Ratan Pal ¹*; Narayan C. Jana ¹

1, Department of Geography, The University of Burdwan, Burdwan, W.B. -713104, India

 

Manuscript link
http://dx.doi.org/10.30493/DAS.2025.510627

Abstract

This research employs the Revised Universal Soil Loss Equation (RUSLE) and Frequency Ratio (FR) models to estimate current soil erosion rates and project future erosion areas in the Silabati River Basin. The average rate of soil erosion was estimated at 4.35 tons/ha/year, significantly below its maximum permissible rate (T) of 12.5 tons/ha/year. However, 12.03% of study area exceeded this threshold. Nineteen soil erosion conditioning factors were taken into account to evaluate soil erosion susceptibility including Rainfall, Elevation, Slope of surface, Aspect of the slope, Topographic curvature, Lithology, Geomorphology, percent of Sand, Silt, Clay and Soil Organic Carbon, Land Use/Land Cover, Vegetation Concentration, Drainage density, Topographic Wetness Index, Terrain Roughness Index, Stream Power Index, and Distance to roads and rivers. High erosion-susceptible locations tend to be found in the upper river basin, where high elevation, steep slopes, high sand and silt content, low clay, and soil organic carbon concentration make it particularly prone to erosion. Moderately vulnerable areas include locales near rivers and lower basin regions with steep slopes and agricultural activities that occupy most of the terrain. The central regions are characterized by dense vegetation that provides erosion protection and are the least vulnerable to erosion. The Receiver Operating Characteristic (ROC) curve was utilized as a validation technique of erosion susceptibility model. The Area Under Curve (AUC) value of 87.40% underscores the model’s accuracy in classifying erosion-susceptible zones within the studied area.

Keywords: Soil erosion, RUSLE, Frequency ratio, Erosion susceptibility

Introduction

Despite the considerable advancements in technology and societal structures, it remains evident that human civilization continues to depend fundamentally on nature and its resources for sustenance. Given the understanding that land degradation caused by human activity and natural causes adversely affects the population, India has established an objective to attain land degradation neutrality over an area exceeding 26 million hectares by the year 2030 [1]. In light of escalating land pressures and evolving land use patterns, it is imperative to assess the risk of soil erosion for the effective management of natural resources, safeguarding the environment, and promoting sustainable practices [2-5]. In this context, the efficiency of GIS in handling large datasets, conducting intricate spatial analyses, and presenting information has significantly transformed the field of hazard studies, with soil erosion being a notable example of this advancement. The incorporation of Geographic Information Systems in hazard studies enhances the capacity for predicting and monitoring environmental issues, ultimately leading to improved land use management practices [6].

Water erosion is a process wherein particles from the Earth’s surface are dislodged by the impact of raindrops or surface runoff and subsequently carried away. This phenomenon primarily results from the direct effects of precipitation and surface water movement [7]. Soil erosion, as an aspect of the evolution of natural landscapes, is a phenomenon recognized as geologic erosion and is, to a certain degree, unavoidable [8]. The gradual process plays a crucial role in the redistribution of soil nutrients. Nonetheless, the issue arises from the transformation of the advantageous aspects of geological erosion into the detrimental effects of accelerated erosion. Soil erosion is a major contributor to global land degradation and has emerged as a critical issue owing to its profound effects on society [9]. The phenomenon of soil erosion results in an estimated depletion of around 74 million tons of vital nutrients from the surface of India’s soil annually, translating to a financial impact of approximately 68 billion Indian rupees (INR) each year [10][11]. This phenomenon occurs when the intrinsic rate of soil erosion accelerates significantly as a result of anthropogenic activities. This acceleration results in a significant increase in land degradation [12][13] by altering the physical, chemical, and organic properties of soil, which in turn reduces agricultural productivity due to the erosion of the productive soil layer from exposed rocks [14], deteriorates water quality, disrupts ecosystems, undermines food security, heightens vulnerability to natural disasters, and encompasses numerous additional consequences [15][16].

Human activities, such as transforming forests for agricultural purposes and inadequate land management practices like overgrazing and expanding cultivated areas, have led to soil erosion rates that are 10 to 100 times greater than their natural rates [1]. As a result of inadequate land use practices, the situation may become increasingly severe in the future [17]. According to Steffen et al. [18], there has been a significant reduction of 38% in global forest areas since the Holocene, which has critically impacted contemporary society [19]. More than 30% of the Earth’s surface has experienced degradation, with soil erosion identified as the primary contributing factor [20][21]. The estimates provided by the United Nations indicate that approximately 20% of the Earth’s land surface experienced degradation over a span of 15 years, from 2000 to 2015 [22], highlighting a significant rate of land degradation in contemporary times.

The National Commission on Agriculture estimated that around 150 million hectares of land in India are subject to soil erosion [23]. However, a later investigation conducted by the National Bureau of Soil Survey & Land Use Planning (NBSS & LUP) revised this figure considerably downward to 119.19 million hectares [24]. Narayan and Babu [25] calculated an average soil erosion rate in India of 16.35 t/ha/yr. As of 2011, the Centre Soil and Water Conservation, Research, and Training Institute (CSWCRTI) estimated that around 68.4% of the land erosion in India was attributed to water-induced factors. In West Bengal, out of a total land area of 8.87 million hectares, approximately 2.2 million hectares, which constitutes around 24.80%, is anticipated to be subject to degradation. Water erosion represents a critical concern, accounting for 54% of this degradation. The considerable population density within the state exerts significant pressure on its constrained land resources, as approximately 75% of the populace possesses an average of merely 0.16 hectares per individual [26].

A variety of methodologies exist for evaluating soil erosion and forecasting its occurrence under diverse physical and environmental circumstances, as well as different management strategies. Notable examples include the Universal Soil Loss Equation (USLE) [27][28], Kinematic Runoff and Erosion Model [29], Process-Based Erosion Model [30], Revised USLE (RUSLE) [31], Water Erosion Prediction Project [32], European Soil Erosion Model [33][34], Soil and Water Assessment Tool [35], Modified Universal Soil Loss Equation [36], Catchment Scale Soil Erosion Model [37], Erosion Productivity Impact Calculator [38], Morgan-Morgan-Finney (MMF) Model [39], along with its various adaptations such as Modified MMF [40], Extended MMF [40], Enhanced MMF [41], Sediment Assessment Tool for Effective Erosion Control [42], and Automated Geospatial Watershed Assessment [43], among others. This study employs the RUSLE, highlighting its straightforward nature, global acknowledgment, and relevance in tropical environments [4][16].

Understanding the rate of erosion in a given area offers significant insights into its present conditions, particularly in regions where agriculture is a crucial factor. In order to effectively address erosion and foster sustainable land use, it is essential to accurately assess both the present conditions and future possibilities. These activities contribute significantly to the preservation of agricultural productivity, the promotion of environmental sustainability, the safeguarding against land degradation, and the maintenance of topsoil nutrient levels. The reduction of sedimentation, alongside the preservation of ecosystem health through the prevention of habitat loss, constitutes significant advantages. Comprehending the processes of erosion is essential for ensuring the stability of infrastructure, managing flood risks, and addressing climate change through the reduction of carbon emissions originating from soil. Furthermore, it contributes to policy formulation and strategic planning, fosters sustainable development, and reduces the economic burdens linked to land restoration and diminished agricultural productivity. This research provides valuable insights for planners and policymakers aiming to mitigate land degradation while effectively managing all land resources within the study area. Therefore, this research aims to ascertain the mean rate of soil erosion, delineate its spatial distribution, and pinpoint areas that may be susceptible to erosion events in the Silabati River Basin, a prominent agricultural region in West Bengal.

Material and Methods

Study area

The River Silabati originates at 23° 14′ 10” N latitude and 86° 38′ 37” E longitude in the Hura block of the Purulia district in West Bengal. The river flows largely southeast and meets the Dwarakeswar River at 22° 40′ 14” N and 87° 46′ 41” E near the city of Ghatal in the Paschim Medinipur district of West Bengal. It travels a total of around 217.28 kilometers from its source. The Silabati River’s catchment area (4011.70 km²) stretches out over three districts in West Bengal: Purulia (55.58 km²), Bankura (1389.18 km²), and Paschim Medinipur (2566.94 km²) [44]. The Silabati River Basin (Fig. 1) has a tropical monsoon climate, with an average rainfall of 1342.39 mm. The temperature ranges from 6°C in the winter to 50°C in the summer [45]. The Jaipanda, Purandar, Champakhali, Sasra, and Ketia Rivers are the main left tributaries of the Silabati River. The Tamal, Kubai, Buriganga, Parang, and Donai Rivers are the main right tributaries [46]. The Shilabati River is a winding stream with an exceptionally sinuous course. The river floods almost every year in the lower reaches of the basin, especially in the areas of Banka, Manik Kundu, Kaskuli, Gadghat, Haldar Bar, Bagpata, Mansatalar Chatal, Patna, Khirpai, Barada, Ranir Bajar, Dalpatipur, Jadupur, Kharar, Palpukur, Baropol, Khasbar, Mansuko, Ghatal, Daspur, and Kushpata [47]. The river basin is a place where the Chota Nagpur Plateau in the west and the lower Gangetic alluvial plain in the east meet [44].

Data acquisition and processing

The monthly average rainfall data spanning forty-one years (1981-2021) was obtained from the Prediction of Worldwide Energy Resources (POWER) [48] website, covering ten locations within and surrounding the study area. A total of one hundred sixty-one soil samples were systematically collected through random sampling (Fig. 1) and subsequently analyzed in the laboratory to obtain the necessary data for the current investigation. The Digital Elevation Model (DEM) derived from the Advanced Land Observing Satellite equipped with Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR) data serves as a basis for the extraction of elevation and slope-related information. The utilization of Landsat 8 satellite imagery facilitates the preparation of various maps, including NDVI, vegetation concentration, and land use/land cover (LULC) assessments. The acquired data underwent processing using Excel and ArcGIS 10.4. The specifics of the data are presented in Table 1.

Selection of models

The RUSLE

The Revised Universal Soil Loss Equation (RUSLE) is considered one of the most effective models for quantifying soil erosion rates in various environmental contexts. The foundational model was established as the Universal Soil Loss Equation (USLE) by Wischmeier and Smith [49]. The objective was to forecast the long-term average annual rates of soil erosion in agricultural fields subjected to various cultivation and management practices throughout the United States. In response to the evolving physio-climatic conditions globally, numerous adjustments were implemented to the model to improve its acceptability, with the RUSLE being one such revision, as detailed by Renard et al. [31]. The Revised Universal Soil Loss Equation remains extensively utilized across the globe, incorporating specific modifications to account for local climatic, topographical, and soil characteristics. The model was employed to determine the rate of soil erosion in India, incorporating equations tailored to account for the specific soil and environmental conditions.

The equation 1 mathematically represents the RUSLE.

A = R × K × LS × C × P                                                                                 (1)

where ‘A’ is the average annual rate of soil erosion, ‘R’ is the rainfall-runoff erosivity factor, ‘K’ is the soil erodibility factor, ‘LS’ is the slope length and steepness factor, ‘C’ is the cover-management factor, and ‘P’ is the support-practice factor.

The R factor is calculated using equation 2 [50].

R = 79 + 0.363P                                                                                           (2)

where ‘R’ is the average annual erosion index, and ‘P’ is the annual average rainfall in mm.

The K factor is calculated using the soil properties method as suggested by Williams [38] (Equation 3 – 7).

K = ƒ_csand × ƒ_ci-si × ƒ_orgc × ƒ_hisand                                                                  (3)

where ‘K’ is the soil erodibility factor, ‘ms’ is the % sand, ‘msilt’ is the % silt, ‘mc’ is the % clay, and ‘orgc’ is the % soil organic carbon.

Equation 8 is used to determine the LS factor [51].

where ‘LS’ is the Slope length and steepness factor, ‘A_S’is the Specific catchment area, m is 0.4 to 0.6, and ‘n’ is 1.2 to 1.3, θ is the slope in degree.

The C factor is estimated using equation 9, designed especially for tropical regions [52].

The P factor is derived from the LULC map of the study area and the P values of different LULC categories are adopted from the ‘United States Department of Agriculture-Soil Conversion Service [53][16].

The Frequency Ratio (FR)

Evaluating susceptibility using statistical models is an effective means of predicting susceptibility as it demonstrates the correlations between known instances and environmental influences [54]. Bonham-Carter [55] introduced the frequency ratio as an effective bivariate statistical measure to assess susceptibility evaluation. GIS integration makes the model suitable for spatial prediction and delineating high erosion-risk areas [56]. Numerous studies have used the model for susceptibility assessments including those on landslide susceptibility [57-59], flood susceptibility [60-62], groundwater susceptibility [63][64], gully erosion susceptibility [65-67], as well as soil erosion susceptibility [68-70].

To calculate the FR, we first determine the area ratio between soil erosion events and non-events in each variable and calculate its ratio against the total area. Next, the area ratios for each variable to the total area are computed, and then each class of factors’ association with soil erosion events is explored. FR value exceeding one indicates a strong positive association and below one signifies weak negative relations that lead to lower erosion susceptibility [70]. The FR is determined using equation 10.

where the number of soil erosion pixels in class ‘i’ of the factor ‘X’ is represented by ‘Npix (SXi)’; the total number of pixels within factor ‘Xj’ is represented by ‘Npix(Xj)’; ‘m’ is the number of classes in factor ‘Xi’; and ‘n’ is the total number of factors in the study area [66][58][71].

Validation of maps

Validating a soil erosion map is challenging due to the difficulty associated with determining the actual rate of soil erosion in any specific location using the plot method, which is costly, laborious, and time-consuming. Instead of employing such an expensive and time-consuming procedure, this study utilized an observation method with local opinions being taken into consideration to validate the prepared map of erosion. To get clear visible evidence of erosion, including the presence of rills and gullies and sediment deposition, the field survey was done from August to September 2024. The soil erosion map is converted into two categories e.g., erosion and non-erosion areas. To differentiate between soil erosion and non-erosion areas, the concept of the maximum permissible soil loss rate (‘T’ value) was used, and it is assumed that the soil erosion rate beyond the ‘T’ value negatively affects society. The ‘T’ value varies significantly due to factors such as soil quality, resistance to erosion, and soil depth at specific locations. According to Mondal and Sharda [72], in India, this value ranges from 2.5 to 12.5 tons per hectare per year (t/ha/yr ). In West Bengal, approximately 88% of the area, including the Silabati River basin, has a ‘T’ value of 12.5 t/ha/yr [26]. Consequently, 12.5 t/ha/yr is used as the threshold to distinguish between erosion and non-erosion areas. Places with a soil erosion rate exceeding 12.5 t/ha/yr are identified as erosion areas, while areas with rates up to 12.5 t/ha/yr are considered as non-erosion areas.

The Receiver Operating Characteristics (ROC) curve provides a useful means of evaluating and assessing the performance and accuracy of a predictive model, especially in classification tasks [73]. It measures the true rate of positive (sensitivity) to falsely high rates (1-specificity) at different thresholds [74]. The Area Under the ROC Curve (AUC) is used to measure a classifier’s ability to differentiate positive from negative classes. Fawcett [75] states that this curve ranges from 0.5 to 1.0; where 1 represents perfect classification while 0.5 denotes random guessing; and between 0.8-0.9 is considered good performance. Bandyopadhyay and Pani [76], and Chakraborty and Pal [77] used the ROC curve efficiently in soil erosion studies.

Preparation of soil erosion inventory map

Preparing a soil erosion inventory map is a crucial initial step in estimating the soil erosion susceptibility of any area [78][79]. To prepare the soil erosion inventory map, 400 correctly classified points from each erosion and non-erosion site (hence the total number of sample points was 800) were randomly selected. According to the assignments of binary values, the soil erosion points were labelled as ‘1’, and non-erosion points were labelled as ‘0’ [80]. Subsequently, the dataset was divided into training (70%) and testing (30%) sets.

Selection of conditioning factors for soil erosion susceptibility assessment

Rainfall

Soil erosion is directly affected by the intensity and duration of rainfall. Nearing et al. [81] examined how rainfall patterns and LULC changes affect soil erosion and found that rainfall was more sensitive to erosion than LULC changes. The rainfall map (Fig. 2 A) is prepared using the average annual rainfall data of forty-one years (1981 – 2021) using the Inverse Distance Weighted (IDW) method.

Relief and slope

The DEM data is utilized to produce elevation (Fig. 2 B), slope (Fig. 2 C), aspect of slope (Fig. 2 D), and topographic curvature (Fig. 2 E) maps. Altitude is integral to soil erosion since its relationship is directly proportional [15]. Steeper slopes increase surface runoff velocity, promoting more erosion processes [82]. As the slope increases, so does its ability to detach soil particles and transport them downslope. Gabarron-Galeote et al. [83] and Raouf et al. [84] have studied how aspects influence soil erosion. Perreault et al. [85] observed that slopes oriented in various directions are subject to varying levels of sunlight, moisture, and wind exposure. These factors can significantly influence vegetation cover, soil moisture content, and the dynamics of water flow across these terrains. The researchers observed the influence of land surface curvatures on soil erosion dynamics and the patterns of water movement across these surfaces. Cotton [86] conducted research into how convex and concave curves influence erosion; convex slopes may shed water quickly, potentially decreasing topside erosion while increasing down-sloping erosion rates; while concave slopes may collect water or sediment accumulation resulting in localized areas with increased erosion or deposition rates.

Land Use/Land Cover (LULC)

Land use intensity is a significant contributor to soil erosion. Studies by Sharma et al. [87]; Munoth and Goyal [88]; and Gashaw et al. [89] indicate an increased rate of soil erosion due to changes and land use intensification. The LULC map is prepared using the Maximum Likelihood Classification method from the Landsat 8 satellite image (Fig. 2 F).

Lithology and geomorphology

The underlying rock structure and its properties significantly influence soil formation, stability, and erodibility rendering it a vital factor in soil erosion study [68]. Kathwas and Patil [90] proved a significant correlation between soil loss and geomorphological landforms. Landform types are crucial in influencing soil erosion rates. The Lithology (Fig. 2 G) and Geomorphology (Fig. 2 H) maps are downloaded from the Geological Survey of India (GSI).

Vegetation concentration

The intensity of vegetation concentration (Fig. 2 I) is a crucial factor in soil erosion. Dense vegetation cover significantly reduces the speed of surface runoff, thereby preventing soil erosion in the area. Additionally, tree roots enhance soil structure stability [91]. In contrast, Ochoa-Cueva et al. [92] found that deforestation and impervious surfaces increase water flow and lead to soil loss. The vegetation concentration of the study area is derived from the NDVI using equation 11.

Where ‘NIR’ and ‘RED’ are the Near Infrared and Red bands of Landsat satellite images respectively.

Soil characteristics

Soil properties, especially sand (Fig. 2 J), silt (Fig. 2 K), clay (Fig. 2 L), and organic carbon (Fig. 2 M) ratios are crucial controlling factors of soil erosion. Higher silt and sand ratios result in loose-textured soil, leading to increased soil erosion. Conversely, clay and organic carbon increase the soil’s cohesive power, rendering it more erosion-resistant. The present study uses the sieving method to determine the sand, silt, and clay percentage. The soil organic carbon (SOC) is determined by following the titration method of Walkley and Black method [93] using equation 12.

Where ‘V’ is the volume of 1 N K₂Cr₂O₇​ solution, ‘B’ is the volume of the FAS reagent for the blank sample, ‘T’ is the volume of the FAS reagent for the soil sample, ‘m’ is the air-dry moisture percentage of the soil sample, ‘W’ is the weight of the soil sample (1 g.), RF is the recovery factor, calculated as 1/0.77 [16].

Terrain Roughness Index (TRI)

Surface irregularities immensely influence soil erosion [94]. Rough surfaces with lumps and irregularities help increase water infiltration while decreasing runoff by creating micro-depressions and channels that enable the water to percolate more easily into the soil. By contrast, smooth surfaces tend to form crusts due to fine soil particles being dispersed and compacted during rain showers, sealing against infiltration and leading to higher surface runoff rates and erosion rates. Bremenfeld et al. [95] and Tarolli et al. [96] emphasize maintaining surface roughness. The TRI (Fig. 2 N) is calculated using equation [97].

where ‘Z_x’ is the elevation of each neighbouring cell, ‘Z_c’ is the elevation of the central cell, and ‘n’ is the number of neighbouring cells.

Topographic Wetness Index (TWI)

The TWI indicates water availability in soil [98]. As a dynamic component, soil moisture frequently changes and varies due to precipitation, temperature, evapotranspiration, infiltration, and runoff [99][100]. Drier soils reduce runoff [101], leading to lower sediment transport than wetter soils and more runoff triggers soil erosion [102]. The TWI (Fig. 2 O) is calculated using the equation 14 [103].

where ‘α’ represents the upslope contributing area per unit contour length, and ‘β’ denotes the slope angle or gradient. The expression ‘tanβ’ describes the frequency distributions of slopes in the steepest downslope direction

Drainage Density (Dd)

The Dd is the length of the streams per unit area [104]. It is an important factor in soil erosion study as high drainage density is closely linked to rugged landscapes and rapid runoff, leading to severe floods and considerable soil erosion [105]. The drainage map is prepared by digitizing topographical maps and subsequently, the Dd map (Fig. 2 P) is prepared from the drainage map using equation 15.

where ‘Lu’ is the total length of the streams, and ‘A’ is the total area.

Stream Power Index (SPI)

The SPI is an essential measure to evaluate soil erosion potential [106] serving as an indicator of potential overland flow and associated erosion [107]. A higher SPI indicates a greater potential for soil erosion and vice versa [108]. The SPI (Fig. 2 Q) is calculated using the equation 16.

where ‘A_s’ is the catchment area and ‘β’ indicates the slope in degree.

Distance from rivers and roads

Panagos et al. [109] considered the proximity to rivers as a significant conditioning factor in soil erosion study. Road construction necessitates altering the terrain, leading to land instability [110]. These impervious surfaces, resulting from such construction, accelerate water flow, thereby exacerbating soil erosion in the surrounding areas [68]. Distance from the river (Fig. 2 R) and road (Fig. 2 S) maps were prepared using the Euclidean distance tool in ArcGIS 10.4 from the drainage network and roadwork maps, respectively. The road network map was downloaded from the Geosadak website and updated through digitization from the topographical maps.

Result and Discussion

Estimating soil erosion

The spatial distribution of the R factor (Fig. 3 A) closely follows rainfall trends; specifically showing that southern and southeastern regions experience the highest rainfall levels while gradually tapering off towards the northwest areas. R factor values ranged from 543.84 to 591.22 MJ.mm/(ha.h.yr), with a mean value of 543.84 MJ.mm/(ha.h.yr). The K factor (Fig. 3 B) ranged between 0.14 and 0.19 ton.ha.h/(MJ.mm), with an average value of 0.17 ton.ha.h/(MJ.mm). The LS factor (Fig. 3 C) varied from 0 to 30.26, with a mean of 0.68. There is an inverse correlation between the C factor and the NDVI. Areas with higher NDVI have lower C values, indicating less erosion due to vegetation’s protective effect and vice-versa. The mean C factor value was 0.39 and ranged between 0.25 to 0.59 (Fig 3 D). The P factor (Fig. 3 E) is responsible for the effect of practices to conserve soil. It assigns values based on various LULC categories, with lower values reflecting more efficiency in erosion control. The P factor value ranged between 0 to 1, with a mean value of 0.19.

After creating all the factors of the RUSLE, each factor map was adjusted to a uniform 30-meter cell size to warrant precision before combination. The resultant map of the soil erosion (Fig. 3 F) displays distinct patterns throughout the basin. High soil erosion is common within the higher reaches due to steep slopes, high elevation, sparse vegetation, higher concentration of silt and sand, lower clay content, and lower amounts of soil SOC. On the other hand, the middle reaches show the lowest rates of erosion due to the dense vegetation cover, which protects from erosion [91]. Lower reaches exhibit moderate erosion rates caused by low elevations, gentle slopes, and the higher concentration of clay content in the soil, reducing erosion, but are negatively affected by intensive farming practices, sparse vegetation, and the predominant presence of urbanization activities.

The results showed that approximately 43.72% of the Silabati River Basin area experiences a very low soil erosion rate (<1 t/ha/yr). In comparison, 28.53% and 9.42% of the area have low (1-5 t/ha/yr) and moderate (5-10 t/ha/yr) erosion rates, respectively. Areas with high (10-20 t/ha/yr) and very high (>20 t/ha/yr) rates of soil erosion account for about 12.01% and 6.32% of the basin, respectively (Table 2). The average rate of soil erosion across the basin is 4.35 t/ha/yr. Considering an erosion threshold of 12.5 t/ha/yr [26] the study area was classified into erosion and non-erosion areas (Fig. 4). Overall, 12.03% of the total area was considered erosion areas and utilized with the inventory map to validate the model.

Soil erosion conditioning factors and their relation with soil erosion susceptibility

Frequency ratio analysis (Table 3) clearly shows that the potential for water erosion of soil is greater with the increase in rainfall, elevation, and slope. The highest FR value of 1.14 was observed in the highest rainfall class, while the lowest FR value of 0.89 was found in the second lowest rainfall class. As for elevation, the lowest FR of 0.91 was observed in areas of lowest elevation areas of less than 40 m from mean sea level (MSL), while the highest (1.55) was recorded in the second highest category of 120 to 160 m from MSL. For slope, the lowest and highest FR of 0.91 and 1.38 were obtained in the lowest (<2˚) and highest (>6˚) classes, respectively. Among the geomorphology classes, the highest and lowest FR values of 4.19 and 0.06 are found in the lateral Bar and Upland (Lateritic), respectively. Other high FR values (>2) were observed in river, dam and reservoir, valley fill, residual hill, point bar, pediplain, and abandoned quarry.

Sand, Silt, and Gravel (Present Day Deposit) have the highest FR value (2.35) among the lithology classes. Two classes, namely Mica Schist / Schist, and Gabbro, scored the FR value of 0, as there are no pixels observed in the Soil Erosion Inventory Map (SEIM) (Fig. 4) corresponded with these lithological classes (Fig. 2 G); indicating these areas might be highly safe in terms of soil erosion. Unclassified Sediment, Clay and Grit, and Conglomerate and lateritic soil obtained low FR values of 0.27 and 0.69, respectively. The high concentration of sand and silt triggers soil erosion as reflected in the FR values. Conversely, with increases in clay and SOC contents, the FR values decrease. The lowest FR values of 0.81 and 0.74 are scored in the areas with the highest clay (>20%) and SOC (>1.0%) contents.

Among the LULC categories, the minimum FR value is held by vegetation cover (0.31), highlighting the minimum rate of soil erosion in this category. Built-up areas recorded a higher FR value of 0.76, while the highest FR values of 1.44 and 1.31 were observed in barren and agricultural lands, respectively. With more detailed inspection to vegetation density (concentration), the highest FR value of 1.13 is found in the area with no vegetation cover (NDVI<0.10), while in densely vegetated (NDVI>0.40) areas scored the lowest (FR 0.92).

When reporting FR values according to aspect of slopes, it is worth noting that the present study area is relatively flat with an elevation range of only 215 m. North and East facing slopes scored higher FR values of 1.07 and 1.01, respectively, while flat areas scored the lowest (0.87). Similarly, the Terrain Roughness Index (TRI) does not vary significantly. The convex topographic curvature with an FR value of 1.02 is more susceptible than concave (FR 1.00) and flat (FR 0.99) topographic curvature. With increasing TWI, the FR value also increased significantly, and the highest FR value of 1.41 was found in the highest TWI class (>12). As for Drainage density (Dd), the highest and lowest FR values of 1.80 and 0.82 are observed in the highest and lowest Dd classes, respectively. In areas of higher erosional power of the river, the highest FR value is observed, while it is lowest in areas of deposition. There is a decreasing trend of FR values as the distance from the road increases; however, the highest FR value (1.28) was found in the class furthest from roads (>4000 m).

Spatial distribution of soil erosion susceptibility

 Frequency ratio analysis was employed to visualize the regional distribution of soil erosion susceptibility and its associated statistics (Fig. 5 and Table 4). The areas categorized as very high and high susceptibility, comprising approximately 7.84% and 7.38% of the total area, respectively, were predominantly situated in the upper sections of the river basin. Multiple variables contribute to the elevated vulnerability in these areas, including significant elevation and slope, high levels of sand and silt, low levels of clay and soil organic carbon (SOC), and increased drainage density. Moreover, 25.05% of the entire basin was classified as moderate. Most of these areas are situated along riverbanks with steep gradients and in the lower reaches of the river basin, where agriculture is prevalent. The center regions predominantly exhibit low (46.32%) and extremely low (13.42%) susceptibility to soil erosion, attributed to a significant presence of vegetation that mitigates erosion risk.

Validation of prepared maps

The cumulative area designated as erosion locations within the research region is approximately 482.66 km², constituting 12.03% of the overall area. The empirical observations and perspectives of local residents are employed to corroborate the veracity of the developed soil erosion map. Initially, 400 locations were randomly chosen as erosion points from the designated soil erosion map indicated as erosion zones. Additionally, 400 points were randomly chosen as non-erosion points from areas classified as non-erosion. All 800 criteria were validated by visual inspection and professional assessments. Regions exhibiting definitive signs of erosion are classified as erosion areas, while those lacking such evidence are designated as non-erosion zones. Of the 400 erosion sites, 67 are misclassified, indicating they were identified as soil erosion regions on the map despite being non-erosion zones. Conversely, 46 sites were inaccurately categorized from the chosen non-erosion points. Of the total 800 points, 113 points are misclassified. The generated soil erosion map exhibits an accuracy of 85.88%, thereby rendering it suitable for subsequent analysis.

The performance of the soil erosion susceptibility model was assessed through ROC curve, resulting in an AUC value of 87.40% (Fig. 6), which indicate that the model is more likely to accurately classify a selected region according to its erosion susceptibility. This score is within the range deemed indicative of satisfactory model performance [75][76]. Consequently, the predictive accuracy is robust, signifying that the spatial modeling methodologies and parameters employed to create the susceptibility map effectively represent the patterns of soil erosion risk throughout the research area. Moreover, there was negligible overlap between true positives and false positives, indicating low false alarm rates that facilitate precise identification of vulnerable areas. This is particularly advantageous in practical applications such as land-use planning and conservation efforts, where accurately predicting susceptible regions enables effective resource allocation decisions.

Conclusion

This research employed RUSLE to forecast soil erosion rates and utilized the statistic-based probabilistic FR to pinpoint regions susceptible to soil erosion.  The average soil erosion rate in the Silabati River Basin is calculated to be 4.35 t/ha/yr, which is below the allowable soil loss rate established for the research region.  Simultaneously, about one-eighth of the Silabati River Basin had soil erosion rates that above the permissible threshold, indicating a concerning predicament.  Vigilance is essential in these areas, and the ongoing implementation of solutions to mitigate soil erosion is imperative.

Acknowledgments

The authors are thankful to NASA Prediction of Worldwide Energy Resources (POWER), Alaska Satellite Facility (ASF), United States Geological Survey (USGS), Geological Survey of India (GSI), Survey of India (SOI), Ministry of Rural Development, India, for providing the required information in digital format. Thanks to Ramkrishna Ashram Krishi Vigyan Kendra, Nimpith, South 24-Parganas, West Bengal – 743338, India for helping in analyzing soil samples.

References

1. Mandal D, Giri N. Soil erosion and policy initiatives in India. Curr. Sci. 2021;120(6):1007–12. DOI
2. Feng X, Wang Y, Chen L, Fu B, Bai G. Modeling soil erosion and its response to land-use change in hilly catchments of the Chinese Loess Plateau. Geomorphology. 2010;118(3):239–48. DOI
3. Zhao G, Mu X, Wen Z, Wang F, Gao P. Soil erosion, conservation, and eco-environment changes in the loess plateau of China. Land Degrad Dev. 2013;24(5):499–510. DOI
4. Thomas J, Joseph S, Thrivikramji K P. Assessment of soil erosion in a tropical mountain river basin of the southern Western Ghats, India using RUSLE and GIS. Geosci. Front. 2018;9(3):893–906. DOI
5. Thomas J, Joseph S, Thrivikramji KP. Assessment of soil erosion in a monsoon-dominated mountain river basin in India using RUSLE-SDR and AHP. Hydrol. Sci. J. 2018;63:542–60. DOI
6. Tralli DM, Blom RG, Zlotnicki V, Donnellan A, Evans DL. Satellite remote sensing of earthquake, volcano, flood, landslide, and coastal inundation hazards. ISPRS J. Photogramm 2005;59:185–98. DOI
7. McIvor I, Youjun H, Daoping L, Eyles G, Pu Z. Agroforestry: Conservation Trees and Erosion Prevention. Encycl. Agric. Food Sys. 2014:208–21. DOI
8. McCool DK, Renard KG. Water Erosion and Water Quality. Adv. Soil Sci. 1990;13:175–85.
9. Tang Q, Xu Y, Bennett SJ, Li Y. Assessment of soil erosion using RUSLE and GIS: a case study of the Yangou watershed in the Loess Plateau, China. Environ. Earth Sci. 2015;73:1715–24. DOI
10. Bhattacharyya R, Ghosh BN, Mishra PK, Mandal B, Rao CS, Sarkar D, Das K, Anil KS, Lalitha M, Hati KM. Soil degradation in India: challenges and potential solutions. Sustainability. 2015;7:3528–70. DOI
11. Lal R. Restoring soil quality to mitigate soil degradation. Sustainability. 2015;7(5):5875–95. DOI
12. FAO. State of food and agriculture 2010–2011. FAO, Geneva. 2011.
13. UNCCD. The economics of desertification, land degradation and drought: Methodologies and analysis for decision-making. 2013.
14. Aslam B, Maqsoom A, Salah Alaloul W, Ali Musarat M, Jabbar T, Zafar A. Soil erosion susceptibility mapping using a GIS-based multi-criteria decision approach: Case of district Chitral, Pakistan. Ain Shams Eng. J. 2021;12:1637–49. DOI
15. Conoscenti C, Maggio CD, Rotigliano E. Soil erosion susceptibility assessment and validation using a geostatistical multivariate approach: a test in Southern Sicily. Nat Hazards. 2008;46:287–305. DOI
16. Pal R, Hembram B, Jana NC. Assessment of soil erosion in the Irga watershed on the eastern edge of the Chota Nagpur Plateau, India. Reg. Sustain. 2024;5(1):1–15. DOI
17. Singh G, Panda RK. Grid-cell based assessment of soil erosion potential for identification of critical erosion prone areas using USLE, GIS and remote sensing: A case study in the Kapgari watershed, India. Int. Soil Water Conserv. Res. 2017;5:202–11. DOI
18. Steffen W, Richardson K, Rockström J, Cornell SE, Fetzer I, Bennett EM, Biggs R, Carpenter SR, De Vries W, De Wit CA, Folke C, Gerten D, Heinke J, Mace GM, Persson LM, Ramanathan V, Reyers B, Sorlin S. Planetary boundaries: Guiding human development on a changing planet. Science. 2015;347(6223). DOI
19. Newbold T, Hudson LN, Arnell AP, Contu S, De Palma A, Ferrier S, Hill SL, Hoskins AJ, Lysenko I, Phillips HR, Burton VJ. Has land use pushed terrestrial biodiversity beyond the planetary boundary? A global assessment. Science. 2016;353(6296):288–91. DOI
20. Nkonya E, Mirzabaev A, von Braun J. Economics of Land Degradation and Improvement – A Global Assessment for Sustainable Development. Springer, Cham, Switzerland. 2016.
21. Anbazhagan M, Nanditha HS, Reshmidev TV. Effect of climate change on soil erosion rate in a tropical Indian catchment. J. Water Clim. Change. 2024;15(3). DOI
22. United Nations. The Sustainable Development Goals Report. 2020.
23. Government of India. Report of the National Commission on Agriculture. Ministry of Agriculture and Irrigation. 1976.
24. National Bureau of Soil Survey and Land Use Planning (NBSS & LUP). Soil survey report. Indian Council of Agricultural Research (ICAR). 2005.
25. Narayan VVD, Babu R. Estimation of soil erosion in India. J. Irrig. Drainage Eng. 1983;109(4):419–34.
26. Lenka NK, Mandal D, Sudhishri S. Permissible soil loss limits for different physiographic regions of West Bengal. Curr. Sci. 2014;107(4):665–70.
27. Wischmeier WH, Smith DD. Predicting rainfall-erosion losses from cropland east of the Rocky Mountains. U.S. Department of Agriculture, Washington. 1965.
28. Azaiez N. Improved Modelling of Soil Loss in El Badalah Basin: Comparing the Performance of the Universal Soil Loss Equation, Revised Universal Soil Loss Equation and Modified Universal Soil Loss Equation Models by Using the Magnetic and Gravimetric Prospection Outcomes. J. Geosci. Environ. Prot. 2021;09(04):50–73. DOI
29. Woolhiser DA, Smith RE, Goodrich DC. Kineros, A Kinematic Runoff and Erosion Model: Documentation and User Manual. USDA Agricultural Research Service. 1990.
30. Nearing MA, Lane LJ, Lopes VL. Modeling soil erosion. In: Lal, R. (Ed.), Soil Erosion Research Methods . Soil and Water Conservation Society. 1994.
31. Renard KG, Foster GR, Weesies GA, McCool DK, Yoder DC. Predicting Soil Erosion by Water: A Guide to Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). USDA Agriculture Handbook No. 703. 1997.
32. Flanagan DC, Nearing MA. USDA-Water Erosion Prediction Project: Hillslope Profile and Watershed Model Documentation. USDA-ARS NSERL Report No. 10. 1995.
33. Morgan RPC, Quinton JN, Rickson RJ, Smith RE. The European soil erosion model (EUROSEM): Documentation and user guide. Centre for Ecology and Hydrology. 1998.
34. Morgan RP, Quinton JN, Smith RE, Govers G, Poesen JW, Auerswald K, Chisci G, Torri D, Styczen ME. The European Soil Erosion Model (EUROSEM): a dynamic approach for predicting sediment transport from fields and small catchments. Earth Surf. Process. Landf. 1998;23(6):527–44. DOI
35. Arnold JG, Srinivasan R, Muttiah RS, Williams JR. Large area hydrologic modeling and assessment part I: Model development. J. Am. Water Resour. Assoc. 1998;34(1):73–89. DOI
36. Williams JR. Sediment-yield prediction with universal equation using runoff energy factor. In Present and Prospective Technology for Predicting Sediment Yield and Sources: Proceedings of the Sediment-Yield Workshop, USDA Sedimentation Laboratory, Oxford, Miss. Agricultural Research Service, US Department of Agriculture. 1975;40(244).
37. Zhang X, Drake N, Wainwright J, Mulligan M. A comparison of Erosion Modelling Techniques in a Pseudo-catchment of the Middle Hills of Nepal. IAHS Publications-Series of Proceedings and Reports-Intern Assoc Hydrological Sciences. 1996;236:467–74.
38. Williams JR. The EPIC Model. In: Singh, V.P., (ed.). Computer Models of Watershed Hydrology . Colorado. Water Resources Publications. 1995;909–1000.
39. Morgan RPC, Morgan DDV, Finney HJ. A predictive model for the assessment of soil erosion risk. J. Agric. Eng. Res. 1984;30(3):245–53.
40. Nearing MA, Hairsine PB. The Future of Soil Erosion Modelling. In Handbook of Erosion Modelling . John Wiley & Sons. 2011.
41. Shi ZH, Ai L, Fang NF, Zhu HD, Zhang HY. Modeling the impacts of integrated small watershed management on soil erosion and sediment delivery: A case study in the Three Gorges Area, China. J. Hydrol. 2012;438–439:156–67. DOI
42. Lim KJ, Engel BA, Tang Z, Choi J, Kim KS, Muthukrishnan S, Tripathy D. Automated Web GIS-based Hydrograph Analysis Tool, WHAT. J. Am. Water Resour. Assoc. 2005;41(6):1407–16. DOI
43. Miller SN, Semmens DJ, Goodrich DC, Hernandez M, Miller RC, Kepner WG, Ebert DW. The Automated Geospatial Watershed Assessment tool. Environ. Model. Softw. 2007;22(3):365–77.
44. Das B, Bandyopadhyay A. Flood Risk Reduction of Rupnarayana River, towards Disaster Management–A Case Study at Bandar of Ghatal Block in Gangetic Delta. J. Geogr. Nat. Disasters. 2015;5(1). DOI
45. Bhunia GS, Maity PK, Shit PK. Spatial Appraisals of Groundwater Recharge Potential Zone Identification Using Remote Sensing and GIS. In Groundwater and Society . 2021;407–27. DOI
46. Roy M, Shamima F, Chatterjee S. Evaluation of Physicochemical and Biological Parameters on the Water Quality of Shilabati River, West Bengal, India. Water Sci. 2021;35(1):71–81. DOI
47. Pal SK. A geomorphological study on regular flood of lower Shilabati river basin and its impact on the arable land. Int. J. Interdiscip. Multidiscip. Stud. 2016;1(1):28–52.
48. NASA. Prediction of Worldwide Energy Resources (POWER). NASA Langley Research Center. 2023. Retrieved on 26th July 2023. Obtained from: Link
49. Wischmeier WH, Smith DD. Predicting rainfall erosion losses from cropland east of the Rocky Mountains: Guide for selection of practices for soil and water conservation. Agricultural Handbook No. 282, U.S. Department of Agriculture. 1965.
50. Babu R, Tejwani KG, Agarwal MP, Bhushan LS. Distribution of erosion index and iso-erosion map of India. Indian J. Soil Conserv. 1978;6(1):1–12.
51. Moore LD, Burch GJ. Physical basis of the length-slope factor in the Universal Soil Loss Equation. Soil Sci. Soc. Am. J. 1985;50(5):1294–8.
52. Durigon VL, Carvalho DF, Antunes MAH, Oliveira PTS, Fernandes MM. NDVI time series for monitoring RUSLE cover management factor in a tropical watershed. Int. J. Remote Sens. 2014;35(2):441–53. DOI
53. USDA-SCS (United States Department of Agriculture-Soil Conservation Service). National Engineering Handbook . Washington. USDA-SCS. 1972.
54. Lee S. Application of logistic regression model and its validation for landslide susceptibility mapping using GIS and remote sensing data. Int. J. Remote Sens. 2004;25(9):1970–85. DOI
55. Bonham-Carter GF. Geographic Information Systems for Geoscientists: Modelling with GIS , Elsevier, Amsterdam, The Netherlands. 1994.
56. Ozdemir A, Altural T. A comparative study of frequency ratio, weights of evidence and logistic regression methods for landslide susceptibility mapping: Sultan Mountains, SW Turkey. J. Asian Earth Sci. 2013;64:180–97. DOI
57. Lee S, Pradhan B. Landslide hazard mapping at Selangor, Malaysia using frequency ratio and logistic regression models. Landslides. 2007;4(1):33–41. DOI
58. Addis A. GIS-Based Landslide Susceptibility Mapping Using Frequency Ratio and Shannon Entropy Models in Dejen District, Northwestern Ethiopia. J. Eng. 2023;2023(1):1062388. DOI
59. Das G, Lepcha K. Application of logistic regression (LR) and frequency ratio (FR) models for landslide susceptibility mapping in Relli Khola river basin of Darjeeling Himalaya, India. SN Appl. Sci. 2019;1(11):1453–522. DOI
60. Samanta RK, Bhunia GS, Shit PK, Pourghasemi HZ. Flood susceptibility mapping using geospatial frequency ratio technique: A case study of Subarnarekha River Basin, India. Model. Earth Syst. Environ. 2018;4:395–408. DOI
61. Sarkar D, Mondal P. Flood vulnerability mapping using frequency ratio (FR) model: A case study on Kulik river basin, Indo-Bangladesh Barind region. Appl. Water Sci. 2020;10(17). DOI
62. Megahed HA, Abdo AM, Rahman MAEA, Scopa A, Hegazy MN. Frequency Ratio Model as Tools for Flood Susceptibility Mapping in Urbanized Areas: A Case Study from Egypt. Appl. Sci. 2023;13(16):9445. DOI
63. Al-Abadi AM, Al-Temmeme AA, Al-Ghanimy MA. A GIS-based combining of frequency ratio and index of entropy approaches for mapping groundwater availability zones at Badra–Al Al-Gharbi–Teeb areas, Iraq. Sustain. Water Resour. Manag. 2016;2:265–83. DOI
64. Balamurugan G, Seshan K, Bera S. Frequency ratio model for groundwater potential mapping and its sustainable management in cold desert, India. J. King Saud Univ. Sci. 2017;29(3):333–47. DOI
65. Rahmati O, Haghizadeh A, Pourghasemi HR, Noormohamadi F. Gully erosion susceptibility mapping: the role of GIS-based bivariate statistical models and their comparison. Nat Hazards. 2016;82:1231–58. DOI
66. Gayen A, Pourghasemi HR, Saha S, Keesstra S, Bai S. Gully erosion susceptibility assessment and management of hazard-prone areas in India using different machine learning algorithms. Sci. Total Environ. 2019;668:124–38. DOI
67. Soren DDL. GIS-Based Gully Erosion Susceptibility Mapping Using Frequency Ratio (FR) Model in Dumka District Jharkhand, India. Int. J. Prog. Res. Eng. Manag. Sci. 2022;2(1):37–53. DOI
68. Tehrany MS, Shabani F, Javier DN, Kumar L. Soil erosion susceptibility mapping for current and 2100 climate conditions using evidential belief function and frequency ratio. Geomatics Nat. Hazards Risk. 2017;8(2):1695–714. DOI
69. Gayen A, Saha S, Pourghasemi HR. Soil erosion assessment using RUSLE model and its validation by FR probability model. Geocarto Int. 2020;35(15):1750–68. DOI
70. Khosravi K, Rezaie R, Cooper JR, Kalantari Z, Abolfathi S, Hatamiafkoueieh J. Soil water erosion susceptibility assessment using deep learning algorithms. J. Hydrol. 2023;618:1–14. DOI
71. Addis A. GIS–based flood susceptibility mapping using frequency ratio and information value models in upper Abay River basin, Ethiopia. Nat. Hazards Res. 2023;3:247–56. DOI
72. Mandal D, Sharda VN. Assessment of permissible soil loss in India employing a quantitative bio-physical model. Curr. Sci. 2011;100(3):383–9.
73. Swets JA. Measuring the accuracy of diagnostic systems. Science. 1988;240(4857):1285–93. DOI
74. Saha S, Gayen A, Pourghasemi HR, Tiefenbacher JP. Identification of soil erosion susceptible areas using fuzzy logic and analytical hierarchy process modeling in an agricultural watershed of Burdwan district, India. Environ. Earth Sci. 2019;78(649). DOI
75. Fawcett T. An introduction to ROC analysis. Pattern Recognit. Lett. 2006;27(8):861–74. DOI
76. Bandyopadhyay S, Pani P. Soil erosion susceptibility analysis in Eastern India using ROC curve method. J. Indian Soc. Remote Sens. 2018;46(2):309–21. DOI
77. Chakraborty R, Pal SC. Evaluating soil erosion models with ROC curves: A case study from Jharkhand. Geocarto Int. 2021;36(3):287–306. DOI
78. Al-Abadi AM. Mapping flood susceptibility in an arid region of southern Iraq using ensemble machine learning classifiers: A comparative study. Arab. J. Geosci. 2018;11(9):218. DOI
79. Ouri AE, Golshan M, Janizadeh S, Cerda A, Melesse AM. Soil Erosion Susceptibility Mapping in Kozetopraghi Catchment, Iran: A Mixed Approach Using Rainfall Simulator and Data Mining Techniques. Land. 2020;9(10):368. DOI
80. Roy J, Saha S. Integration of artificial intelligence with meta classifiers for the gully erosion susceptibility assessment in Hinglo river basin, Eastern India. Adv. Space Res. 2021;67(1):316–33. DOI
81. Nearing MA, Jetten V, Baffaut C, Cerdan O, Couturier A, Hernandez M, Le Bissonnais Y, Nichols MH, Nunes JP, Renschler CS, Souchere V, van Oost K. Modeling response of soil erosion and runoff to changes in precipitation and cover. Catena. 2005;61(2–3):131–54. DOI
82. Ziadat FM, Taimeh A. Effect of rainfall intensity, slope, land use and antecedent soil moisture on soil erosion in an arid environment. Land Degrad Dev. 2013;24:582–90. DOI
83. Gabarron-Galeote MA, RuizSinoga JD, Quesada MA. Influence of aspect in soil and vegetation water dynamics in dry Mediterranean conditions: functional adjustment of evergreen and semideciduous growth forms. Ecohydrology. 2013;6:241–55.
84. Raouf A, Peng Y, Shah TI. Integrated use of aerial photographs and LiDAR images for landslide and soil erosion analysis: a case study of Wakamow Valley, Moose Jaw, Canada. Urb Sci. 2017;1(2):20. DOI
85. Perreault LM, Yager EM, Aalto R. Effects of gradient, distance, curvature and aspect on steep burned and unburned hillslope soil erosion and deposition. Earth Surf. Processes Landf. 2017;42(7):1033–48. DOI
86. Cotton CA. The Erosional Grading of Convex and Concave Slopes. Geogr. J. 1952;118(2):197–204.
87. Sharma A, Tiwari KN, Bhadoria PBS. Effect of land use land cover change on soil erosion potential in an agricultural watershed. Environ. Monit. Assess. 2010;173:789–801. DOI
88. Munoth P, Goyal R. Impacts of land use land cover change on runoff and sediment yield of Upper Tapi River Sub-Basin, India. Int. J. River Basin Manag. 2020;18(2):177–89. DOI
89. Gashaw T, Taffa Tulu T, Argaw M, Worqlul AW. Modeling the impacts of land use–land cover changes on soil erosion and sediment yield in the Andassa watershed, upper Blue Nile basin, Ethiopia. Environ. Earth Sci. 2019;78:1–22. DOI
90. Kathwas AK, Patil N. Geomorphic Control on Soil Erosion – A Case Study in the Subarnarekha Basin, India. Pol. J. Soil Sci. 2021;LIV(1):1–24. DOI
91. Zhou P, Luukkanen O, Tokola T, Nieminen J. Effect of vegetation cover on soil erosion in a mountainous watershed. Catena. 2008;75:319–25. DOI
92. Ochoa-Cueva P, Fries A, Montesinos P, Rodrıguez Dıaz JA, Boll J. Spatial estimation of soil erosion risk by landcover change in the Andes of southern Ecuador. Land Degrad Dev. 2015;26(6):565–73. DOI
93. Walkley A, Black IA. An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. Soil Sci. 1934;37:29–38. DOI
94. Matranga G, Francesco Palazzi F, Leanza A, Milella A, Reina G, Cavallo E, Biddoccu M. Estimating soil surface roughness by proximal sensing for soil erosion modeling implementation at field scale. Environ. Res. 2023;238:1–14. DOI
95. Bremenfeld S, Fiener P, Govers G. Effects of interrill erosion, soil crusting and soil aggregate breakdown on in situ CO₂ effluxes. Catena. 2013;104:14–20. DOI
96. Tarolli P, Cavalli M, Masin R. High-resolution morphologic characterization of conservation agriculture. Catena. 2019;172:846–56. DOI
97. Riley SJ, DeGloria SD, Elliot R. A terrain ruggedness index that quantifies topographic heterogeneity. Intermt. J. Sci. 1999;5(1–4):23–7.
98. Iqbal J, Read JJ, Thomasson AJ, Jenkins JN. Relationships between soil–landscape and dryland cotton lint yield. Soil Sci. Soc. Am. J. 2005;69:872–82. DOI
99. Daly E, Porporato AA. Review of Soil Moisture Dynamics: From Rainfall Infiltration to Ecosystem Response. Environ. Eng. Sci. 2005;22:9–24. DOI
100. Brady NC, Weil RR. The Nature and Properties of Soils ; 13th ed. Prentice Hall: Upper Saddle River, NJ, USA. 2002.
101. Chen X, Kumar M, McGlynn BL. Variations in streamflow response to large hurricane-season storms in a southeastern US watershed. J. Hydrometeorol. 2015;16(1):55–69. DOI
102. Flanagan DC, Foster GR, Moldenhauer WC. Storm pattern effect on infiltration, runoff, and erosion. Trans. ASAE. 1988;31(2):414–20.
103. Beven KJ, Kirkby MJ. A physically based, variable contributing area model of basin hydrology. Hydrol. Sci. J. 1979;24(1):43–69. DOI
104. Horton RE. Drainage basin characteristics. Trans Am Geophys Union. 1932;13:350–61.
105. Strahler AN. Quantitative analysis of watershed geomorphology. Trans Am Geophys Union. 1957;38(6):913–20. DOI
106. Seutloali K, Dube T, Mutanga O. Assessing and mapping the severity of soil erosion using the 30-m Landsat multispectral satellite data in the former South African homelands of Transkei. Phys. Chem. Earth 2016;100:296–304. DOI
107. Moore ID, Gessler P, Nielsen G, Peterson G. Soil attribute prediction using terrain analysis. Soil Sci. Soc. Am. J. 1993;57:443–52.
108. Kakembo V, Xanga W, Rowntree K. Topographic thresholds in gully development on the hillslopes of communal areas in Ngqushwa Local Municipality, Eastern Cape, South Africa. Geomorphology. 2009;110:188–94. DOI
109. Panagos P, Karydas CG, Gitas IZ, Montanarella L. Monthly soil erosion monitoring based on remotely sensed biophysical parameters: a case study in Strymonas river basin towards a functional pan-European service. Int. J. Digit. Earth. 2012;5:461–87. DOI
110. Swanson FJ, Dyrness C. Impact of clear-cutting and road construction on soil erosion by landslides in the western Cascade Range, Oregon. Geology. 1975;3:393–6.

Cite this article:

Pal, R., Jana, N. C. Predicting the spatial patterns of soil erosion hazard using RUSLE and frequency ratio in the Silabati River Basin, eastern India. DYSONA – Applied Science, 2025;6(2): 422-444. doi: 10.30493/das.2025.510627