Introduction

Water erosion is responsible for a significant portion of the estimated 2000 t km−2 year−1 of topsoil removal, making it a primary contributor to global soil erosion (Cherif et al., 2023). Intense precipitation, rushing rivers, and other water-related activities contribute to soil surface erosion, which is transported downstream. This has far-reaching consequences for agriculture, human activities dependent on fertile soil, and the health of aquatic ecosystems downstream. While less frequent, wind erosion can also substantially impact, particularly in desert regions where the top layer of soil is susceptible to being carried away by strong wind gusts due to limited moisture availability (Chuenchum et al., 2020; Fadhil, 2013). Soil erosion is a significant concern that will persist in the future, particularly in low and middle-income countries, notably in tropical and sub-tropical region (Lal, 2001). Each year, water and wind erosion damage approximately 1.094 and 0.549 billion hectares of land, respectively (Lal, 2003). Evaluating and predicting soil loss is a crucial subject in soil science. It is a joint exercise to predict soil loss rates and calculate the transport sediment production using RUSLE (Gayen et al., 2020; Liou et al., 2022; Neupane et al., 2023).

As a consequence of anthropogenic activities, including shifting land uses, woodland loss, excessive grazing, and intensive farming practices, soil erosion has become a significant environmental issue that has worsened during the twentieth century (Hossini et al., 2022; Tiruneh and Ayalew, 2015; Meadows, 2003). In addition, climate change is also a contributing factor that alters precipitation patterns, causing more intense rainfall and increasing the risk of soil erosion (Li & Fang, 2016). Steep slopes and shallow soils are particularly at risk of soil erosion, making them more vulnerable to aeolian and fluvial erosion (Mahala, 2020). Soil erosion is a complicated problem that requires a multi-faceted approach, including implementing erosion control practices measures, such as building check dams and ecological restoration to alleviate climate change effects by enhancing agricultural system resilience, and sustainable land use exercises (Ashraf et al., 2017; Hoyos, 2005; Pradeep et al., 2015; Yuliang & Yun, 2002).

Park et al. (2011) emphasize the significance of quantities analysis in assessing soil erosion risk and determining efficient control of erosion techniques. This approach facilitates measuring and anticipating soil erosion rates, enabling informed land management choices and prioritizing conservation endeavors. Fluvial erosion is harmful to agricultural productivity as it removes the topmost fertile soil layer containing organic material and critical nutrients for plant growth. Every year, 75 billion tons of soils are lost from agricultural land around the world (Akça et al., 2022). There are locations and circumstances where fluvial erosion rates exceed 100 t ha1 year1 (Osman & Osman, 2014). Consequently, crop production may be impacted. Loss of soil structure and decreased soil fertility ultimately affect food security and ecosystem health. Therefore, stopping soil erosion and promoting acceptable land management exercises are essential to maintain healthy soil and productive agricultural land (Bag et al., 2022; Gilani et al., 2022; Luetzenburg et al., 2020; Piacentini et al., 2018).

Soil erosion modeling is a valuable tool for evaluating erosion’s magnitude and spatial distribution, providing crucial insights for guiding land-use decisions and implementing effective management strategies to mitigate its impacts (Mitasova et al., 2013; Toubal et al., 2018). Using diverse models, including the RUSLE model and its derivatives, researchers and land managers can estimate erosion rates, patterns, and those underlying issues contributing to soil erosion over extensive land. This approach also enables the identification of erosion hotspots, i.e., areas with severe soil erosion (Gelagay & Minale, 2016). Developing appropriate land-use strategies and practices, i.e., conservation farming, contour cropping, and terracing, becomes feasible using erosion-susceptible area maps generated through these models. These measures aim to prevent soil loss while preserving the fertility of soils (Karydas et al., 2020; Millington, 1986; Xie et al., 2021). Ricci et al. (2020) undertook a study to identify effective Best Management Practices (BMPs) to minimize soil erosion in the Puglia region of Italy. Four management scenarios were implemented: contour farming (BMP1), no-tillage (BMP2), reforestation (BMP3), and contour farming with reforestation (BMP4). Combining contour farming and reforestation (BMP4) was shown to be the most effective method, resulting in a 38% reduction in soil erosion (from 5.95 to 3.70 t ha−1) followed by BMP2 (29%), BMP1 (22%), and BMP3 (15%).

Identifying factors contributing to soil detachment is paramount when designing erosion control measures (Hossini et al., 2022). The RUSLE model is widely employed in soil erosion modeling due to its efficiency and straightforwardness. This model incorporates five crucial variables, namely rainfall, soil type, steepness, slope length, LULC management, and calculation of soil erosion with support practice. GIS/RS data sets can be used to estimate these variables in different forms. By considering these physical variables, the RUSLE model allows the creation of continuous soil erosion prediction (Biswas & Pani, 2015; Habtu & Jayappa, 2022).

Pakistan is confronted with various challenges encompassing natural hazards, socio-economic development, and the repercussions of climate change. It is situated in a geographically precarious region. The country frequently experiences natural calamities, i.e., floods, drought, landslides, and earthquakes. These calamities substantially impact Pakistan’s economy and infrastructure, resulting in loss of life, displacement of people, and damage to property and agriculture (Gilani et al., 2022). According to the Global Climate Risk Index, Pakistan was the fifth most vulnerable to climate change in light of its vulnerability to environmental hazards and its capacity to manage and adapt to these risks (Eckstein et al., 2019).

Fluvial erosion is an important ecological concern in Pakistan. It contributes to poor water quality by carrying suspended sediment and pollutants in river water. These consequences extend beyond the immediate site, including sediment deposition in reservoirs, reduced water storage capacity, and equipment damage. These environmental impacts have significant economic and social implications for the population (Swarnkar et al., 2018; Walling, 1983). Erosion occurring in the young mountains of Hindukush, Himalayas, and Karakorum substantially contributes to carrying sediment load by River Indus. Consequently, the sediment eventually accumulates in the Tarbela Dam. On average, the Indus River transports approximately 250 million tons of silt annually (Kalhoro et al., 2021). Understanding the reason for soil loss is crucial to growing actual erosion management strategies. Moreover, identifying erosion-prone areas can help prioritize conservation efforts (Arabameri et al., 2018). The significant effects of sedimentation in reservoirs underscore the importance of soil erosion management to ensure sustainable water and energy resources, food security, and environmental well-being.

The economic, food security, and ecosystem implications of sedimentation and siltation in reservoirs are substantial. Specifically, in the context of Pakistan, the loss of capacity in the Tarbela reservoir due to sedimentation can adversely affect hydropower production, water supply, agricultural productivity, and fisheries in downstream regions. Moreover, the impacts of sedimentation on shared rivers between Pakistan and neighboring countries can give rise to cross-border conflicts over water resources (Ul-Hussan et al., 2020). The prime objectives are: (1) To recognize the key factors contributing to soil erosion within the research area. (2) To delineate areas that are particularly vulnerable to erosion. (3) To understand the RUSLE model efficiency in soil loss estimation.

Materials and methods

Study area

The Mansehra District, in the Hazara Division of Khyber Pakhtunkhwa Province, Pakistan encompassing a total area of 4,125 km2, is home to an estimated population of 1,555,742 residents as per the 2017 census. This population density equates to approximately 377.15 people per km2. Between 1998 and 2017, the district exhibited an average annual growth rate of 2.47%. Geographically, the district (Fig. 1) is situated within the coordinates of latitude 34°14′05′′ to 35°11′15′′ N and longitude 72°09′10′′ to 72°48′15′′ E. It shares borders with Muzaffarabad district to the east, Kohistan and Battagram districts to the north, Abbottabad and Haripur districts to the south, and Shangla and Buner districts to the west. The district’s strategic location holds significant importance in tourism and transportation (Arif et al., 2022).

Fig. 1
figure 1

A Pakistan administrative map. B Khyber Pakhtunkhwa map and location of Mansehra district, and location of Siran River Watershed. C The location of Siran River watershed

Full size image

The Karakoram Highway, traversing through the district, is a major transportation route connecting Pakistan with China to the northern region and acts as a getaway between these two countries. Due to its diverse topography and location, the district serves as a catchment area for several significant valleys, including Kaghan, Batagram, and Balakot (Inam, 2021). Effectively managing soil erosion and other environmental concerns in this area is crucial to ensure sustainable development and preserve its natural beauty. Mansehra district encounters diverse climatic conditions, encompassing winter snowfall and heavy rainfall during the monsoon season.

Notably, maize and wheat are the primary crops cultivated in the district, contributing significantly to its agricultural production (Ahmad et al., 2007; Khwaja et al., 2018). The Kunhar and Siran rivers are crucial in providing water resources for the district. Originating from the enchanting Lake Saif-ul-Malook, the Kunhar River traverses Balakot and Garhi Habibullah before joining the Neelum River (Taneez et al., 2021). On the other hand, the Siran River originates from the Mandagucha Glacier and meanders through the Pakhal Valley, ultimately flowing into the Tarbela Dam. Two significant canals, namely the Siran Canal at Dharial in the upper part and Siran Canal at Shinkari in the lower part, have been constructed along the course of the Siran River, serving as vital irrigation channels (Qaisar et al., 2008; Raza et al., 2016).

The watershed of Siran River is situated within the latitude range of 34°34′30″ and 34°45′15″ and the longitude range of 73°15′30″ and 73°25°15″ east. Stretching approximately 130 km, the river joined with the Indus River at Tarbela in the Hazara division. Siran River catchment exhibits diverse landscapes comprising valleys, mountains, plains, and lakes. The geological formations in the region encompass a broad period, ranging from the Precambrian age to the present era, and exhibit remnants of both Cambrian and Precambrian rocks (Awan et al., 2021).

Methodology

The prime goal of this research is to estimate soil loss resulting from water erosion using the RUSLE model, which incorporates multiple factors associated with soil erosion. To achieve this objective, it encompassed activities such as data collection, data processing and analysis, generation of RUSLE factors maps, creation of a soil erosion risk map, and analysis and discussion of results. The methodology flow diagram is presented in Fig. 2.

Fig. 2
figure 2

The methodology flow diagram

Full size image

This study employed the RUSLE model based on secondary data sources to determine the average annual soil loss. This model considers several important factors to estimate soil loss, i.e., rainfall, soil type, slope length, steepness, maintenance factor, and LULC management. The RUSLE model is empirically defined as follows (Renard et al., 1996):-

$$\mathrm{A}=\mathrm{R}\times \mathrm{K}\times \mathrm{LS}\times \mathrm{C}\times \mathrm{P}$$
(1)

where,

A:

= yearly mean soil loss (t/ha every year)

R:

= precipitation erodiblity factor (MJ/ha every year)

K:

= soil erodibility factor (t ha h MJ−1mm−1)

LS:

= slope length and slope steepness factor

C:

= cover management factor

P:

= support practice factor

Rainfall erosivity (R) factor

The influence of runoff is considered while calculating R factor, determining the power of rainfall on soil erosion. According to Wischmeier and Smith (1978) and Cardoso et al. (2020), the value of R-factor for a given rainfall event is determined by multiplying the rainfall energy (E) by 30-min rainfall intensity (I-{30}) in millimeters per hour. The rainfall impact on soil loss is acknowledged by various factors, counting the intensity and amount of rainfall, duration, size, and raindrop distribution (Jadhao et al., 2023; Yan et al., 2008; Ziadat & Taimeh, 2013).

However, estimating rainwater erosivity can pose challenges, particularly in regions where long-term rainfall data is not readily available, including specific areas of Pakistan. Simplified approaches have been developed to overcome this limitation, utilizing easily accessible meteorological information to predict the R-factor. These models establish statistical relationships between precipitation characteristics, such as intensity, duration, frequency, and rainfall erosivity. Although these simplified models may not capture the full range of erosivity variability, they can provide valuable estimations of rainfall erosivity factor in data-scarce regions. The precision of these approaches may vary depending on local climatic conditions and should be validated through field measurements whenever possible (Morgan, 2009).  Panagos et al. (2017) used Global rainfall erosivity database to prepare erosivity map at 30 arc Second based on gaussian process regression.

Several studies have utilized annual rainfall data to compute the erosivity factor (R), such as Awan et al. (2021), Aslam et al. (2021), Chen et al. (2022), Patriche et al. (2023), Angulo-Martínez and Beguería (2009), Panagos, Ballabio, et al. (2015a), and Panagos et al. (2017). In this study, Eq. 2 was employed due to its simplicity, as supported by Chuenchum et al. (2020), Aslam et al. (2021), Musabbir et al. (2023) (Ghosal & Das, 2020), (Maurya & Tanwar, 2021), and (Li et al., 2022).

$$\mathrm{R}=79+0.363\times \mathrm{AAP}$$
(2)

where R represents the factor of rainfall erosivity, while AAP is the average annual rainfall.

Rainfall data for the meteorological stations of Kakul (Abbottabad), Haripur, and Balakot has been collected over 20 years from 2000 to 2020. The gathered data was organized in an attribute table and linked to a location map of the meteorological stations. The Kriging technique was utilized to interpolate point data and generate a continuous surface. Subsequently, the masking tool was used to acquire rainfall data from the interpolated surface of a specific study area (Usowicz et al., 2021).

Soil erodibility (K) factor

Wischmeier et al. (1971) defined “soil erodibility” as the innate propensity of soil eroded by water or wind. It is referred to as a K-factor, the quantities measurement of the resistance to soil erosion. Different soil types may show varied degrees of erosive sensitivity. A variety of physiochemical characteristics of soil affect how easily it erodes, and these characteristics, taken as a whole, can show how resistant the soil is to erosion from surface runoff and rainfall (Khan, 2021). Two physical variables that could influence the K-factor are soil texture, which defines the proportions of sand, silt, and clay components in soil, and soil structure, which specifies how the soil components are grouped into clusters. Soil compaction, slope steepness and length, and vegetation cover can also impact soil erodibility. Additionally, soil’s chemical characteristics, including organic matter content, nitrogen levels, and pH, might impact soil erodibility.

According to Kim et al. (2012), a K-factor value is provided for a specific type of soil type on a conventional study parcel (22.18 m length, 1.83 m width, with a slope gradient is 9%). Typically, the K-factor’s value ranges from 0.01 to 1. To calculate the value of the K-factor for different soil types, Eq. 3 was used by Sharpley & Williams, 1990in the current study. Data was downloaded from the website (https://www.fao.org/soils-portal/soil-survey/soil-maps-and-databases/faounesco-soil-map-of-the-world) to account for soil types in the research region.

$$\mathrm{K}=\mathrm{fcsand}\times \mathrm{fcl}\hbox{--} \mathrm{si}\times \mathrm{forgc}\times \mathrm{fhisand}$$
(3)

where, fcsand, fcl, forg, and fhishand represent a fraction of course sand, a fraction of clay, a fraction of organic matter, and a fraction of sand in the soil, respectively. The fcsand has a low soil erodibility factor, while the factor for soil with high coarse content and less sand content has a high value for soil. The clay and silt having a high proportion ratio in the soil produced by fcl-si result in low soil erosion. High organic matter content in the soil, as indicated by the forgc percentage, leads to reduced soil erosion, particularly in soils with an exceptionally high sand proportion.

$$\mathrm{fcsand}=\left(0.2+0.3\times\exp\right)-0.256\times ms\times\left(1-\frac{msilt}{100}\;\right)$$
(3.1)
$$fcl- si=\left[\frac{msilt\ }{mc+ msilt}\ \right]\ 0.3$$
(3.2)
$$forgc=\left[1.0-\frac{0.25\times orgc}{orgc+\mathit{\exp}\left[\left(3.72-2.95\times orgc\right)\right]}\ \right]$$
(3.3)
$$fhisans=1-\frac{0.7\times \left(1-\frac{ms}{100}\right)}{\left(1-\frac{ms}{100}\right)+\mathit{\exp}\Big(-5.51+22.9\times \left(1-\frac{ms}{100}\right)}$$
(3.4)

where, ms and msilt are the percent sand and silt content with a diameter particle of 0.05–20 mm and 0.02–0.05 mm. While on the other hand, mc and orgc are the percent of clay and organic matter of content with a diameter particle less than 0.002 mm and soil layer in %, respectively. The DSMW was used to obtain the texture of soil data, i.e., sand, silt, clay, and organic material proportions for the study area mentioned earlier. Subsequently, MS Excel was used for the calculation of each type of soil erodibility.

Slope length and steepness factor (LS)

The length of slope and gradient is crucial in soil erosion, affecting LS factor (Chauhan et al., 2020; Khan, 2021). The soil erosion proportion is significantly influenced by terrain’s characteristics, particularly the length of slope and gradient. Steeper slopes are more vulnerable to erosion due to the water runoff of higher velocity, exerting greater pressure on the topsoil. (Chauhan et al., 2020; Khan, 2021).

In addition to steep gradients, longer slope lengths increase soil erosion as rainwater has more time to erode the soil. To determine the LS factor, an effective approach is to utilize a digital elevation model (DEM) (Chauhan et al., 2020; Khan, 2021). In the present investigation, the LS factor was determined using Moore and Burch (1986)'s algorithm (Eq. 4) with the aid of the 30-m ASTER GDEM (Chauhan et al., 2020). This factor is essential for estimating soil loss and understanding erosion dynamics (Chauhan et al., 2020; Khan, 2021).

$$LS=\frac{{\left( slope\ lenght\right)}^{0.4}}{22.13}\times \frac{\left(0.01745\ \mathit{\sin}{\theta}^{1.4}\right)}{0.0896}\times 1.4$$
(4)

where the length of the slope is a flow of accumulation × Digital elevation model cell size used and θ = degree in slope.

The slope length is calculated using the slope parentage and size of pixel of the digital elevation model. Arc-Hydro tool of ArcMap 10.8 was utilized to create accumulation flow from DEM. Das (2021) point out that factor of LS increases as slope length and gradient increase. Areas where flow accumulation takes place are identified as highly vulnerable to soil loss and erosion.

Cover and management factor (C)

In soil erosion simulation, the C-factor represents crop cover and management practices, encompassing the influence of crop cultivation, vegetative cover, and soil erosion mitigation techniques on soil loss. When the land lacks vegetation and is bare, the C-factor is typically assigned a value of 1.0, indicating a lack of protection against soil erosion. However, as vegetative cover increases, the C-factor decreases, signifying greater protection against soil erosion (Egbueri et al., 2022; Obiahu & Elias, 2020; Rawat & Singh, 2018). It is worth noting that when water bodies are considered, the C-factor is often assigned a value of 0 since there is no soil to preserve in such areas. Additionally, vegetation plays a vital role in regulating the erosive force of raindrops. When vegetation covers the soil, it acts as a barrier, preventing raindrops from directly impacting the soil surface and slowing down their erosive effect. Therefore, maintaining a healthy vegetation cover is essential for reducing soil erosion and preserving soil health (Erenstein, 1999).

Sentinel-2 images with a 10-m spatial resolution were acquired from the USGS website and analyzed using ArcGIS 10.8 to determine the C-factor. Maximum likelihood supervised classification was applied to analyze the images into LULC classes. The classified image has been rationalized and evaluated using open-source Google Earth Pro. The accuracy was evaluated for overall accuracy and Kappa coefficient. The overall accuracy was 72.78%, and Kappa coefficient was 0.70 or 70%. The task of C-factor values to each LULC class was created on existing literature (Ebabu et al., 2022; Wan et al., 2022). Typically, C-factor values range between 0 and 1, with higher values indicating greater surface protection through crop or vegetation coverage or various management techniques. It is essential to consider the accuracy of the LULC classifications and the reliability of the extracted values from the published literature. These factors are crucial to ensure that the computation of the C-factor is reliable and accurately reflects the conditions of each LULC class.

Support practice factor (P)

Support practice with the P-factor component in the RUSLE model assesses the efficiency and preservation of soil practices in mitigating soil loss. It is considered landscaping, contour farming, conservation cultivation, and reducing soil erosion due to land management practices.

P-factor ranges from 0 to 1, where 1 value denotes fully implemented practices that efficiently prevent soil erosion, and 0 indicates the absence of such practices. The P-factor is determined based on the study region’s types and extent of preservation practices.

For example, terrace cultivation would have a higher P-factor than ineffective or poorly maintained practices. Calculating the P-factor in the RUSLE can involve various approaches depending on the specific preservation practices employed and data availability (Panagos, Borrelli, & Meusburger, 2015b). In the case of the Siran basin study, DEM was divided into five slope categories. Assuming the implementation of terrace agriculture in the basin, the required P-factor values for each slope class were determined from Koirala et al. (2019) and Shin (1999) studies. Subsequently, P-factor was computed for the study area using the technique specified by Shin (1999).

Results and discussion

Rainfall erosivity (R) factor

The regional distribution of rainfall and R-factor for the study area is illustrated in Figs. 3 and 4, respectively. The data presented in Table 1 were collected from the Regional Meteorological Office Peshawar and other sources, covering the study region and surrounding areas from 2010 to 2020. Table 2 displays the predicted R-values for the Siran watershed and the corresponding areas within different classes. The investigation findings indicate that 58.33% of the study region falls within R-values 434.2–487. The eastern portion of the research region, which comprises 12.05% of the Siran watershed, exhibits the highest R-value while experiencing the lowest soil erosivity.

Fig. 3
figure 3

R-factor classes and associated area under class in the Siran Basin

Full size image
Fig. 4
figure 4

A Mean annual rainfall in mm. B Rainfall erosivity factor map. C Pakistan map showing study area

Full size image
Table 1 Average annual rainfall (mm) for Siran River basin and neighboring meteorological stations.
Full size table
Table 2 R-factor scores for the Siran Basin and associated area under each class
Full size table

Soil erodibility (K) factor

The Siran basin has two main soil types: Eutric Cambisols (be) and Lithosols (I). Eutric Cambisols are the predominant soil type, covering 86.36% of the total area. These soils are highly fertile and widely recognized as ideal for agriculture. They exhibit a loam texture, roughly equal amount of sand, silt, and clay. The A-horizon of Eutric Cambisols is typically yellowish-brown, although it can occasionally be a vivid red. The best Cambisols are found on rocky terrain with tree cover.

On the other hand, Lithosols are thin soils with poorly developed horizons. They form either in mountainous regions or due to the resistance of underlying parent rocks to weathering. Lithosols are commonly found on rocky slopes where deep soil development is hindered by erosion caused by rainfall. Vegetation cover in Lithosols is minimal, mainly comprising shrubs or grasses. These soils are unsuitable for farming due to their shallow depth, steep slopes, and high susceptibility to erosion. In the Siran watershed, Lithosols account for 13.65% of the area. Tables 3 and 4 provide information on the soil types, texture, K-value, and respective areas.

Table 3 Soil types and associated K-factor score for Siran Basin
Full size table
Table 4 Soil types, composition (%), associated K-factor score, and area under each type for Siran Basin
Full size table

Figure 5A illustrates that soils dominate the soil types in the study area, covering a substantial portion of the study area. In contrast, Lithosols are found in smaller patches, primarily sited in the upper stream region of the Siran River basin. As shown in Fig. 5B, most of the research region exhibits a low K-factor score of 0.0204, indicating a low susceptibility to soil erosion. However, despite occupying only 13.65% of the total areas, the highest K-factor value in lithosols exhibit is 0.0184%. The higher concentration of sand in the lithosols soil can be attributed to its elevated value. Soils with a higher sand content are extremely vulnerable to erosion compared to those with higher clay or organic matter content. The reduced binding ability of sands makes them easily transported by water or wind. Thus, the higher sand content in this area’s Lithosols soil may contribute to the observed elevated soil erodibility factor values.

Fig. 5
figure 5

A The spatial distribution of soil types. B Soil types and K-factor

Full size image

Slope length and steepness factor (LS)

LS factor is crucial in assessing potential for soil loss as it considers flow concentration, slope length, and gradient. Fig. 6A visualizes the DEM, while Fig. 6B displays the slope in percentage within the research region. Figure 6C, D depicts the flow accumulation grid and the geographical distribution of LS factor across the research area. The data presented in Table 5 are the values of the LS factor for the Siran watershed ranging from 0.1 to 1439. A significant portion of the area (95.89%) exhibits an LS factor ranging from 0.1 to 2. However, a higher LS factor range of 30.1–1439 is concentrated in a small area of 3.87 km2 (0.22%), primarily associated with steep terrain along streams and rivers. The analysis reveals that vertical slopes are predominantly located in the northeastern region of the study area, indicating a greater likelihood of soil loss in these specific locations

Fig. 6
figure 6

A The DEM of Siran River watershed. B The slope in percentage. C The flow accumulation grid. D Spatial distribution of LS factor

Full size image
Table 5 Siran Basin LS-factor scores and associated area under each class
Full size table

Land cover and management factor (C)

Table 6 provides a comprehensive overview of the LULC categories identified in the research region, the percentage of each class, and their corresponding values of the C-factor. The C-factor is a decisive component of the RUSLE equation with the influence of vegetation and management practices on soil loss. It is typically scaled from 0 to 1, with 0 indicating minimal erosion potential and 1 suggesting a higher likelihood of erosion. Most of the study region, comprising over 80%, comprises agricultural and barren land. These land types generally exhibit higher C-factor values, indicating increased susceptibility to soil loss.

Table 6 Siran Basin, LULC classes, associated C-factor scores, and area under each class
Full size table

In contrast, forested areas display lower C-factor values, indicating a reduced risk of soil erosion. Table 7 and Fig. 7 further provide detailed information on the extent of each C-factor class, which ranges from very low to very high erosion potential. Notably, only a small portion (0.16%) of the research area, considered for 62.34% of the total size, is classified as having an exceptionally high risk of soil loss. While specific locations within the research region present a significant erosion risk, the remaining areas generally demonstrate a moderate to high erosion potential.

Table 7 Showing C-factor values, area in km2, and percentage of Siran River.
Full size table
Fig. 7
figure 7

A The LULC of Siran River watershed. B Spatial distribution of C-factor values

Full size image

Support practice factor (P)

Shin (1999) and Koirala et al. (2019) highlighted the P-factor, which characterizes essential acts in assessing the soil erosion risk by considering the effectiveness of support practices in mitigating erosion. The values of the P-factor are visualized in the study. Figure 8A, B depicts the slope in degrees and the geographical representation of the P-factor value, respectively. These visualizations offer insights into the steepness of slopes and spatial variation in reducing soil erosion to check the effectiveness of support practices. Table 8 shows slope classes, their respective areas, and the corresponding P-factor scores. The P-factor score ranges from 0 to 1, where 0 represents highly effective support practices that resist erosion, while 1 indicates lower resilience to erosion.

Fig. 8
figure 8

A The spatial distributions of slope classes (slope in degree) for the Siran River watershed. B The spatial distributions of support practice (P-factor) values for the Siran River watershed

Full size image
Table 8 Siran basin slope classes, area, and associated P-factor scores of each class
Full size table

Annual soil loss

The study’s key findings are summarized in Table 9 and Figs. 9–12. This research reveals an average annual soil loss of 0.154 million tons, equal to 0.871 tons per hectare or 0.871 tons per hectare. The study region is based on soil erosion intensity and is classified into five classes, i.e., very high, high, moderate, low, and very low. Our analysis demonstrates that a significant portion of the research area has very low soil erosion, which is 72.94% with a rate of 0.021 tons per hectare per year, contributing to the total annual soil loss is 1.67% only. Areas characterized by low to moderate soil erosion rates of 0.021–0.5 tons/ha/year and 0.51–1 tons/ha/year account for approximately 15% of overall annual soil loss, occupying 24.67% of the Siran basin.

Table 9 Showing percentages and erosion severity of Siran River
Full size table
Fig. 9
figure 9

Siran River watershed, soil loss intensity, and area in km2

Full size image
Fig. 10
figure 10

A Watershed level annual soil loss. B Sub-watershed level of Siran River annual soil loss tons/ha/year. C Sub-watershed level and annual soil loss tons/ha/y vs geology

Full size image
Fig. 11
figure 11

A R-factor. B K-factor. C C-factor. D P-factor. E LS-factor. F Annual soil loss and erosion hotspot areas (R*K*LS*C*P)

Full size image
Fig. 12
figure 12

Major rivers of KP and their assessment with Siran River

Full size image

In contrast, regions exhibiting extremely high soil erosion rates (over 10 tons per hectare per year) are found in only 0.15% of the total area, yet they donate to 75.76% of total yearly soil loss. These areas are often located on rocky terrain near streams and rivers, with high LS and C-factors indicating poor crop management practices. The study highlights the critical role of actual erosion control measures in this highly susceptible area to mitigate and preserve soil loss health. Our research findings reveal the significant issue of soil erosion in Siran Basin, particularly the areas characterized by limited vegetation cover and steep slopes. To address this problem and ensure sustainable land use, the study provides valuable insights for implementing effective soil conservation and management strategies. Notably, areas identified as soil erosion hotspots exhibit high erosion rates due to sparse forest cover, shallow sandy or gravelly soils, improper agricultural practices, and the absence of soil conservation measures. These identified areas demonstrate medium to high soil erodibility, indicating their susceptibility to erosion.

Additionally, heavy rainfall contributes to the basin area, with approximately 26.14% of the total area exhibiting slopes exceeding 26.9% and corresponding P scores of 0.2. Based on these findings, it is crucial to prioritize implementing targeted soil preservation and management measures in these vulnerable areas. This will help mitigate soil erosion, promote sustainable land use practices, and safeguard the long-term health and productivity of the Siran Basin’s soil resources. Most erosion hotspots in the Siran River basin are concentrated in the northern regions. This spatial distribution is depicted in Fig. 9, which showcases the area in km2 and the intensity of soil loss. Additionally, Fig. 10A, B presents soil loss annually at the watershed level, while Fig. 10C highlights the association between geology and annual soil loss.

Various parameters were considered to assess the annual soil loss by using RUSLE model as illustrated in Fig. 11A–E. This figure depicts the factors utilized in the annual soil loss calculation. Furthermore, Fig. 11F provides the geographical distribution of erosion-prone locations, commonly referred to as erosion hotspots, within the Siran River watershed. Tables 9 and 10 and Fig. 11A, B present the Siran River basin annual soil loss at the sub-watershed level. Our analysis shows that sub-watersheds 1, 2, and 3 have rates of 207.30, 166.23, and 477.91 tons/ha/year annual loss, respectively. These sub-watersheds predominantly consist of granite as their underlying geology.

Table 10 Sub-watershed of Siran River and annual soil loss tons/ha/year
Full size table

Granite undergoes chemical weathering through hydrolysis of its feldspar component, forming weathered material known as Grus (Chung et al., 2020). Granite also undergoes chemical weathering even under icy conditions, known as polar weathering. Moreover, urban areas such as Mansehra City, Attar Sheesha, Chitta Batta, etc., primarily contribute to WS-6, which exhibits a very high soil erosion rate. The surrounding area of WS-6 is covered by Quaternary sediments, which consist of loose deposits of gravel, sand, silt, and clay. These quaternary deposits are composed of loose, unconsolidated rock that is highly susceptible to erosion.

Different studies have been carried out in various Pakistan regions to estimate the annual soil loss using the RUSLE model. Ullah et al. (2018) presented a study on the district Chakwal, which revealed a yearly soil loss of 97.81 m tons ha−1 year−1, averaging 70 to 230 tons/ha/year. Maqsoon et al. (2020) focused on the Chitral River and estimated 58 tons/ha/year mean yearly soil loss. In the Ghabbir basin, Ashraf et al. (2017) found average annual soil loss is 22 tons/hectare/year. Butt et al. (2011) investigated the Simly basin in the Margalla hills and reported an average yearly soil loss of 14 tons/ha/year.

According to Batool et al. (2021), the Potohar basin experiences recorded 325.39 tons/ha annual soil loss. Gilani et al. (2021) conducted a study on soil erosion in the KP province of Pakistan and estimated yearly soil loss of 11.78 and 21.97 tons/ha/year. Zaidi and Khan (2018) conducted a study on the Kunhar River and found an average soil loss of 21.18 tons/ha/year. In the present study, the average soil loss was calculated to be 0.871 tons/hectare/year, with a total yearly soil loss of 0.154 million tons in the Siran River watershed. Table 11 shows Gilani et al. (2021)’s projections of annual soil loss for Pakistan and its seven administrative regions. The analysis reveals that the estimates of annual soil loss obtained in the current research are significantly lower than other predictions. This difference could be attributed to the relatively unexplored and lush green nature of the Siran Valley, in contrast to the Kunhar and Chitral River watersheds.

Table 11 Siran River and average soil loss of various regions of Pakistan
Full size table

Results comparison and validation

Moreover, regarding water discharge, the Siran River is relatively smaller than other local rivers. Figure 12 provides an overview of the basin areas of significant rivers within the Khyber Pakhtunkhwa province, illustrating the relative size of the Siran River. The environmental conditions in the Kunhar and Siran watersheds exhibit considerable similarities and flow in roughly parallel directions. However, unlike the Kunhar basin, the Siran River basin remains largely unexplored. Nevertheless, the growing number of tourists has rapidly expanded motels and lodges in the Naran region. It is estimated that Naran alone has approximately 1500 hotels of varying sizes. These human activities contribute to sediment production, eventually transported downstream by the Kunhar River.

Conclusions

The primary objective of this research was to count the Siran River basin’s annual soil loss using the integration of RUSLE, GIS, and RS techniques. The study finding reveals that a small portion of the Siran River basin, particularly the northern part, has experienced significant soil loss, while the major portion of the basin has low to moderate levels of soil loss. The annual soil loss estimated in the basin is 0.154 million tons, equal to 0.871 tons per year per hectare. The spatial distribution analysis identifies specific areas in the far north of the basin as erosion hotspots, indicating a higher susceptibility to soil erosion as linked to the southern part. This form of soil erosion is relevant to the challenges faced by less economically developed countries, where soil erosion is a significant problem. For instance, over 30 to 32.80 million of land in India is under river erosion. Similarly, in Iran, soil loss occurs at a rate of 24 tons per acre annually. In Pakistan, soil erosion affects over 16 million hectares, accounting for approximately 20% of the country’s total geographic area. Of this total, 11.2 million hectares are at risk of water erosion.

On a global scale, approximately 20 billion tons of sediments are produced annually, with more than 80% of this sediment ultimately reaching the world’s oceans. The remaining 20% is deposited in reservoirs and dams. These figures highlight the widespread and significant impact of soil erosion worldwide. Agriculture plays a vital role in Pakistan’s economy, with a significant portion of agricultural output relying on the country’s abundant water resources, 50% of which is derived from surface water. The effective utilization of these water resources is closely linked to the management of surface waters. One of Pakistan’s most important water storage facilities is the Tarbela Dam, constructed on the Indus River. Over time, the dam has faced challenges due to soil erosion and changes in the surrounding region’s LULC dynamics. The continuous accumulation of sediment within the dam has led to a reduction in its water storage capacity.

Tarbela Dam, the largest earth-filled reservoir in Pakistan, serves dual purposes of hydroelectricity generation and irrigation. According to Wallingford (1998), the average annual sediment influx into the dam is approximately 81,000 million cubic meters (mm³), with varying annual influxes of fine sediment ranging from 100 to 300 metric tons (MT) between 1967 and 1996. Pakistan Water and Power Development Authority (WAPDA) predicts that the Tarbela Dam will continue to lose water storage capacity, having already experienced a significant decline. Neglecting this issue over an extended period could lead to a major catastrophe. Sedimentation is essential in influencing the reservoir’s storage capacity, emphasizing the importance of accurately assessing sediment inflow into the reservoir.

The findings of this study are expected to provide valuable understanding for governments and relevant organizations in making informed decisions to mitigate soil loss and address the sedimentation issue. Addressing these challenges aims to ensure the long-lasting sustainability of water resources and optimize the functioning of critical infrastructure such as the Tarbela Dam.