Evaluation of soil erosion rate using geospatial techniques for enhancing soil conservation efforts

Abstract

According to reports, Ethiopia is one of the countries in sub-Saharan Africa with the worst affected by soil erosion. It has both on-site and off-site consequences on biophysical and socioeconomic settings in an area. The study area is heavily affected by soil erosion forming diverse erosion structures, particularly in the upper course of the watershed. Hence, this work seeks to estimate the geographically distributed annual soil loss rate and mapping of soil erosion hazard hotspot areas in the watershed using the Revised Universal Soil Loss Equation (RUSLE) adapted to Ethiopian conditions. The RUSLE parameters, such as rainfall erosion factor (R-factor), soil erodibility factor (K-factor), slope steepness and slope length factor (LS-factor), land cover factor (C-factor), and conservation practice factor (P-factor) were considered as data input for the analysis to quantify the soil loss rate in the study area. A digital elevation model (DEM) with a 12.5 × 12.5-meter resolution was employed for catchment delineation and determination of the LS factor.The mean yearly rainfall data from the surrounding rain gauge stations was used to analyze the R-factor. The results of the current conditions showed that the average typical soil loss rate from the entire watershed is 23.8 t ha-¹ yr-¹, and the quantity of soil loss from the study area ranged from 0 to 776.71tan /ha-¹ yr-¹. Nonetheless, Tiro Afeta experiences mean soil erosion at a rate of roughly 50.2 t ha-¹ yr-^1, exceeding the acceptable threshold of 11 t ha-¹ yr-¹. Determining the sustainability of soil production requires assessment, particularly in cases where significant yearly soil erosion occurs. Due to intensive agricultural activities in the Xiro Afeta watershed, significant soil erosion is predominantly occurring in thissteep upper region. Consequently, this area urgently requires appropriate soil protection measures.

Introduction

Soil erosion is a process by which dirt is removed from the surface of the ground by wind, water, transit, and deposition. (Bennett and Chapline 2020). Furthermore, it aggravates the physical, chemical, and biophysical properties of the soil by speeding up soil degradation (Lal 2012). Soil erosion can be caused by a variety of factors, such as steep slopes, excessive rainfall following an extended dry spell, incorrect land cover patterns, and natural disasters (Wassie 2020). It’s also possible that the intrinsic characteristics of soil make it more prone to erosion. An instance of an intrinsic feature of this type is a thin layer of silty topsoil that contains little organic components.

it causes both regional effects like the loss of top fertile soil and off-site effects like dam silting, lake ecosystem disruption, polluted drinking water, and increased downstream flooding, land erosion is one of the most important environmental concerns facing the planet. Soil erosion has increased nationwide even though these effects were acknowledged as a global issue in the 20th century (Dotterweich 2013). Studies show that 80% of agricultural fields worldwide experience soil erosion, which compromises the fields’ capacity to yield crops(Lal 1998). Additionally, alarming data regarding the issue of erosion is presented, showing that between 10 and 40 times as much soil from agricultural fields is lost to erosion as is recovered(Pimentel et al. 1993).

Ethiopia experiences severe soil erosion due to its rough and steep terrain as well as more frequent and intense rainstorms. In addition, human activity, inadequate land use, overgrazing, poor agricultural practices, rapid population growth, a substandard agricultural system, and deforestation all contribute to the nation’s extensive soil degradation(Titilola and Jeje 2008).The effects of the nation’s insufficient watershed management measures include the destruction of natural systems, significant sediment deposition in lakes and reservoirs, loss of soil fertility, and river sedimentation (Biswas 1990). The Aleltu River watershed, with its significant untapped potential, suffers from severe environmental degradation primarily due to deforestation. The relentless expansion of agriculture has transformed landscapes into areas of ecological distress. Large-scale forest clearing for agriculture disrupts delicate ecosystems, jeopardizing the region’s natural balance. Continuous agricultural pressure exacerbates deforestation, as farms encroach upon previously untouched areas, driven by the insatiable demand for land. This agricultural expansion leads to soil erosion, biodiversity loss, and deteriorating water quality along the Aleltu River and its tributaries. The region is caught in a cycle of unsustainable farming practices, making sustainable growth uncertain without concerted efforts to curb deforestation and promote environmentally friendly farming methods. Maintaining the Aleltu River ecosystem’s integrity is crucial for the long-term sustainability of agricultural livelihoods and biodiversity preservation.

The Aleltu River, which is the principal river in the study area, is a significant tributary of the Gibe River. Medium-sized tributaries of this river comprise larger sub-catchments that extend over 200 km. Each sub-catchment produces varying amounts of silt and is more or less prone to erosion depending on how the sub-catchments are currently set up.Therefore, conducting this research aids in locating the areas of the specified watershed where soil erosion is most severe. It is vital to research and determine the most prone places in an erosion-vulnerable area before implementing intervention methods. Over the past few decades, many soil scientists have developed and effectively applied a range of prediction models to quantify and assess soil erosion. . These models are employed to determine erosion risk zones and forecast the rate of soil erosion. Using remotely sensed data, GIS, and RUSLE are among the most widely used empirical methods for estimating soil erosion(Biswas 1990). Numerous academics have scrutinized the outcomes of this model, verifying its effectiveness in estimating soil erosion rates and delineating erosion risk areas across the globe (13). This study was focused on the Aleltu River watershed, an area significantly impacted by soil erosion. Surprisingly, no previous research has been conducted in this region. While much attention has been given to the northern part of our country due to its high slopes, it’s important to note that the western part of some areas also exhibits steep terrain, resulting in substantial soil loss annually. The Aleltu River watershed is among these affected areas, yet it has remained largely unexplored in terms of research until now.

Combined, remote sensing, GIS, and RUSLE allow for the cell-by-cell determination of soil erosion loss. Because it links with GIS, has access to relevant data, and is more straightforward than other conceptual and process-based models. Even though the RUSLE model was created after the parameters were evaluated and verified in a range of soil, climatic, and management scenarios in the United States, numerous attempts have been made to calibrate and validate its use for other countries, including Ethiopia.

Materials and methods

The research domain

The investigation’s focus was the Aleltu River watershed in Southwestern Ethiopia. This watershed, which covers an area of around 2390.5 km², is one of the main sub-basins of the Gibe (known as the Gibe in Ethiopia. The research area lies in the latitude range of 7054’00” to 80 38’00” N and the longitude range of 36,053’ 30” to 37020’ 50” E. One of the largest tributaries of the Gibe River is the Aleltu River, which rises in the eastern Oche region and flows southwest to join it (Fig. 1).

Fig. 1

Location map of Aleltu River watershed

Climate and topography

The research area, which is a sub-basin of the Gibe River basin, is composed of a variety of topographical features, from mountainous to flat, with altitudes that range from 1515 to 3011 m above mean sea level. The height-dependent climatic conditions vary. Two agroclimatic zones (Ethiopian: “weyna dega and dega”) with elevation differences of 1500–2300 and above 2300 m, respectively, make up the catchment of the higher elevation “dega” in the research region, the maximum and minimum temperatures are roughly 27.9 and 13.03degrees Celsius, respectively. At Kola, the lowest height of the research region, the maximum and minimum temperatures are 31.3 and 7.4 degrees Celsius, respectively. The chosen recorded maximum and minimum temperatures from the upper and lower regions of the study area are displayed in Table 1.

Table 1 Mean Temperature Data 190 to 2019

The study area boasts a highly complex forest ecosystem teeming with a diverse array of wildlife. This region is rich in both flora and fauna, featuring an abundance of various plants, vegetables, and fruits. The dense forest provides a thriving habitat for many species, contributing to the area’s ecological diversity.

Agricultural activity in the area is robust, with farms producing a variety of crops. Notably, cassava and coffee are particularly productive, highlighting the region’s fertile soil and favorable growing conditions. The interplay between the rich natural environment and active farming practices underscores the area’s vibrant and dynamic landscape. The annual average rainfall intensity varies among the locations as follows:

Sokoru

Maximum 1992.9 mm, minimum 507.79 mm.

Limu

Maximum 1945.1 mm, minimum 1111 mm.

Seka

Maximum 1989 mm, minimum 879.9 mm.

Bedele

Maximum 9311.4 mm, minimum 997 mm.

Sire

Maximum 1235.09 mm, minimum 936.8 mm.

Jimma

Maximum 1999.56 mm, minimum 945 mm.

Silkam

Maximum 1789.7 mm, minimum 1005.6 mm.

These values highlight the significant variability in rainfall intensity across different locations in the study area (Figs. 2, and 3).

The rainfall distribution and intensity also vary across the elevation variation. The area receives its maximum rainfall from May to September (Fig. 2).

Fig. 2

Mean annual rainfall of the study area for the years 1990–2019 with the integration of temperature

Fig. 3

Map of the agro-climate zone

Soil and geology

Three different types of geological terrains largely shaped the regional geology of the research area. From oldest to youngest, they are Precambrian rocks, Paleozoic to Mesozoic rocks, and Quaternary sediments. Precambrian intrusive rocks, primarily huge granite with coarse-grained texture, cover much of the study region. The rocks are topped with a thick layer of soil with a color range of black to brownish (Table 2).

Table 2 Major soil types and their characteristics

During this study, various types of equipment and software were employed to collect and analyze the data, as detailed in Table 3.

Table 3 Software and tools used for this study

Soil erosion analyzing procedure

The soil erosion vulnerability state in the research region was analyzed using RUSLE in a GIS framework, utilizing elements derived from soil data, topographic maps, satellite image, digital elevation models, meteorological data, and other pertinent studies. To predict the rate of soil loss in a spatial domain, individual RUSLE components such as R, L, S, C, and P were built in the GIS database and combined cell by a cell grid. The majority of secondary data, such as satellite images, DEM, meteorological data, and soil data, were acquired from various governmental organizations( Ethiopian Meteorology Agency, Ethiopian Geology Institution, and Ethiopian Map, agency) to calculate the required regionally dispersed yearly average soil loss rate. Field observations were also conducted to collect the main data, which comprised essential details about the current land management strategies employed in the research region.

The raster calculator’s multiplication technique was used to merge the data layers or maps of the RUSLE model’s R, K, LS, C, and P variables produced from the collected data in the ArcGIS database to calculate the overall rate of soil erosion (Almouctar et al. 2021). The empirical equation of The USLE model is given by Eq. (1).

$$\:\text{A}=\text{R}\text{*}\text{K}\text{*}\text{L}\text{S}\text{*}\text{C}\text{*}\text{P}$$

(1)

Where, A = Computed annual soil loss per unit area in [t ha^-1 yr^-1], R = rainfall factor in [MJ mm ha^-1hr^-1yr^-1], K = soil erodibility factor (soil loss per erosion index unit for a specified soil measured on a standard plot of 22.1 m long, with uniform 9% slope, The ratio of soil loss from the field’s slope length and steepness to the normal slope length of 22.1 m and steepness of 9% slope is known as the “slope length and steepness factor,” or LS (dimensionless). C = land use and land cover factor (ratio of soil loss from a specified area with specified cover and management to that from the same area in tilled continuous fallow) (dimension less), and P = support practice factor (ratio of soil loss with a support practice like; Contour tillage, strip-cropping, terracing to soil loss with row tillage parallel to the slope (dimensionless). The study follows different study designs (Fig. 4).

Fig. 4

Flow chart of the determinations of soil loss using RUSLE in Arc GIS.R-Factor estimation

R-Factor

The amount of soil erosion caused by average yearly precipitation and runoff at a certain place is measured numerically by the R-factor(Efthimiou 2018). To account for soil erosion brought on by raindrop impacts and the ensuing overland flow, RUSLE and its predecessor USLE were developed (Römkens et al. 2015). The rate of soil loss is strongly correlated with various factors related to rainfall. Raindrops can dislodge soil particles from the surface, while the intensity, duration, and patterns of rainfall, whether from a single storm or a series of storms, play crucial roles in soil erosion. Additionally, the pace and volume of runoff resulting from rainfall contribute significantly to soil loss.

The rainfall kinetic energy and intensity during a 30-minute period, which was collected by employing automatic graphic recorders to record rainfall, can also be used to determine this component. Globally, these empirical formulations have been created and are in use .

R stands for the erosion factor (MJ mm ha-¹hr-¹yr-¹), The R-factor for six representative metrological stations in the research area was determined using this methodology. Ethiopia’s National Meteorological Agency (NMA) provided the rainfall data from the chosen stations. Data on rainfall was gathered during 30 years (1990–2019), with some missing records and monthly averages included.

, The yearly average rainfall was collected from Sokoru, Limu Genet, Seka Town, Bedele Town, Sire, Jimma, and Silkam were the rainfall stations used in the sample. The annual average rainfall at each metrological station for 30 years was converted to an R-factor value using Eq. (2). The names of the metrological stations utilized in this investigation, along with the average annual precipitation, are displayed in Table 4.

However, some of the rainfall stations were out of the study area to cover the whole study area while interpolation was

$$\:R=\:-0.812+(0.562\times\:\text{P})$$

(2)

Where; R is the erosion factor (MJ mm ha^− 1 hr^− 1 yr^− 1), and P is the mean annual precipitation (mm).

Table 4 Rain gauge stations with respective average rainfall

As shown in Fig. 5 the rainfall point data interpolation procedure was carried out using Arc GIS 10.3 Inverse Distance Weighted (IDW) to create a data surface from a dispersed set of point data. Ultimately, the R-factor values were interpolated and clipped in the GIS database to produce a corrosive map.

Fig. 5

Map of the mean annual rainfall of the study area interpolated

K-Factor

The K-factor, determined for a length of 22.1 m and a specific slope rainfall of 9%, serves as an indicator of soil susceptibility to particle separation and transportation (Efthimiou 2020). Ceasing soil erosion is influenced by various physical and biological factors, including the presence of organic matter, aggregate stability, shear strength, infiltration capacity, and chemical composition (Mare 2014). These intrinsic soil characteristics, such as silt content, sand composition, organic matter content, soil structure, and permeability, collectively influence the value of the K-factor (Mare 2014). Soil colors further contribute to the variation in K-factor values, with certain colors indicating higher susceptibility to water erosion than others (Table 5).

Table 5 Soil color and respective K-factor values (45)

On a range of 0.15 to 0.3, the K-factor for the research location is established based on the soils that are present and their complementary colors. Soils that are least prone to water erosion are represented by the lower number (0.15), whereas those that are most prone to water erosion are represented by the larger value (0.3). The soil map for the study’s watershed was gathered from the soil class and provided by Ethiopia’s Ministry of Water Resources, Irrigation, and Electricity (MoWIE). The identified varieties of soil in the study area were Orthocrisols, Dystric Nitisols, Pellicvertisols, and Dystric fluvisols. Therefore, among these soil types, Orthic Acrisols are the most prone to water erosion. This type of soil is characterized by clay accumulation, low base saturation, an acidic nature, and distinct horizons. These properties present challenges for agriculture due to their acidity and nutrient limitations, but they can be managed with appropriate soil amendments and conservation practices. Table 6 lists the colors of several soil types derived from many literary sources.

Table 6 The research area’s soil types, together with their physical color and associated K-factor values

Finally, after the resulting shape file attribute table was edited and the K-factor values, the map changed to a grid file or raster format with a cell size of 12.5 × 12.5 m resolution in ArcGIS to generate an erodibility factor map.

LS-factor

The distance from the place of origin of the overland flow to the point at which the slope sufficiently decreases to permit the onset of deposition or the entry of runoff water into a clearly defined channel is known as the slope length (Mare 2014). Because of the accumulation of downslope runoff, runoff water velocity generally increases with slope length. Slope steepness is therefore defined as the gradient from the site of flow origin to the point at which either the slope decreases sufficiently to let deposition occur or runoff water enters a well-defined channel; the higher the predicted erosion, the steeper the gradient.

Slope steepness has typically been demonstrated to be a significant element influencing the degradation of soil in RUSLE model parameters because the steeper the field’s slope, the more water is forced down the hill, the faster the water runs, and the larger the soil loss from water erosion. The ratio of soil loss resulting from the field’s slope length and steepness to the standard slope length and steepness of 22.1 m and 9% slope represents the combined component’s effects(Mccool et al. 1987). This feature is a major contributor to soil erosion since it increases with slope length and steepness, which also increases concentrated flow velocity and soil stability.

12.5 m x 12.5 m digital elevation model (DEM) data from the study area were used to compute the LS factor. The United States Geological Survey (USGS) produced the DEM data that was utilized, and it is publicly accessible on the Internet raster lay. It was created using the spatial analysis tool in ArcGIS from DEM data. After the fill operation, the Arc Hydro tools of the ArcGIS extension were utilized to process and generate the flow direction and accumulation map from the DEM. These results were then utilized as an imputation for the LS-factor calculation.

The raster calculator of Arc GIS uses the following equation, Eq. (3), to create the S-factor map for LAN. Different GIS experts are developing different methods for determining the LS factor at different points in time (Lewis et al. 2005)25.

$${\rm{LS = (power }}\left( {{\rm{flow length/22}}{\rm{.1}}} \right){\rm{, 0}}{\rm{.3* power }}\left( {{\rm{slope/9, 1}}{\rm{.3}}} \right)$$

(3)

Where λ = Flow length; and S = Slope in percent.

Two steps in raster calculator:

• 1. Determination of (λ0.3) * (S/9)1.3

• 2. Division of the result of step one by 22.1(Mogesie, 2014)

The values of exponents range from 0.3 to 1.3. Specifically, the exponent for 0.3 ranges from 0.2 to 0.6, and the exponent for 1.3 ranges from 1.0 to 1.3, with lower values used for prevailing sheet flow and higher values for prevailing rill flow. The length and slope of the standard USLE plot are 22.13 m (72.6 feet) and 0.09 radians (5.14°), respectively (Mogesie 2014).

C-Factor estimation

The Land Use and Land Cover (LU/LC) component, which shows how land cover and its management affect soil erosion rate, is the second most significant factor influencing soil erosion (after topography).

This factor must be approximated, as much as possible, using real LU/LC data that appropriately represents the current research region (Table 7). This study was conducted using the 2020 land use/land cover classification map which is based on USGS. A maximum likelihood method was used to classify the Aleltu River’s Landsat 8 watershed image. Four different land use and land cover categories were visible when the research region was taken out of this LU/LC map. Grassland, Bareland, Forest, and Cultivated landwere visible. 69% of the land is covered by cultivated farms. Forest and grassland make up the second and third percent of the total area, with 23.3 and 7.6%, respectively (Table 8).

Table 7 The date of acquestioon and level of processing of the LU/LC.

Following classification, each land use and land cover class’s corresponding C-factor values were determined (Table 8). These values were sent to the relevant LU/LC groups by earlier research findings. These values were then entered into the LU/LC map’s attribute table to generate the C-factor map in the Arc GIS database.

Table 8 LU/LC class and their accompanying C-factor values

P-Factor estimation

The conservation practice factor represents the impacts of soil conservation techniques that reduce water flow quantity and rate, promote infiltration, and hence diminish erosion. The first issue is that the P-Factor ranges from 0 to 1, but not exclusively within this range. The second issue is that when there is downslope cultivation, it indicates a lack of support practices such as contouring and terracing. The absence of these support practices is represented by a P-Factor value of 1, not 0. Because of this, the consequences of this component rely on the actual farming that is done in the area by farmers or other stakeholders.

The primary impacting elements are terracing, contouring, and strip cropping because they lessen the slope’s steepness and decrease the strength of rainfall-runoff erosion, which promotes infiltration (Bhat et al. 2019). These practices offer a significant advantage against erosion because they permit surface runoff to flow at a lower velocity and be less concentrated in a channel. Numerous scholars have endeavored to examine the most common physical management techniques, including contouring while farming and contouring in conjunction with terracing (Mairura et al. 2021).

Table 9P-factor values for corresponding conservation practice for two cases (if only contouring practice is commonly practiced and if both contouring and terracing practice were fully developed), within a given range of slope gradient in percent.

Table 9 P-factor for commonly practiced soil conservation activities

During the site visit, data regarding the research area’s practices was acquired about soil conservation and management. Information gathered during the site visit indicated that contour plowing was the most widely used method of conserving soil and water. However, shows that the most popular technique for reducing soil erosion in the research region is contour plowing, which is followed by the construction of a few dirt and stone bunds.

Soil conservation is a well-defined practice that farmers in cultivated regions have long employed as their primary means of controlling soil erosion. To develop a green economy that is climate change resilient, efforts have been made over the past six years to apply management principles both nationally and in the study region. In a limited section of the research area, certain small-scale trench excavation and afforestation operations have also been recorded using this methodology.

Consequently, not every effective soil and water conservation strategy in the watershed area was implemented. Furthermore, this study compares the rate of soil degradation in the study area under current soil management methods, such as contour plowing solely with the rate under fully implemented use of industry-standard technological solutions, such as terracing in conjunction with plowing. The topography (slope) of the site played a significant role in choosing conservation efforts, as steep and moderate slopes would not experience the same levels of soil erosion under the same management strategy. The study area revealed that most parts have steep slopes (greater than 16%) with P-Factor values ranging from 0.4 to 0.9, indicating a significant need for soil conservation.

As a result, the study location’s slope gradient limits determined the values of the conservation practice factor. Seven slope gradient ranges, each with a corresponding P-factor value, are used to divide the research region (Table 9).

Point elevation data sets known as DEMs are available for free download from the internet. These data consist of point elevation or values in addition to x and y grid coordinates. It is a compilation of raster data created in different methods for different scales or resolutions of maps.

For this investigation, however, a resolution or pixel size of 12.5 × 12.5 m was chosen. The point elevation data are particularly useful as an input to the GIS for generating critical derivative products such as slope and flow accumulation. In this study, the DEM data were used to determine the agro-climatic zone of the catch, locate the watershed that produces the outlet near the Gibe River confluence site, and supply important RUSLE variables such as LS and P-factor.

Results

RUSLE model parameters

R-Factor

Within the research region, a variance in the mean annual rainfall quantity was observed to cause a variation in rainfall erosion. This resulted in a variation in the measured rainfall erosivity values at the chosen rainfall stations, with Sokoru recording 1507.76 mm and Sire town 1854 mm. The high mean rainfall is about 1845 mm and the lowest is about 1596.16 mm. The high R-factor is 1028.77 and the lowest is 888.923 MJ mm ha-¹hr-¹yr-¹. (Figure 6).

Fig. 6

R-factor map

K-Factor

The K-factor values of the present soils in the research area vary, ranging from 0.15 to 0.3 t ha-¹MJ-¹mm-¹, suggesting a variety in soil erodibility. The soil is more likely to erode when the K-factor is close to 1 and more capable of preventing erosion when it is close to 0. Consequently, the largest K-factor values of 0.3 are found in orthocrisols which make up roughly 13% of the total area, With a K-factor of 0.15 is 2.8% area coverage, pellic vertisols have the lowest values, suggesting that the earth ‘ss surface is less prone to erosion (Fig. 7a, and b).

Fig. 7

Major Soil types in the study area (A) respective K-factor map (B)

LS-Factor

The lower, flatter part, 0 LS-factor, and the higher, steeper sector, 214 LS-factor, correspond to the research area. Figure 8 shows that the center and southern portions of the research area had lower LS factors. Larger LS factors have been discovered to be more common in the rocky and steep areas of the research location. Those with lower LS-factor values than those with higher LS-factor values would have less severe soil erosion as anticipated by this factor (Fig. 8).

Fig. 8

LS- factor map of the study area

C-Factor

From the classified LU/LC image, the area of each LU/LC class was calculated and presented in Fig. 9. Based on the calculation, it was observed that the highland area was covered with forest at about 23.3%, grassland at about 7.6%, and cultivated land at 69% which corresponds with lower C-factor values. Over the study area, forest and grassland which have C-factor values of 0.001 and 0.05, collectively cover an area of only 30.9%, and about 69% of the study area was covered with cultivated land (Fig. 9a, and b).

Fig. 9

Maps of LU/LC (A) and corresponding map of C- factor (B)

P-Factor

The P-factor value of the land management plan now implemented in the research region ranges from 0.4 to 0.9 for various slope gradients. The results show that the center and most of the northern of the study region have lower P-factor values than the rest of the region, which has higher P-factor values. The slope map of the research region indicates that the middle of the region contains a high level and moderate slope ranging from 2 to 16%, while the southern, eastern, and western regions have slopes higher than 16%. Figure 10 displays the P-factor value. Greater P-factor values were concentrated in the higher and outer regions of the subject matter area, whereas lower P-factor values were clustered in the middle of the research zone.

Fig. 10

P-factor for existing condition (A) for the imaged conditions (B)

Estimated average annual soil loss for the existing condition

For the entire research region, the pixel-based modeling results indicated an average yearly soil loss rate of 23.8 t ha-¹ yr-¹. However, in certain areas, the annual soil erosion rate reaches as high as 776.71 t ha-¹ yr-¹, as illustrated in Fig. 11.

Fig. 11

Total soil loss rate map of the study area

Based on the most prone areas to erosion, sites with very little impact, areas with slight impact, and other relevant patterns in erosion conditions, the watershed was categorized into five severity groups. About 63.6% of the watershed was in less soil erosion, with rates ranging from very slight to modest (3 to 10.9 t ha-¹ yr-¹). Consequently, there is a reasonable chance of soil erosion in these locations. Three categories were created out of the remaining research region: 41.8, 67.6, and 124 Moderate Severe, and Very severe soil erosion risk regions respectively (Table 10).

Table 10 Soil erosion severity class and corresponding percent coverage area (Mustefa et al. 2020)

The estimated maximum amount of soil loss that can occur from a single plot of land without causing soil deterioration is between 5 and 11 t ha-¹ yr-¹. Therefore, it can be inferred that there is little risk of soil erosion in the middle sections of the research area, which comprise approximately 63.6% of the total area. This is attributed to the finding that the erosion limit of the soil in this region is 11 t ha-¹ yr-¹.

Prioritization of soil erosion vulnerable area

The boundaries of the districts within the watershed of the Aleltu River are shown in Fig. 12, along with the respective severity class of soil erosion for each district. The Tiro Afeta district was deemed a highly significant site susceptible to soil erosion based on the results. The erosion rate in this district was the greatest in the entire research area, ranging from 0 to 820.16 t ha-¹ yr-¹ at a mean of 50.2 t ha-¹ yr-¹. The average yearly soil losses in the Limu Seka and Cora Botor districts are 10 and 20.2 t ha-¹ yr-^1, respectively, making them less vulnerable. For a variety of causes, this led to more severe soil erosion in some areas of the study area than in others (Fig. 12). The differences in physical circumstances between the regions were one of the main factors. The annual average, minimum, and maximum rates of soil loss for each district are displayed in Table 11.

Fig. 12

Boundaries of districts in the study area and severity class map

Table 11 Shows the vulnerability of soil erosion at the district level

Discussions

The significant geographical heterogeneity in topography, variation in land cover, land use, and increased rainfall volatility are all identified as major contributors to the significant spatial variability in soil degradation within the watershed, as suggested by the research. These factors collectively represent the primary sources of soil erosion in the study region.

The reduction in soil loss vulnerability values was influenced by several factors, including lower rainfall erosivity (882–973 MJ mm ha^− 1 hr^− 1 yr^− 1), lower K-factor values, and lower LS-factor values ranging from 0 to 0.1. The central areas of the research site exhibited generally level and moderately sloped terrain. The results indicated that an area greater than 10.5 km² (or 0.44%) of the study region experienced significant soil erosion. There are two levels of erosion severity identified: very severe erosion (> 50 t ha^− 1 yr^− 1) and severe erosion (30–50 t ha^− 1 yr^− 1). The severe erosion is primarily located in the southern half of the watershed, with some isolated locations in the eastern and western portions. This pattern is due to the higher erosive strength of rainfall from increased rainfall intensity and higher LS-factor.As the LS factor increases, soil loss increases(Michalopoulou et al. 2022). The LS factor in the study area ranged from 0 to 214.042, with the maximum values due to the steep slopes, which were greater than 20%, in the southern, some northern, western, and eastern parts of the area. Only the central part of the study area has gentle slopes.

The research area is subjected to exceptionally erosive rains that originate in the study zone’s southernmost portion. It steadily decreases toward the few parts of the east and center of the study region in the direction of Limmu Genet. Limmu Gennet, the midway location, shares similar characteresics. Shares the values of erosion that are in the middle of each region’s highest and lowest values. Between the highest and lowest erosion rates, shares the geographically scattered erosion values. Sire town has recorded 1028.77 MJ mm ha-¹hr-¹yr-¹ R-factor which is the maximum from all study area districts while Sokoru has less R-factor which is about 888.923 MJ mm ha-¹hr-¹yr-¹ This maximum R-factor increase showed that the erosive power is maximum.

Soil erosion decreases with a lower K-factor, but increases as the K-factor rises(Lin et al. 2019). The computerized soil map of the research area showed four different types of soil, each with a variety of properties. Dystric nitisol, which is nearly always present, makes up the majority of the soil in the region. The majority of this kind of soil is found at the basin’s northern and central borders. The second-highest concentration of dystricfluyisols is found in the watershed’s central, western, and northern portions. In the study area, 97% of the region had a K-factor between 0.2 and 0.3. Most of these soil types were located in the central and southwest sections of the watershed, with only a small percentage in the northern area. Consequently, the watershed is somewhat vulnerable to soil erosion.

As the C-factor increases, soil vulnerability also rises(Pushpalatha et al. 2017). In the study area, about 69% was covered by agricultural land with a C-factor of 0.18 (maximum), contributing to soil vulnerability. Soil erosion from this area was expected to be high because the soil was exposed to the first rainfall events without any cover. The cultivated land covers most of the central parts with some scattered distribution in the Southern and Northern parts of the study area.

An increasing P-factor indicates poor land management, leading to higher soil erosion(Tian et al. 2021). The P-factor value of the current land management plan ranges from 0.4 to 0.9 across various slope gradients. The central and northern parts of the study region have lower P-factor values, while the higher and outer regions have higher P-factor values. Greater P-factor values were concentrated in the higher and outer regions of the study area, whereas lower P-factor values were clustered in the middle of the research zone.

The results generally showed closed results when compared to earlier studies on a few Ethiopian basins and watersheds, both in terms of the expected soil loss rate and the spatial patterns. Determined that the rate of soil loss in Ethiopia’s highlands ranged from 0.0 to 300 t ha-¹ yr-¹(Shiferaw 2012). Other pertinent research found that using the same model, the range of the soil loss rate is 0 to 237 t ha-¹ yr-¹. Demonstrates how the soil loss rate, which ranges from 0 to 203 t ha^− 1 yr^− 1, varies from the watershed near the research region (Usman et al. 2023).

Therefore, the average soil degradation rate (23.8 t ha-¹ yr-¹) for the entire watershed in the present investigation is comparable to the findings from the Jabi Tehinan watershed in the North-Western highlands (about 30.6 t ha-1 year-1), and the Wondo Genet watershed (about 26 t ha^− 1 yr^− 1) (Gebreegziabher et al. 2023). Using the same projection model, an average of 27.5 t ha^− 1 yr^− 1 was determined for the entire Upper Blue Nile (Haregeweyn et al. 2017).

Other research suggests that the rates of erosion in other Ethiopian watersheds are marginally higher than those found in this study. The Chemoga watershed in the Blue Nile Basin in the North-Western highland experiences notable rates of erosion, with an average soil rate of 93t ha-¹ yr-¹(Bewket and Teferi 2009). Furthermore, 47.4 t ha-¹ yr-¹ is the average rate of soil loss in the Koga watershed of the Blue Nile basin (Gashaw et al. 2021). A few other studies show very low average rates of soil erosion, which contrasts with the results of the present and past studies in the highlands. It was found that the Medego watershed in the northern highlands has a rate of 9.63 t ha-¹ yr-¹ (Gashaw et al. 2021). It was also stated that the Zingin watershed has a rate of 9.10 t ha-¹ yr-¹ (Gashaw et al. 2018). The reason for this discrepancy in results is the real condition of the watersheds. The results were lower since the research region had a moderate slope and was primarily flat. In general, the highest soil erosion covers the maximum area which existed in Tiro Afeta district, and the lowest soil erosion coverage covers the minimum area which existed in Limu Sega district (Fig. 13).

Fig. 13

Coverage of soil erosion

Conclusions

This study aimed to provide a thorough evaluation of the distribution of erosion in the watershed under the current conditions as well as recommended techniques for managing the watershed. The current major water-induced soil erosion in the studied area was validated by the study’s findings. The study’s conclusions show that, under the existing circumstances, the annual rate of soil loss ranges from 0 to 776.71 t ha-¹ yr-¹, with an average annual soil loss of 23.8 t ha-¹ yr-¹. This is much more than the maximum amount of soil loss that is allowed, which is 11 t ha-¹ yr-¹. Such losses could affect the long-term viability of the study area’s land productivity, as well as pose a problem with eutrophication and excessive sedimentation at the Aleltu River’s anticipated reservoirs downstream.

In regions of the watersheds with steep slopes, especially in the Xiro Afeta, the average annual soil loss is 50.2 t ha-¹ yr-¹, and the overall rate of soil erosion can reach 820.6 t ha-¹ yr-¹. The districts of Limu Kosa, Cora Botor, and Limu Saqa are located in the central and eastern parts of the research area, respectively, with lower rates of erosion and lesser vulnerability to erosion. The calculated soil erosion rate was compared to earlier estimations and reports from the area and found to be reasonable to validate the study’s conclusions. The anticipated amount of soil loss and its geographical distribution may facilitate the establishment of a comprehensive and sustainable land management strategy through conservation planning for the research area’s prioritized soil erosion risk regions.

The advice emphasizes how critical it is to put in place extensive and long-lasting soil and water conservation measures inside the watershed that are specifically designed to take into account the special features of each stream order and agricultural sector. To avoid irreversible land degradation, special care is needed for places experiencing high to very severe soil loss, such as the Tiro Afeta district. Watershed management techniques must also be used to moderate soil erosion sites to prevent additional deterioration and erosion and to protect these sensitive places from environmental deterioration.

It is recommended that local populations quickly use soil conservation measures in agricultural areas, such as contour plowing, mulching, strip cropping, terracing, multiple cropping, and other traditional methods. To adopt effective conservation measures and to promote sustainable land management practices, community engagement is essential. Additionally, with a dedication to frequent updates and improvements, stakeholders and decision-makers are urged to create both short- and long-term natural resource management systems. By taking a proactive stance, these systems are certain to remain effective in safeguarding soil and water resources and fostering ecosystem resilience.

Finally, the possible advantages of contour plowing in conjunction with terracing highlight the significance of additional study and investigation into cutting-edge conservation strategies. These methods show promise in higher soil erosion rates and maintaining the watershed’s ecological integrity. By working together and taking proactive conservation measures, stakeholders may achieve sustainable land management objectives and protect the resilience and health of the watershed ecosystem.