23\u0026ndash;92. (In Serbian)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarević I, Sibinović M, Manojlović S, Batoćanin N, Petrović AS, Srejić T (2021) Geological approach for landfill site selection: A case study of Vršac Municipality, Serbia. Sustainability 13:7810. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su13147810\u003c/span\u003e\u003cspan address=\"10.3390/su13147810\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen J, Li F, Fan Z, Wang Y (2016) Integrated application of multivariate statistical methods to source apportionment of watercourses in the Liao River Basin, Northeast China. Int J Environ Res Public Health 13:1035. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijerph13101035\u003c/span\u003e\u003cspan address=\"10.3390/ijerph13101035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCopernicus Land Monitoring Service (2023) High resolution layer: Tree cover and forests. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://land.copernicus.eu/en/products/high-resolution-layer-forests-and-tree-cover\u003c/span\u003e\u003cspan address=\"https://land.copernicus.eu/en/products/high-resolution-layer-forests-and-tree-cover\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 7 Mar 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eĆurić V, Durlević U, Ristić N, Novković I, Čegar N (2022) GIS application in analysis of threat of forest fires and landslides in the Svrljiški Timok basin (Serbia). Glasn Srp Geogr Drust 102(1):107\u0026ndash;130. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/GSGD2201107C\u003c/span\u003e\u003cspan address=\"10.2298/GSGD2201107C\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDalir P, Naghdi R, Jafari S, Tsioras PA (2025) Comparative assessment of woody species for runoff and soil erosion control on forest road slopes in harvested sites of the Hyrcanian forests, Northern Iran. Forests 16:1013. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/f16061013\u003c/span\u003e\u003cspan address=\"10.3390/f16061013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Vente J, Poesen J (2005) Predicting soil erosion and sediment yield at the basin scale: Scale issues and semi-quantitative models. Earth-Sci Rev 71:95\u0026ndash;125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.earscirev.2005.02.002\u003c/span\u003e\u003cspan address=\"10.1016/j.earscirev.2005.02.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeđanski V, Durlević U, Kovjanić A, Lukić T (2024) GIS-based spatial modeling of landslide susceptibility using BWM-LSI: A case study\u0026mdash;City of Smederevo (Serbia). Open Geosci 16:20220688. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1515/geo-2022-0688\u003c/span\u003e\u003cspan address=\"10.1515/geo-2022-0688\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDildabek D, Aliaskarov D, Kaimuldinova K, Bakanov N, Yegizbayeva A, Shuangjie Z (2025) Projected land use land cover dynamics and tourism suitability assessment in the selected Kazakhstan area. J Geogr Inst Jovan Cvijic SASA 75(3):461\u0026ndash;469. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/IJGI2503461D\u003c/span\u003e\u003cspan address=\"10.2298/IJGI2503461D\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDominici R, Larosa S, Viscomi A, Mao L, De Rosa R, Cianflone G (2020) Yield erosion sediment (YES): A PyQGIS plug-in for the sediments production calculation based on the erosion potential method. Geosciences 10:324. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/geosciences10080324\u003c/span\u003e\u003cspan address=\"10.3390/geosciences10080324\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eĐorđević M, Đokić M, Manić M, Vesković J, Dragović R, Smičiklas I, Dragović S, Onjia A (2026) Advanced GIS-based RUSLE modeling for soil erosion estimation in the Toplica River Basin, Serbia. Geosciences 16:83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/geosciences16020083\u003c/span\u003e\u003cspan address=\"10.3390/geosciences16020083\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDragičević N, Karleuša B, Ožanić N (2018) Modification of erosion potential method using climate and land cover parameters. Geomat Nat Hazards Risk 9:1085\u0026ndash;1105. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/19475705.2018.1496483\u003c/span\u003e\u003cspan address=\"10.1080/19475705.2018.1496483\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurlević U, Ilić V, Valjarević A (2025) Wildfire susceptibility mapping using deep learning and machine learning models based on multi-sensor satellite data fusion: A case study of Serbia. Fire 8:407. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/fire8100407\u003c/span\u003e\u003cspan address=\"10.3390/fire8100407\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurlević U, Momčilović A, Ćirić V, Dragojević M (2019) GIS application in analysis of erosion intensity in the Vlasina River Basin. Glasn Srp Geogr Drust 99(2):17\u0026ndash;36. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/GSGD1902017D\u003c/span\u003e\u003cspan address=\"10.2298/GSGD1902017D\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurlević U, Srejić T, Valjarević A, Aleksova B, Deđanski V, Vujović F, Lukić T (2025) GIS-based spatial modeling of soil erosion and wildfire susceptibility using VIIRS and Sentinel-2 data: A case study of Šar Mountains National Park, Serbia. Forests 16:484. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/f16030484\u003c/span\u003e\u003cspan address=\"10.3390/f16030484\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurlević U, Valjarević A, Novković I, Vujović F, Josifov N, Krušić J, Komac B, Djekić T, Singh SK, Jović G et al (2024) Universal snow avalanche modeling index based on SAFI\u0026ndash;Flow-R approach in poorly-gauged regions. ISPRS Int J Geo-Inf 13:315. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijgi13090315\u003c/span\u003e\u003cspan address=\"10.3390/ijgi13090315\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEfthimiou N, Lykoudi E, Karavitis C (2017) Comparative analysis of sediment yield estimations using different empirical soil erosion models. Hydrol Sci J 62:2674\u0026ndash;2694. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/02626667.2017.1404068\u003c/span\u003e\u003cspan address=\"10.1080/02626667.2017.1404068\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEfthimiou N, Lykoudi E, Panagoulia D, Karavitis C (2016) Assessment of soil susceptibility to erosion using the EPM and RUSLE models: The case of Venetikos River catchment. Glob Nest J 18:164\u0026ndash;179. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.30955/gnj.001847\u003c/span\u003e\u003cspan address=\"10.30955/gnj.001847\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEnvironmental Systems Research Institute (2024) Sentinel-2 10 m land use/land cover time series (2024). Esri Living Atlas. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://livingatlas.arcgis.com/landcoverexplorer/\u003c/span\u003e\u003cspan address=\"https://livingatlas.arcgis.com/landcoverexplorer/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 9 Mar 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEstrada-Carmona N, Harper EB, DeClerck F, Fremier AK (2017) Quantifying model uncertainty to improve watershed-level ecosystem service quantification: A global sensitivity analysis of the RUSLE. Int J Biodivers Sci Ecosyst Serv Manag 13:40\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/21513732.2016.1237383\u003c/span\u003e\u003cspan address=\"10.1080/21513732.2016.1237383\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEuropean Space Agency (2024) Copernicus global digital elevation model. Distributed by OpenTopography. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5069/G9028PQB\u003c/span\u003e\u003cspan address=\"10.5069/G9028PQB\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 2 Mar 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFern\u0026aacute;ndez G, Merch\u0026aacute;n L, S\u0026aacute;nchez J\u0026Aacute; (2025) Spatial representation of soil erosion and vegetation affected by a forest fire in the Sierra de Francia (Spain) using RUSLE and NDVI. Land 14:793. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land14040793\u003c/span\u003e\u003cspan address=\"10.3390/land14040793\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGajić Protić A, Krunić N, Protić B (2024) Detecting Serbia\u0026rsquo;s settlement patterns: A fuzzy logic-based approach to rural\u0026ndash;urban area delimitation for spatial planning. Land 13:1981. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land13121981\u003c/span\u003e\u003cspan address=\"10.3390/land13121981\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGajić A, Krunić N, Protić B (2021) Classification of rural areas in Serbia: Framework and implications for spatial planning. Sustainability 13:1596. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su13041596\u003c/span\u003e\u003cspan address=\"10.3390/su13041596\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGavrilović S (1972) Inženjering o bujičnim tokovima i eroziji [Engineering of torrents and erosion]. J Constr (Spec Issue) :1\u0026ndash;292 (In Serbian)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeoSrbija (2026) Geoportal of Serbia\u0026mdash;Cartographic data viewer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://a3.geosrbija.rs\u003c/span\u003e\u003cspan address=\"https://a3.geosrbija.rs\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 9 Mar 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGocić M, Dragićević S, Radivojević A, Martić Bursać N, Stričević L, Đorđević M (2020) Changes in soil erosion intensity caused by land use and demographic changes in the Jablanica River Basin, Serbia. Agriculture 10:345. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriculture10080345\u003c/span\u003e\u003cspan address=\"10.3390/agriculture10080345\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGolijanin J, Nikolić G, Valjarević A, Ivanović R, Tunguz V, Bojić S, Grmuša M, Lukić Tanović M, Perić M, Hrelja E, Stankov S (2022) Estimation of potential soil erosion reduction using GIS-based RUSLE under different land cover management models: A case study of Pale Municipality, B\u0026amp;H. Front Environ Sci 10:945789. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fenvs.2022.945789\u003c/span\u003e\u003cspan address=\"10.3389/fenvs.2022.945789\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonzalez-Ollauri A, Mickovski SB (2020) The effect of willow (Salix sp.) on soil moisture and matric suction at a slope scale. Sustainability 12:9789. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su12239789\u003c/span\u003e\u003cspan address=\"10.3390/su12239789\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodrich DC, Burns IS, Unkrich CL, Semmens DJ, Guertin DP, Hernandez M, Yatheendradas S, Kennedy JR, Levick LR (2012) KINEROS2/AGWA: Model use, calibration, and validation. Trans ASABE 55:1561\u0026ndash;1574. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.13031/2013.42264\u003c/span\u003e\u003cspan address=\"10.13031/2013.42264\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrčić M, Sibinović M, Ratkaj I (2024) The entropy as a parameter of demographic dynamics: Case study of the population of Serbia. Acta Geogr Slov 64:23\u0026ndash;39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3986/AGS.11441\u003c/span\u003e\u003cspan address=\"10.3986/AGS.11441\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaffas K, Hrissanthou V (2018) Soil erosion, streambed deposition and streambed erosion\u0026mdash;Assessment at the mountainous terrain. Proc 2:626. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/proceedings2110626\u003c/span\u003e\u003cspan address=\"10.3390/proceedings2110626\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKosmas C, Gerontidis St, Marathianou M (2000) The effect of land use change on soils and vegetation over various lithological formations on Lesvos (Greece). CATENA 40:51\u0026ndash;68. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0341-8162(99)00064-8\u003c/span\u003e\u003cspan address=\"10.1016/S0341-8162(99)00064-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKostadinov S, Braunović S, Dragićević S, Zlatić M, Dragović N, Rakonjac N (2018) Effects of erosion control works: Case study\u0026mdash;Grdelica Gorge, the South Morava River (Serbia). Water 10:1094. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/w10081094\u003c/span\u003e\u003cspan address=\"10.3390/w10081094\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKovačević J, Cvijetinović Ž, Lakušić D, Kuzmanović N, Šinžar-Sekulić J, Mitrović M, Stančić N, Brodić N, Mihajlović D (2020) Spatio-temporal classification framework for mapping woody vegetation from multi-temporal Sentinel-2 imagery. Remote Sens 12:2845. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs12172845\u003c/span\u003e\u003cspan address=\"10.3390/rs12172845\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLazarević R (1983) Soil erosion map of SR Serbia 1:500,000\u0026ndash;Interpretation. Institute for Forestry and Wood Industry, Belgrade, Serbia. (In Serbian)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLazarević R (1985) A new method for determining the erosion coefficient (Z). Erozija \u0026ndash; Stručno-informativni bilten 13:53\u0026ndash;61 (In Serbian)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLovrić N, Tošić R (2018) Assessment of soil erosion and sediment yield using erosion potential method: Case study\u0026mdash;Vrbas River Basin (B\u0026amp;H). Glasn Srp Geogr Drust 98(1):1\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/GSGD180215002L\u003c/span\u003e\u003cspan address=\"10.2298/GSGD180215002L\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLukić S, Baumgertel A, Obradović S, Kadović R, Beloica J, Pantić D, Belanović Simić S (2022) Assessment of land sensitivity to degradation using the MEDALUS model\u0026mdash;A case study of Grdelica Gorge and Vranjska Valley (Southeastern Serbia). iForest 15:163\u0026ndash;170. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3832/ifor3871-015\u003c/span\u003e\u003cspan address=\"10.3832/ifor3871-015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalušević I, Ristić R, Radić B, Polovina S, Milčanović V, Nešković P (2025) A historical overview of methods for the estimation of erosion processes on the territory of the Republic of Serbia. Land 14:405. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land14020405\u003c/span\u003e\u003cspan address=\"10.3390/land14020405\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManojlović S, Antić M, Šantić D, Sibinović M, Carević I, Srejić T (2018) Anthropogenic impact on erosion intensity: Case study of rural areas of Pirot and Dimitrovgrad municipalities. Serbia Sustain 10:826. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su10030826\u003c/span\u003e\u003cspan address=\"10.3390/su10030826\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManojlović S, Sibinović M, Srejić T, Novković I, Milošević MV, Gatarić D, Carević I, Batoćanin N (2022) Factors controlling the change of soil erosion intensity in mountain watersheds in Serbia. Front Environ Sci 10:888901. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fenvs.2022.888901\u003c/span\u003e\u003cspan address=\"10.3389/fenvs.2022.888901\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartinović M, Ratkaj I (2015) Sustainable rural development in Serbia: Towards a quantitative typology of rural areas. Carpath J Earth Environ Sci 10:37\u0026ndash;48\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaruna M, Crnčević T, Milojević MP (2019) The institutional structure of land use planning for urban forest protection in the post-socialist transition environment: Serbian experiences. Forests 10:560. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/f10070560\u003c/span\u003e\u003cspan address=\"10.3390/f10070560\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilevski I, Aleksova B, Lukić T, Dragićević S, Valjarević A (2024) Multi-hazard modeling of erosion and landslide susceptibility at the national scale in the example of North Macedonia. Open Geosci 16:20220718. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1515/geo-2022-0718\u003c/span\u003e\u003cspan address=\"10.1515/geo-2022-0718\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilovanović B, Ducić V, Radovanović M, Milivojević M (2017) Climate regionalization of Serbia according to the K\u0026ouml;ppen climate classification. J Geogr Inst Jovan Cvijic SASA 67(2):103\u0026ndash;114. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/IJGI1702103M\u003c/span\u003e\u003cspan address=\"10.2298/IJGI1702103M\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMilovanović B, Schuster P, Radovanović M, Ristić Vakanjac V, Schneider C, Milivojević M (2018) Spatial\u0026ndash;temporal variability of air temperatures in Serbia in the period 1961\u0026ndash;2010. J Geogr Inst Jovan Cvijic SASA 68(2):157\u0026ndash;175. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/IJGI1802157M\u003c/span\u003e\u003cspan address=\"10.2298/IJGI1802157M\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinistry of Mining and Energy of the Republic of Serbia (2026) GeoLISS: Geological information system of Serbia\u0026mdash;Geological map viewer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://geoliss.mre.gov.rs/vebkarte/geo300.html\u003c/span\u003e\u003cspan address=\"https://geoliss.mre.gov.rs/vebkarte/geo300.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 8 Mar 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicosia A, Carollo FG, Di Stefano C, Palmeri V, Pampalone V, Serio MA, Bagarello V, Ferro V (2024) The importance of measuring soil erosion by water at the field scale: A review. Water 16:3427. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/w16233427\u003c/span\u003e\u003cspan address=\"10.3390/w16233427\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOpenStreetMap Team (2026) OpenStreetMap export. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.openstreetmap.org/#map=7/44.226/15.562\u003c/span\u003e\u003cspan address=\"https://www.openstreetmap.org/#map=7/44.226/15.562\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 13 Feb 2026\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOro V, Stanisavljevic R, Nikolic B, Tabakovic M, Secanski M, Tosi S (2021) Diversity of mycobiota associated with the cereal cyst nematode Heterodera filipjevi originating from some localities of the Pannonian Plain in Serbia. Biology 10:283. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/biology10040283\u003c/span\u003e\u003cspan address=\"10.3390/biology10040283\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOshunsanya SO, Yu H, Odebode AM, Edem ID, Oluwatuyi TS, Imasuen EE, Odeyinka DE (2025) Comparative assessment of fractional and erosion plot methods for quantifying soil erosion and nutrient loss under vetiver grass technology on two contrasting slopes in rainforest agroecology. Agriculture 15:1762. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriculture15161762\u003c/span\u003e\u003cspan address=\"10.3390/agriculture15161762\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePanagos P, Borrelli P, Poesen J, Ballabio C, Lugato E, Meusburger K, Montanarella L, Alewell C (2015) The new assessment of soil loss by water erosion in Europe. Environ Sci Policy 54:438\u0026ndash;447. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.envsci.2015.08.012\u003c/span\u003e\u003cspan address=\"10.1016/j.envsci.2015.08.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePanagos P, Ballabio C, Poesen J, Lugato E, Scarpa S, Montanarella L, Borrelli P (2020) A soil erosion indicator for supporting agricultural, environmental and climate policies in the European Union. Remote Sens 12:1365. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs12091365\u003c/span\u003e\u003cspan address=\"10.3390/rs12091365\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePanagos P, Borrelli P, Matthews F, Liakos L, Bezak N, Diodato N, Ballabio C (2022) Global rainfall erosivity projections for 2050 and 2070. J Hydrol 610:127865. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jhydrol.2022.127865\u003c/span\u003e\u003cspan address=\"10.1016/j.jhydrol.2022.127865\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePantić M, Maričić T, Milijić S (2024) Visualising the relevance of climate change for spatial planning by the example of Serbia. Appl Sci 14:1530. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/app14041530\u003c/span\u003e\u003cspan address=\"10.3390/app14041530\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePapageorgiou N, Hadjimitsis D, Danezis C, Lasaponara R (2025) Assessment of soil erosion risk in cultural heritage sites: A bibliometric analysis. Heritage 8:307. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/heritage8080307\u003c/span\u003e\u003cspan address=\"10.3390/heritage8080307\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerovic V, Jaramaz D, Zivotić Lj, Cakmak D, Mrvic V, Milanovic M, Saljnikov E (2016) Design and implementation of WebGIS technologies in evaluation of erosion intensity in the municipality of Niš (Serbia). Environ Earth Sci 75:211. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12665-015-4857-x\u003c/span\u003e\u003cspan address=\"10.1007/s12665-015-4857-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePetrović AM, Manojlović S, Srejić T, Zlatanović N (2024) Insights into land-use and demographical changes: Runoff and erosion modifications in the highlands of Serbia. Land 13:1342. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land13091342\u003c/span\u003e\u003cspan address=\"10.3390/land13091342\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePinto CC, Calazans GM, Oliveira SC (2019) Assessment of spatial variations in the surface water quality of the Velhas River Basin, Brazil, using multivariate statistical analysis and nonparametric statistics. Environ Monit Assess 191:164. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10661-019-7281-y\u003c/span\u003e\u003cspan address=\"10.1007/s10661-019-7281-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePolovina S, Radić B, Ristić R, Milčanović V (2024) Application of remote sensing for identifying soil erosion processes on a regional scale: An innovative approach to enhance the erosion potential model. Remote Sens 16:2390. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs16132390\u003c/span\u003e\u003cspan address=\"10.3390/rs16132390\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePrăvălie R, Borrelli P, Panagos P, Ballabio C, Lugato E, Chappell A, Miguez-Macho G, Maggi F, Peng J, Niculiță M, Roșca B, Patriche C, Dumitrașcu M, Bandoc G, Niță I-A, B\u0026icirc;rsan M-V (2024) A unifying modelling of multiple land degradation pathways in Europe. Nat Commun 15:3862. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-024-48252-x\u003c/span\u003e\u003cspan address=\"10.1038/s41467-024-48252-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQGIS Development Team (2026) QGIS geographic information system v3.40.09 with GRASS [software]. Open Source Geospatial Foundation Project. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://qgis.osgeo.org\u003c/span\u003e\u003cspan address=\"http://qgis.osgeo.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 30 Jun 2025\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRao AU, Sabhahit N, Ananda LU, Bhandary RP (2026) Role of soil erosion in instability of slopes along coastal Karnataka. Geotechnics 6:21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/geotechnics6010021\u003c/span\u003e\u003cspan address=\"10.3390/geotechnics6010021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRatknić M, Braunović S, Rakonjac Lj, Ratknić T (2025) The afforestation strategy of the Republic of Serbia. Belgrade, Serbia\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaux J, Copard Y, Laignel B, Fournier M, Massei N (2011) Classification of worldwide drainage basins through the multivariate analysis of variables controlling their hydrosedimentary response. Glob Planet Change 76:117\u0026ndash;127. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gloplacha.2010.12.005\u003c/span\u003e\u003cspan address=\"10.1016/j.gloplacha.2010.12.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaza A, Ahrends H, Habib-Ur-Rahman M, Gaiser T (2021) Modeling approaches to assess soil erosion by water at the field scale with special emphasis on heterogeneity of soils and crops. Land 10:422. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/land10040422\u003c/span\u003e\u003cspan address=\"10.3390/land10040422\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRistić R, Kostadinov S, Abolmasov B, Dragićević S, Trivan G, Radić B, Trifunović M, Radosavljević Z (2012) Torrential floods and town and country planning in Serbia. Nat Hazards Earth Syst Sci 12:23\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/nhess-12-23-2012\u003c/span\u003e\u003cspan address=\"10.5194/nhess-12-23-2012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmanov Z, Duisenbayev S, Zulpykharov K, Laiskhanov S, Turymtayev Z, Kozhayev Z, Taukebayev O (2025) Soil salinization and its impact on the degradation of agricultural landscapes of the Talas District, Kazakhstan. J Geogr Inst Jovan Cvijic SASA 75(2):233\u0026ndash;250. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2298/IJGI2502233S\u003c/span\u003e\u003cspan address=\"10.2298/IJGI2502233S\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpalevic V, Barovic G, Vujacic D, Curovic M, Behzadfar M, Djurovic N, Dudic B, Billi P (2020) The impact of land use changes on soil erosion in the River Basin of Miocki Potok, Montenegro. Water 12:2973. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/w12112973\u003c/span\u003e\u003cspan address=\"10.3390/w12112973\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSrejić T, Manojlović S, Sibinović M, Bajat B, Novković I, Milošević MV, Carević I, Todosijević M, Sedlak MG (2023) Agricultural land use changes as a driving force of soil erosion in the Velika Morava River Basin, Serbia. Agriculture 13:778. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriculture13040778\u003c/span\u003e\u003cspan address=\"10.3390/agriculture13040778\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSrejić T, Manojlović S, Sibinović M, Lukić T, Kričković E, Durlević U (2025) Impact of the agri-geographical transformation of rural settlements on the geospatial dynamics of soil erosion intensity in municipalities of Central Serbia. Open Geosci 17:20250857. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1515/geo-2025-0857\u003c/span\u003e\u003cspan address=\"10.1515/geo-2025-0857\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStanišić M, Lovrić M, Nedeljković J, Nonić D, Pezdevšek Malovrh Š (2021) Climate change governance in forestry and nature conservation in selected forest regions in Serbia: Stakeholders classification and collaboration. Forests 12:709. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/f12060709\u003c/span\u003e\u003cspan address=\"10.3390/f12060709\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStefanidis S, Stathis D (2018) Effect of climate change on soil erosion in a mountainous Mediterranean catchment (Central Pindus, Greece). Water 10:1469. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/w10101469\u003c/span\u003e\u003cspan address=\"10.3390/w10101469\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStefanidis SP, Proutsos ND, Tigkas D, Chatzichristaki C (2025) Erosion-based classification of mountainous watersheds in Greece: A geospatial approach. Sustainability 17:8710. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/su17198710\u003c/span\u003e\u003cspan address=\"10.3390/su17198710\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStojković S, Marković D, Durlević U (2023) Snow cover estimation using Sentinel-2 high spatial resolution data: A case study\u0026mdash;National Park Šar Planina (Serbia). In: Ademović N, Mujčić E, Mulić M, Kevrić J, Akšamija Z (eds) Advanced Technologies, Systems, and Applications VII. Lecture Notes in Networks and Systems, vol 539. Springer, Cham, Switzerland, pp 389\u0026ndash;399. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-031-17697-5_39\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-17697-5_39\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSukmawijaya A, Akber MA, Wang Z, Azizan FA, Bell M, Abdul Aziz A (2026) Spatial assessment of water balance and soil erosion under land-use change in Chieng Hac, Northern Vietnam. Remote Sens 18:998. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs18070998\u003c/span\u003e\u003cspan address=\"10.3390/rs18070998\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eT\u0026ouml;lgyesi C, B\u0026aacute;tori Z, Erdős L (2014) Using statistical tests on relative ecological indicator values to compare vegetation units\u0026mdash;Different approaches and weighting methods. Ecol Indic 36:441\u0026ndash;446. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecolind.2013.09.002\u003c/span\u003e\u003cspan address=\"10.1016/j.ecolind.2013.09.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTošić I, da Silva ASA, Filipović L, Tošić M, Lazić I, Putniković S, Stosic T, Stosic B, Djurdjević V (2025a) Trends of extreme precipitation events in Serbia under global warming. Atmosphere 16:436. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/atmos16040436\u003c/span\u003e\u003cspan address=\"10.3390/atmos16040436\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTošić R, Dragićević S, Lovrić N (2012) Assessment of soil erosion and sediment yield changes using the erosion potential model. Carpath J Earth Environ Sci 7:147\u0026ndash;154\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTošić R, Dragićević S, Lovrić N, Langović M (2025b) Mapping soil erosion intensity using erosion potential method (EPM): Case study\u0026mdash;Bosnia and Herzegovina. Carpath J Earth Environ Sci 58:413\u0026ndash;426. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.26471/cjees/2025/020/343\u003c/span\u003e\u003cspan address=\"10.26471/cjees/2025/020/343\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrendafilov B, Minchev I, Trendafilov A, Blinkov I (2024) Comparison of EPM with RUSLE for soil erosion modeling in the Strumica River Basin. Geogr Environ Sustain 17:44\u0026ndash;49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.24057/2071-9388-2024-0580\u003c/span\u003e\u003cspan address=\"10.24057/2071-9388-2024-0580\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUnited Nations Convention to Combat Desertification (2022) Global land outlook, 2nd edn. Land restoration for recovery and resilience. UNCCD, Bonn, Germany\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUrošev M, Kovačević-Majkić Š, Štrbac D, Milanović Pešić A, Milijašević D, Jakovljević D, Petrović A (2017) Waters of Serbia. In: Radovanović M (ed) Relief of Serbia. Geographical Institute Jovan Cvijić SASA, Belgrade, Serbia, pp 161\u0026ndash;234. (In Serbian)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang J, Wang X, Yan Y, Wang L, Hu H, Ma B, Zhou H, Liu J, Gan F, Fan Y (2025) Peak soil erosion risk in mixed forests: A critical transition phase driven by Moso bamboo expansion. Agriculture 15:1772. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriculture15161772\u003c/span\u003e\u003cspan address=\"10.3390/agriculture15161772\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWani AA (2024) Comprehensive analysis of clustering algorithms: Exploring limitations and innovative solutions. PeerJ Comput Sci 10:e2286. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7717/peerj-cs.2286\u003c/span\u003e\u003cspan address=\"10.7717/peerj-cs.2286\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu D, Tian Y (2015) A comprehensive survey of clustering algorithms. Ann Data Sci 2:165\u0026ndash;193. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s40745-015-0040-1\u003c/span\u003e\u003cspan address=\"10.1007/s40745-015-0040-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang A, Zhang L, Zhang S, Zhan Z, Shi J (2022) Research on the temporal and spatial characteristics, spatial clustering and governance strategies of carbon emissions in cities of Shandong. Front Environ Sci 10:1024122. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fenvs.2022.1024122\u003c/span\u003e\u003cspan address=\"10.3389/fenvs.2022.1024122\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eŽivotić L, Vuković Vimić A (2022) Soil degradation and climate change in Serbia. Belgrade, Serbia\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"1447d7b5-010b-407f-82b2-5957c493caf5","identifier":"10.13039/501100004564","name":"Ministarstvo Prosvete, Nauke i Tehnološkog Razvoja","awardNumber":"451-03-33/2026-03/200172 and 451-03-33/2026-03/200091","order_by":0}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Geographical Institute “Jovan Cvijić”, Serbian Academy of Sciences and Arts","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":"Geosciences, Sentinel-2, Land use, Agricultural land, Erosion potential model (EPM), Agglomerative Hierarchical Clustering (AHC)","lastPublishedDoi":"10.21203/rs.3.rs-9569900/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9569900/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSoil erosion represents a complex geomorphological and geological process that negatively affects the environment, the quality of natural resources, and the safety of the population. In this study, a mechanical soil erosion map for the territory of Serbia was produced using the Erosion Potential Model (EPM), combined with remote sensing data (Sentinel-2) and Geographic Information Systems (GIS). The analysis used contemporary geospatial data on lithology, land use, and terrain slope, with a 30 m spatial resolution. The average erosion intensity at the national level is 0.239, corresponding to the weak erosion class. More than half of the territory (53.8%) is affected by very weak erosion, while 24.2% is characterized by weak erosion intensity. Additionally, 12.7% of Serbia falls within the medium erosion intensity class. The results further indicate that 7.6% of the territory is affected by intensive erosion, while 0.7% is exposed to excessive erosion. Approximately 1% of the territory is exposed to sediment accumulation. Multivariate analysis of geographical conditions showed that the highest values of the erosion coefficient (Z) were determined by land use (r = 0.826). In contrast, the lowest values were associated with terrain slope (r = −0.805). In addition to the national-scale assessment, spatial differentiation of the results was performed at the local (municipal) level. The most susceptible municipalities were identified and analyzed, as well as those characterized by specific natural and anthropogenic conditions. Municipalities were subsequently grouped into six clusters based on six indicators using Agglomerative Hierarchical Clustering (AHC), according to their susceptibility to mechanical erosion. The results were validated using the ROC-AUC metric, which showed perfect predictive power (100%). This study significantly contributes to decision-making at both national and local levels by providing a scientific basis for developing strategies for sustainable forest management and soil conservation, highlighting the important role of forests in reducing soil erosion and maintaining ecosystem stability.\u003c/p\u003e","manuscriptTitle":"GIS-Based Soil Erosion Susceptibility Mapping in Serbia Using the EPM Model and Satellite Remote Sensing: A National-Scale Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-06 18:00:21","doi":"10.21203/rs.3.rs-9569900/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":"fd5ad68e-183b-4426-8405-0dc261fb9fb0","owner":[],"postedDate":"May 6th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":67272673,"name":"Geomorphology"},{"id":67272674,"name":"Geographic Information Systems"}],"tags":[],"updatedAt":"2026-05-06T18:00:21+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-06 18:00:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9569900","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9569900","identity":"rs-9569900","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}