Study area

This study area was conducted in the Fincha catchment, Abay Basin, Ethiopia as shown in Figure 1. This delineated watershed is geographically found between 37°0.06′00″E and 37°33′18″E longitude and 09°21′11″N to 10°01′00″N latitude and covers total area of 2148 ha. The major landforms fall in rolling terrain class (2%–10%) which covered about 46%, rolling terrain about 27%, and hilly terrain about 23%. Only small pockets of land which accounts about 1% from total area was covered by flat plain and 1% also characterized by highly rugged mountainous and rolled topography with steep slopes(Olika & Iticha, 2019). The meteorological (recorded precipitation) stations contributing to the outlet were identified using Thiessen Polygon. Accordingly, the contributing stations namely: Alibo, Shambu, haro, Neshi, Homi, Hareto, Gabate, Kombolcha, and Wayu were identified. Since the intensity of rainfall and the infiltration capacity of the soil, geologic, and hydrogeologic settings are the main causes for groundwater recharge, the exploration of potential zones considers the rainfall stations. The soils of the area were dominated by Haplic Vertisols, Eutric Cambisols, and Nitisols.

Figure 1.

Geographical location of the study area.

Data

To quantify the annual soil loss in the catchment, recorded rainfall of eleven (11) stations (Table 1), Digital Elevation Model (DEM), existing soil map, and LANDSAT imageries were used to derive the key factors of RUSLE model. The point-based recorded rainfall data of the selected stations were converted into areal rainfall using an advanced interpolation technique called Kriging method. For this purpose, spatial analyst tool in GIS environment was used to convert the point data into areal using an ordinary kriging method (equation (1)). The selected significant factors (R-factor, P-factor, K-factor, LS-factor, and C-factor) were derived from the areal rainfall, LULC, slope, and soil data. Rainfall, geology, and soil were obtained from Ethiopian Ministry of Water (MW) and LULC was generated from the LANDSAT imageries retrieved from  https://earthexplorer.usgs.gov/ for the 2020 datasets. Digital Elevation Model (DEM) (12.5 × 12.5 m) was downloaded from Alaska University official website:  https://search.asf.alaska.edu/.

(1)

Table 1.

Rainfall Stations in and Surrounding Fincha Subbasin.

Where Z^(Xo) and Z^(Xi) are the interpolated values and observed values of the individual points, N is the total number of observations, and lgr;i the weight of each observation.

Soil loss models

A bunch of documents are available regarding the different soil loss models; however, RUSLE is effectively and intensively used (Fayas et al., 2019; Markose & Jayappa, 2016; Mengistu et al., 2016) worldwide in several catchments. The reliability and capability of RUSLE model was evaluated by (Prasannakumar et al., 2012), and it has been used in the catchment where the impact of soil erosion is very high. RUSLE model is the most popular soil loss estimation model that can be modeled in GIS platform. The raster calculator in spatial analyst tool of GIS environment can easily be used to estimate the rate of soil loss by combining the key significant factors.

Description of RUSLE model

RUSLE is an empirical erosion model recognized as an effective tool to quantify the mean annual soil losses and its risk (Chuenchum et al., 2020). This model is applicable for arable land and limited to a long-term rainfall record. The RUSLE model receives the physical features and surface dynamic changes such as R-factor, P-factor, K-factor, LS-factor, and C-factor as criteria for the quantification of annual soil loss. Based on the physical evidences, and availability of data in this catchment, the selected significant factors were prepared and used the RUSLE model (equation (2)) to estimate the annual soil loss. The raster calculator in the spatial analyst tool of GIS environment has a capability to generate the soil loss map, for this purpose, the flowchart showing the detailed procedures and data needed are summarized in Figure 2.

(2)

Where A the total annual soil loss (t/ha per year), t is the thickness of lost soil

R Rainfall erosivity factor (MJmm ha⁻¹ h⁻¹ year⁻¹)

K Soil erodibility factor (t haMJ⁻¹ mm⁻¹)

LS Slope length and steepness factor (dimensionless)

C Over and management factor (dimensionless)

P Support practice factor (dimensionless)

Figure 2.

Detail flowchart of the steps and data needed in the study.

Rainfall erosivity factor (R)

Rainfall erosivity factor R describes the relationship between the rainfall intensity and the soil responses to it (Kayet et al., 2018). According to the information obtained from the national meteorology of Abay Basin in which the current study is found, varies from 1353.65 to 2030.93 mm. The rainfall erosivity factor (R) is an important factor of the RUSLE model to estimate the annual soil loss. The values of rainfall erosivity factor for this study was derived from fifteen (15) years recorded rainfall of 11 meteorological stations (Refer Table 1). Distributed rainfall in Kriging interpolation technique was used to generate the rainfall erosivity factor (R) (Dessalegn et al., 2017) using the empirical equation (Bekele & Gemi, 2021; Hategekimana et al., 2020) as shown in (equation (3)).

(3)

Where R Rainfall erosivity factor (MJmm ha⁻¹ h⁻¹ yr⁻¹)

P Annual mean precipitation (mm)

Soil erodibility factor (K)

The soil erodibility factor (K) is a quantitative description of the susceptibility of soil to erosion. The degree of tolerating the effect of intensive rainfall is reached when soil particles are detached and transported. The cohesive force between the soil particles is the restoring force, and if the force due to the moving surface runoff is much greater than the cohesive force, the soil particles are exposed to erosion (Girmay et al., 2020; Seifu et al., 2020). The permeability of the soil and the degree of erodibility are interconnected parameters (Kayet et al., 2018). The ability of soil particles in persisting against rainfall is different in different soil types and this property is expressed in terms of erodibility factor (Ayenew et al., 2018). The soil types and the corresponding values of the erodibility factor (K) in the study area.

Topographic factor (LS)

The severity of the spatial variability of soil erosion relies on the topographic conditions of an area. The steepness or the flatness of an area governs the degree of the erodibility of soil particles. The speed of the water flowing over soil and the slope of the area are dependent parameters (Kayet et al., 2018). The length of the slope and slope steepness of the area in the study area was generated using Digital Elevation Model (DEM) of 12.5 × 12.5 m and LS-factor was generated in ArcGIS version 10.4. Flow accumulation and slope (%) are commonly used input parameters with a fixed cell size with regression equation(Abdulkadir et al., 2019) (Eq.4).

(4)

Where LS Slope length and steepness factor (dimensionless)

Support practice (P) and cover and Management factors(C)

The impact of agricultural or cropping system and the practice of crop management on the rate of soil erosion is defined using the C-factor (Hategekimana et al., 2020). It also characterizes the influence of soil disturbing activities, plant sequence, and productivity level, soil cover and subsurface bio-mass on soil erosion. Cover and management factor is important in developing conservation plans (Mohamed & Shantha, 2019). The other key significant factor used in RUSLE model is the support and conservation practices factor called P-factor (Iticha & Takele, 2018). This parameter estimates the support and conservation practices and the corresponding to a particular support practice in up and down slope farming (Markose & Jayappa, 2016). LUC types are the primary inputs for the estimation of P and C values in a given catchment, and therefore in this study, for the LULC and the corresponding C and P values (Table 2 and Figure 3) were assigned based on their influences on soil erosion.

Table 2.

Soil Types and Corresponding Erodibility Factor (K-Values) in the Catchment.

Figure 3.

Major Land Use Land Cover (LULC) in the catchment.

Estimation of soil loss

In this study, the rate of annual soil loss ranging from 0 to 76.5 t ha⁻¹ yr⁻¹ was quantified sing RUSLE model and GIS tools. The total area of the study area is 21.48 km² (2148 ha) and the corresponding annual soil loss was estimated as164, 322 t ha⁻¹ yr⁻¹. The vulnerability of soil to the erosion is very in this catchment as we can see from soil severity map generated in RUSLE model. The estimated annual soil loss (0 to 76.5 t ha⁻¹ yr⁻¹) in this study was much higher than the total annual soil formation rate estimated by (Girmay et al., 2020) that ranges from 2 to 22 t ha⁻¹ yr⁻¹. This indicates that the sensitivity to erosion is highly increased due to increment values of support practice and cover management practice factors. The surface dynamic change of the area should be managed and actions are important to control the LULC to minimize the volume of the surface runoff that in turn increases the erodibility of the soil. Rainfall erosivity factor (R-factor), soil erodibility factor (K-factor), support practices factor (P-factor), cover management practice (C-factor), and the topographic factor (LS-factor) derived from the existing data and detailed information about each factor were briefed in the next sections.

Rainfall erosivity factor (R-factor)

The rainfall intensity and the nature of the surface are the primary reasons for the soil erosion (Fayas et al., 2019). The speed of the surface runoff is governed by the slope of the area. The degree of the detachment of soil particles is dependent of rainfall erosivity. The classifications of the rainfall of the thematic maps in both approaches revealed almost the same categories. This indicates that the ANN model can also be used as an alternative in generating a thematic map. The rainfall in the study area was reclassified into five (5) major categories as (1,425.18–1,586.69 mm), (1,586.69–1,679.21 mm), (1,679.21–1,766.98 mm), (1,766.98–1,873.74 mm), and (1,873.74–2030.30 mm), and the corresponding rainfall erosivity were (0.00–1.34 MJmm ha⁻¹ h⁻¹ yr⁻¹), (1.34–3.42 MJmm ha⁻¹ h⁻¹ yr⁻¹), (3.42–5.96 MJmm ha⁻¹ h⁻¹ yr⁻¹), (5.96–8.83 MJmm ha⁻¹ h⁻¹ yr⁻¹), and (8.83–11.96 MJmm ha⁻¹ h⁻¹ year⁻¹) respectively. As we can see from Figure 4, the southwestern part of the catchment gets an annual rainfall ranging from 1,873.74–2030.30 mm, and there is high sensitivity to erosion. The values of the rainfall erosivity factor are also very high in the southwestern and middle portions of the catchment.

Figure 4.

Rainfall classification and derived R factor. (a) Existing rainfall categories. (b) R-factor.

Soil erodibility factor (K)

Soil erodibility factor (K-factor) is one of the key parameters of soil erosion modeling that is commonly used in RUSLE model (Panagos et al., 2014). The relationship between the rainfall intensity and the ability of soil to erode is expressed as erodibility factor. Six (6) soil types namely: Chromic Luvisols, Eutric Cambisols, Eutric Leptosols, Eutric Vertisols, Haplic Alisols, and Haplic Arenosols with the corresponding percentage of coverage as 17.27%, 11.27%, 39.94%, 21.28%, 2.51%, and 1.16% were identified in this catchment. Of these soil groups, Eutric Leptosols and Eutric Vertisols cover about 61.22% of the entire region, and the erosion status is severe in these soil groups, which in turn indicates a high erodibility of the soil in the catchment (Fenta et al., 2020). The values of soil erodibility factor (K) ranging from 0.13 to 0.45 were estimated in the study area (Table 3 and Figure 5).

Table 3.

Cover and Management (C) and Support Practice (P) Factors in the Study Area.

Figure 5.

Major soil types and derived K factor. (a) Soil categories. (b) K-factor.

Topographic factor (LS)

In a given catchment, the speed and volume of surface runoff are governed by the topographic conditions (Bekele & Gemi, 2021). The topographic factor is the combination of hill-shade (S) and length of the stream (L). The concept behind the importance of the topographic factor (LS) in estimating the soil loss is that the soil erosion increases with the slope steepness, and the length over which the sensitivity to erosion is valid (Desalegn & Mulu, 2021). The topographic factor (LS) generated from the Digital Elevation Model (DEM) ranges from 0 to 167. The slope length and steepness (LS) increase the velocity of surface runoff causing detachment of soil particles, which in turn lead to erosion. In this catchment, the slope ranges from 0% to 560%, and due to the steepness of the slope, the soil loss is very visible especially for the slope values of more than 11%. As we can see from the Figure 6, the upper portion of the catchment is classified as steep slope whereas the southern part of catchment is low-lying areas. The surface runoff with high velocity is emerged from the upper part of the catchment is causing a severe erosion at the low-lying part of the study area. The ranges of slope (degree) and the corresponding LS-factor values generated in this sudy was presented in Figure 6a and b.

Figure 6.

The slope classification and the derived LS factor. (a) Slope categories. (b) LS factor.

Support practice (P) and cover and management factors(C)

Soil erosion modeling is also depedent of the surface dynamic changes such as cover and management (C) and support practice (P). These parameters are used in the Revised Universal Soil Loss Equation (RUSLE) to quantify the annual soil loss. Support and conservation practices factor (P factor) and cover and management factor (C factor) derived from the major LULC (Figure 7) of the catchment are presented in Figure 7a and b respectively. These factors revealed that the lower part of the catchment is sensitive to soil erosion due to the weakness of soil conservation practices (Chalise et al., 2019; Panagos et al., 2015).

Figure 7.

The derived C factor and P factor in the catchment. (a) C factor. (b) P factor.

The severity of this soil loss was categorized based on the qualitative classification namely; Very Low (red), Low (green), Moderate (yellow), and High (blue) and the corresponding soil loss class was presented in Figure 8a and b respectively. Although the general rate of annual soil loss in this particular study is classified into four, the degree of severity is different from watershed to watershed. The detailed spatial variability of the soil loss for the individual watershed and the computed percentage of the annual soil loss for each watershed are presented in Table 4.

Table 4.

Summary of Soil Loss (ton/yr) and Loss Percentage of Each Watershed.

Figure 8.

Qualitative based spatial variation of soil erosion severity map in Fincha catchment. (a) Qualitative soil loss categories. (b) Quantitative soil loss categories.

The catchment was distributed and divided into small watershed division labeled as W1, W2, W3. . .. W20. As we can see from Figure 9, the severity for the different soil loss classifications obtained in RUSLE model and the qualitative classification of the soil loss as Very Low (0–15 t ha⁻¹ yr⁻¹), Low (15–45 t ha⁻¹ yr⁻¹), Moderate (45–75 t ha⁻¹ yr⁻¹), and High (>75 t ha⁻¹ yr⁻¹) were identified. In terms percentage coverage of soil erosion severity in the catchment, a total eight (those are labeled as W3, W5, W7, W8, W10, W11, W17, and W20) out of twenty of watershed divisions are very vulnerable to the soil erosion as revealed in Table 4 and Figure 10.

Figure 9.

Soil loss class and qualitative classifications.

Figure 10.

The spatial variability of the soil loss of the individual watershed.

The percentage coverage of the soil loss severity for the individual watershed under the qualitative classification is 10%, 15%, 30%, and 45% were identified as very low, Low, moderate, high, respectively. As revealed in the severity map generated in RUSLE model (Figure 10), the cropland and grassland which cover 65% of the total area are highly vulnerable to the soil erosion. The spatial variability of annual soil loss shown in Figure 9 showed that the majority of cropland and grassland in the lower part of the catchment is very sensitive to soil loss and this fact is observed in the detailed severity of each watershed (Figure 10). The spatial variation of the soil loss severity map generated in the RUSLE model has a paramount role to alert land resources managers and all stakeholders in controlling the effects via the implementation of both structural and non-structural mitigations. The results of the RUSLE model can also be further considered along with the catchment for practical soil loss quantification that can help for protection practices.