Soil Erosion and Sediment Export under Rainfed Agriculture Dominated Landscapes: Evidence from the

Didessa Watershed, Southwestern Ethiopia.

Deresa Abetu Gadisaa,b,c,d, Abreham Berta Aneseyeee, Hirpa Abduro Ogoa,b,c,d Habtamu
Eticha Olanaa,b,c, Jiacheng Huanga,b,c, Sisi Yuc, Junyao Suna, Xue Yana,b,c *
a
State Key Laboratory of Plant Diversity and Specialty Crops, Wuhan Botanical Garden
Chinese Academy of Sciences.
b
University of Chinese Academy of Science, Beijing, 100049, China
c
Sino-Africa Joint Research Center, Chinese Academy of Sciences
d
Ethiopian Biodiversity Institute (EBI), 30726, Addis Ababa, Ethiopia.
e
Department of Natural Resource Management, College of Agriculture and Natural
Resource, Wolkite University, Ethiopia.
*Corresponding author:
[email protected]/[email protected]/[email protected]/
[email protected]
Abstract
Global environmental degradation resulting from unsustainable human activities has
become a critical development challenge in the 21st century. Didessa watershed in
southwestern Ethiopia characterized by rugged topography dominated with subsistence
rainfed agriculture illustrates this vulnerability. This study aims to quantify the
spatiotemporal soil loss (SL) and sediment export (SE) (1987-2023) using RUSLE–InVEST
model, identify erosion hotspot sub-watersheds, and estimate reclamation cost to inform
sustainable watershed management. Multi temporal satellite imagery (1987, 2000, 2013,
and 2023), rainfall and soil data were used to estimate SL and SE (t/ha/yr). The study
revealed, 67% of the watershed was occupied with agricultural land. The overall soil loss in
the watershed was increased by 192%.
There was a substantial increase in mean annual SL 163% (8.89 in 1987 to 23.44 t/ha/yr in
2023) and SE 367% (from 2.1 to 9.8 t/ha/yr), illustrating deterioration of the watershed.
ANOVA results showed a statistically significant difference in both SL (F 227, p< 0.0001)
and SE (F 213.62, p< 0.0001) across the study period, showing a steady rise. The rates of
erosion concisely varied among the 26 sub-watershed, with the four (SW-1, SW-17, SW-20
and SW-26) having sever erosion levels (50 -63t/ha), two (SW-19 and SW-6) the lowest
(<20 t/ha) and the rest sub-watersheds having moderate rates. Rainfall, sub-watershed size,
slope gradient and elevation were significantly determining (p <0.05) the erosion rate across
years and land uses. The estimated cost to reclamation these degraded soils and restore the
ecosystem services was about USD 24 million. The severe SL and SE from the watershed
pose a serious threat to agricultural sustainability, water resources, and ecosystem function.
Thus, urgent conservation actions are essential to mitigate the escalating environmental
degradation and its associated economic impacts.
Keywords: Agricultural landscape, InVEST SDR model, Ecosystem service,
Environmental degradation, Sustainable land management
1. Introduction
Environmental degradation is a serious global concern caused by unsustainable agriculture,
deforestation, overgrazing, and impacts of climate change such as floods and droughts
(Atieku & Segbefia, 2024; Kaiser, 2021; Shukla et al., 2019). It's degrading soil fertility and
vegetative cover, with the most devastating effects experienced in susceptible regions such
as Sub Sahara African (SSA) countries (Tully et al., 2015; Yengoh, 2020). Ethiopia
particularly, in the erosion sensitive highlands degradation is accelerated by high rates of
population growth, deforestation for fuel wood and agricultural lands, and incapable
investment in soil conservation (Bishaw, 2001; Nyssen et al., 2004; Wassie, 2020). The
main major contributing components include rain fed agriculture on hilly terrain, low
adoption of sustainable land management (SLM) practices, land tenure insecurity, poor-
resource exploitation, and intense rainfall (Hurni et al., 2005; Teshome et al., 2016).
The staggering scale of land degradation is estimated 33% of the planet's land area is
moderately to severely degraded, and this results in the loss of over 24 billion tonnes of
productive soil annually and economic losses of $6.3 trillion per year due to diminished
ecosystem services and agricultural production (FAO, 2015; Gupta, 2019; Hossain et al.,
2020; Initiative, 2015). The Ethiopian Highlands (which contributes 85% of the Nile's
flow), and the Nile Basin more generally, suffer from catastrophic soil erosion, at an
estimated 1.3 to 1.9 billion tonnes per year (Hurni et al., 2010a; Kidane & Alemu, 2015).
Unmanageable sedimentation inflicts heavy costs on down river infrastructure; e.g., Sudan's
Roseires Dam loses storage capacity 5-10 times faster than predicted, undermining water
security(Bishaw, 2001). In Ethiopia, over 50% of agriculturally suitable highland areas are
degraded, with soil loss rates of 20-200 t/h/yr far greater than the threshold of <10 tonnes
contributing for the estimated 2-3% annual reduction in Agricultural GDP (Haile, 2004;
Hurni et al., 2010a; Tamene, Adimassu, Aynekulu, et al., 2017; Tamene & Vlek, 2008).
The effects of land degradation are ecological, social, and economic in nature. It endanger
food security, causes nearly 10% of anthropogenic greenhouse gas emissions via the loss of
soil carbon, intensify biodiversity loss, and imposes high economic burden (Initiative, 2015;
Pandit et al., 2018; Shukla et al., 2019; UNICEF, 2023). At the local level, in the Nile
Basin, reservoir siltation decrease water storage, cripples hydro power and irrigation
operation, decline water quality, and increases transboundary tensions over scarce water
resources (Assessment, 2020; Attia & Nasr, 2024; Bishaw, 2001). In Ethiopia, it hinders the
Sustainable Development Goals (SDGs) especially; intensifying food insecurity (SDG 2),
rural poverty (SDG 1), climate vulnerability (SDG13), and desertification (SDG 15)
(Boerger et al., 2021; Ethiopia, 2018; Hurni et al., 2005; Nyssen et al., 2004). With such
cascading effect, the control of soil erosion and land degradation is not environmental
issues but a global stability, regional cooperation, and national survival concern especially
for agriculture dependent country's like Ethiopia.
The measurements of soil erosion have undergone important progress in the past decades,
from easy empirical to well informed spatial temporal frameworks that combine a number
of analytical approaches. This was pioneered by the development of the Universal Soil Loss
Equation (USLE) in the 1960s, which provided simple annual loss predictions but had no
spatial and temporal information (Wischmeier & Smith, 1978). The subsequent integration
of geographic information systems (GIS) with the Revised Universal Soil Loss Equation
(RUSLE) in the 1990s was a significant advance, which permitted spatially explicit
mapping of erosion and evidenced basic landscape heterogeneity (Renard, 1997). These
improvements were particularly useful in vulnerable regions like the Ethiopian highlands,
where RUSLE uses identified severe pattern deterioration, with a 50% loss rate exceeding
50 t/ha/yr on half the agricultural land a finding that specifically accounted for observed
decreases in farm productivity and food security (Hurni et al., 2005; Tamene, Adimassu,
Aynekulu, et al., 2017).
Soil erosion poses a serious environmental impact in Ethiopia, specially, in its highland
regions, where it undermines agricultural productivity, degrades water quality, and
compromises long term ecosystem sustainability. Currently, the InVEST model has been
progressively applied to estimate soil erosion and sediment transport changes across various
Ethiopian watersheds. For instance, the Akaki River catchment (Yohannes et al., 2024); the
Winike watershed (Aneseyee et al., 2020) and the Abbay (upper Blue Nile) Basin (Gashaw
et al., 2021), all of which report significant rates of soil loss and sediment export. Disregard
to this progress, there is a demand of comprehensive studies addressing erosion dynamics in
southwestern Ethiopia particularly in the Didessa watershed, a region characterized by
pronounced rain fed agricultural practices, topographic variability, diverse land use
pressures, and socioeconomic vulnerability. Disregard, these knowledge gap hinders the
improvement of region specific conservation plan of action and highlights the need for
targeted research to support sustainable watershed management in this underrepresented
area.
Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) Sediment Delivery
Ratio (SDR) model integration with Revised Universal Soil Loss Equation (RUSLE) and
Geographic Information Systems (GIS) is a significant methodological advance in soil
erosion and sediment transport assessment (Cong et al., 2020; Sharp et al., 2020). This
approach overcomes the limitations of traditional methods by measuring sediment delivery
routes through spatially explicit connectivity, value of sediment retention as an ecosystem
service, and economic costs associated with land degradation. In places with limited data,
like the Didesa watershed, this approach offers valuable policy insights (Gashaw et al.,
2018). Hence, the recent approach is applicable in both data rich and data limited
environments (Aneseyee et al., 2020; Caro et al., 2020; Cong et al., 2020; Hamel et al.,
2024; Hamel et al., 2020; Hamel et al., 2021; He et al., 2023; Posner et al., 2016; Wu et al.,
2022; Xiao et al., 2023). It pinpoints significant source areas that modify the most to
sediment downstream and evaluate the cost effectiveness of different conservation options
crucial for making the best use of limited watershed management resources(de Magalhães
et al., 2022). Current studies exhibit the model’s quality to produce clear, location specific
information, maps highlighting conservation priorities, and economic estimates of sediment
buildup in dams (Cong et al., 2020; Wu et al., 2022). This close the gap between scientific
research and real world policy decisions in erosion-prone regions. This study aims to assess
soil erosion and sediment dynamics in the Didessa watershed with the following objectives;
(1) Spatiotemporal analysis of annual soil loss and sediment export trend using RULES-
InVEST modeling, (2) Identify the vulnerable subwatersheds, (3) asses soil loss and
sediment export severity within the watershed to set conservation priority and (4) valuation
of restoration cost to inform sustainable watershed management in support of sustainable
development goals.
2. Materials and methods

2.1. Description of the study area

The study was conducted in the Didessa Watershed, situated in the Oromia Regional State
of southwestern Ethiopia, inside the upper reaches of the Nile River Basin (Fig.1). The
watershed covers an area of 1780 km² and characteristics by diverse topography, with
elevations ranging from 1470 m to 3031 m (mean=1852.63 m) and slopes varying from 0°
to 64°. It receives an annual mean rainfall of 1717 mm, with mean maximum and minimum
temperatures of 28.7 °C and 11.7 °C, respectively (Fig. 2). Rain fed agriculture and animal
husbandry are the main source of economic activities, experienced across all attitudinal
ranges using integrated farming systems. The watershed connects with six administrative
districts, having remnants of moist evergreen Afromontane forests, selected and conserved
as a national forest priority areas (Bayissa, 2020; Jiru, 2021). However, the forests have
been degraded and fragmented due to high rain fed agricultural practices, land use/land
cover change and human encroachment (Jiru & Molla, 2022).
Figure 1. Study area site
Figure 2. Climatic Characteristics of the Didessa Watershed
2.2. Soil loss and sediment delivery modeling
The study watershed's soil loss and sediment delivery were measured and mapped using the
InVEST model. This model apply several information, considering raster land use types, a
Digital Elevation Model (DEM), the Revised Universal Soil Loss Equation (RUSLE)
factors. The summary of the InVEST model, utilized to assess soil loss and sediment in this
study is illustrated in (Fig. 3). The raster and vector data used as model input were projected
to the UTM coordinate system, WGS84 datum, zone 37 N. While the model provides
reliable outputs, it is heavily dependent on the USLE, which is constrained to overland
(sheet and rill) erosion processes and is also sensitive to soil erodibility (K) and Index of
connectivity (ICO) limits (Cavalli et al., 2013; Renard, 1997; Vigiak et al., 2012).
Figure 3. Framework employed for soil loss and sediment Delivery Ratio (SDR) analysis
using the RUSLE-InVEST Model in Didesa watershed
2.2.1 Soil loss estimation
The SDR is linked with the RUSLE (Panagos et al., 2016; Wischmeier & Smith, 1978),
RUSLE utilize five input variables to estimate soil loss (Eq. 1).
USLEi= (R*K*LS*C*P)i Eq. 1
Where USLEi= Mean annual soil loss (t/ha/yr), R=Rainfall erosivity (MJ mm/ha/year), K =
Soil erodibility (MJ mm/t/ha/yr), LS = Topographic factor (slope length and gradient)
(dimensionless), C = land cover (dimensionless) and P = land management practices
(dimensionless).
A) Erosivity factor (R)
Rainfall data records were collected from meteorological stations to determine rainfall
erosivity (R factor), one of the main causes of soil erosion, using a spatial regression
equation (Eq. 2) and average annual rainfall (Hurni, 1985). The meteorological information
comprises data for 1987 - 2023 that were collected from five meteorological stations
(Jimma, Gatira, Agaro, Denbi, Bedelle, and Sigmo) around and within the study watershed.
The Inverse Distance Weighted (IDW) interpolation method was subsequently applied to
produce a spatially distributed raster map of rainfall erosivity. This raster map function as
an input for the InVEST model, specifically for the reference years 1987, 2000, 2013, and
2023 (Supplementary figure.1).
R = -8.12 +0.562P Eq. 2
Where R = erosivity factor; P = mean annual rainfall in mm/year
B) Soil erodibility factor (K)
To estimate soil erodibility factor (K) for the watershed, the area was stratified based on
land use, topography, soil types and vegetation cover to ensure the collection of
representative soil samples collection. A total of 211 composite soil samples (Fig 4a) were
collected and analyzed for organic carbon content and texture in the soil Laboratory of
Jimma university, College of Agriculture and Veterinary Medicine. The organic carbon
content was determined using a Walkley-Black method (Walkley and Black 1993), while
soil texture was analyzed using the Bouyoucos hydrometer method (Bouyoucos, 1927).
The K- factor was derived using the formula Eq. 3 (Renard (1997) (Supplementary table.1),
a widely recognized and applied method for estimating soil erodibility in tropical regions.
Then, a spatially distributed raster map of the K- factor for the entire watershed was
generated using IDW interpolation (Fig. 4a) and subsequently used as input for InVEST
model.
K=¿ ¿ ¿ Eq. (3)
Where, K = soil erodibility, OM = Soil organic content (%),a = organic matter (%); b = Soil
structure code: (1) very structured or particulate, (2) fairly structured, (3) slightly structured,
and (4) solid; c = Soil profile permeability code: (1) rapid, (2) moderate to rapid, (3)
moderate, (4) moderate to slow, (5) slow, and (6) very slow, and
M =(silt ( % )+ very fine sand ( % ))¿).
C) Digital Elevation Model (DEM)
DEM was used to evaluate the impact of watershed topographic features, particularly slope
length and gradient on soil erosion and sediment export. For accurate assessment of soil
erosion and sediment export the best quality, fine resolution DEM data are essential. A 30m
resolution DEM image was obtained from United States Geological Survey (USGS) data
server (https://earthexplorer.usgs.gov). The DEM underwent processing to fill sinks and
was subsequently clipped to the study watershed boundary and the study area watershed
soil types were identified from soil map of Ethiopia (Fig. 4b) using the of spatial analysis
tools in ArcGIS 10.8.1.

Figure 4. The soil erodibility factor of the watershed (A) and Major soil types (B).
D) Management Practices factor (P)
Mapping degraded and preserved areas was necessary for measurement of the management
practices factor (P), which was established based on the kind of conservation measures used
within the watershed.The P- factor values were evaluated using the interaction between
major land cover and slope, as outlined by Wischmeier and Smith (1978), and further
validated through field observation to characterize existing conservation practices. The P-
values ranged from 0, indicating areas with efficient conservation measures and minimum
erosion hazard, to 1, correspond areas with minimal or no conservation activity (Hurni,
1985; Wischmeier & Smith, 1978).
E) Cover management (C) factor
The cover management factor (C) serves as a quantitative measure of the influence of
vegetation cover, and management practices on soil erosion processes (Panagos et al.,
2015). LU/LC data provide critical insights into cropping patterns, fallow land, forests, and
water bodies, which are important for development planning and erosion assessment
(Ganasri & Ramesh, 2016). Remote sensing and GIS techniques were utilized to generate
LU/LC maps of the watershed for the year 1987, 2000, 2013 and 2023 (Supplementary
figure.2 and table.2). Landsat images for these years, spaced approximately 13 years apart
to capture LU/LC dynamics influenced by significant policy shifts, were acquired from the
United States Geological Survey (USGS) (https://earthexplorer.usgs.gov). To check up
classification accuracy, 500 Ground Control Points (GCP) were collected from the major
LU/LC types. Image classification and mapping were conducted using QGIS (version
3.36.3), ERDAS Imagine (version 16.8.0.2100), and ArcMap 10.8.1. A supervised
classification approach, utilizing the Mahalanobis distance algorithm, applied for LU/LC
change classification was used. Classification accuracy was assessed using overall
accuracy, the kappa coefficient (KC), and user’s and producer’s accuracy metrics (Lillesand
et al., 2015).. The kappa coefficient results were interpreted following the guidelines of
(Viera & Garrett, 2005) (Supplementary table.3). The average C-factor for each LU/LC
class was calculated using the Normalized Difference Vegetation Index (NDVI) and the
following equation (Eq. 4) (Durigon et al., 2014).
(-NDVI+1)
C-factor = Eq. 4
2
The C-value ranged from 0.33 to 0.76 in the watershed. A higher NDVI value indicates
higher vegetation cover and vice versa.
Where: NIR = the surface spectral reflectance in the near-infrared band and RED = surface
spectral reflectance in the red band. This factor value ranges between 0 and 1, zero refers to
very strong land cover effects and one indicates no cover or bare surface (Negese, 2021).
2.2.2. Sediment export
The InVEST SDR model uniquely enables the analyse of soil loss and sediment export from
each land use type contributing to streams, reservoirs, and other water bodies (Cong et al.,
2020). The sediment delivery, is estimated using combinations of factors such as index of
connectivity (IC), which used to calculate the sediment Delivery Ratio (SDR), as a function
of the up slope area (Ud) and down slope flow path (Dd) to the streams, derived from DEM
(Borselli et al., 2008). The Sediment delivery ratio for a pixel i is computed directly from
the IC using a sigmoid function (Borselli et al., 2012; Vigiak et al., 2012) as described in
Eq. 5 below. This investigation qualify insights into the hydrological connectivity between
an area and the stream network.
SDR max
SDRi = 1+ exp( ICo−ICi ) Eq.5
k
Borselli k (kb) and Borselli ICo (ICo and ICi) are parametric quantities that define the shape
of the SDR-IC relationship, with a default value of 0.8 to minimize the number of
parameters.
The sediment export from a pixel i (Ei), represents the portion of eroded sediment that that
reaches a stream, calculated by using Eq.6. The total catchment sediment export (Et) is
estimated by using Eq.7 below.
Ei = uslei *SDRi Eq. 6

Et= ∑ Ei Eq.7
i

2.3. Model validation

To evaluate the applicability and reliability of the InVEST Sediment delivery ratio model,
simulated sediment delivery rates were checked with observed data collected from gauged
stations collected by the Ministry of Ethiopian Water and Energy (MOE). Model
performance was assessed using mostly adopted statistical metrics, including Nash-Sutcliffe
efficiency (NSE), Coefficient of Determination (R 2) and Average Percent Bias Error
(PBIAS) (Moriasi et al. (2007b). These metrics were measured using Eq. 9 and 10 below.

PBIAS=¿ Eq. 8
NSE=1−¿ Eq. (9)
The model exhibit better performance when PBIAS and R² values are low. Specifically, the
model exhibits very good performance when R2 and RMSE values are > 0.75 and PBIAS is
±10%. Conversely, the model is reasoned dis satisfactory and inapplicable if R2 and RMSE
are below 0.50 and PBIAS is below 25% then the model is unsatisfying and not applicable
(Moriasi et al., 2007a; Munoth & Goyal, 2019).
The InVEST model provides annual sediment delivery rates, whereas the observed data are
 typically registered on a daily timescale. To check consistent comparison, daily observed
data were aggregated into annual sediment delivery rates (t/ha) using an empirical
relationship between stream flow (m3/s) and sediment concentration (mg/L) (Aneseyee et
al., 2020; Sadeghi et al., 2008), as computed by the following equation (Eq. 10).
c
SC=b ×Q Eq. 10
Where: SC (ton/day) = sediment loss, Q stream flow (m 3/s), b and c are determined from
sediment concentration (g/ml) and stream flow.
2.4. Economic cost of soil erosion rehabilitation
The replacement costs of fertilizer (such as Nitrogen and Phosphorus nutrients) lost due to
soil erosion from the agricultural lands was estimated by multiplying the volume of eroded
soil (t/ha) by the cost of equivalent nutrients in Diammonium Phosphate (DAP) and urea
fertilizers (Selassie & Belay, 2013). According Kefyalew (2011), the average use
DAP/Urea fertilizer of 100kg/ha will be expected to restoration of soil fertility and the
current market price of 4,032 (ETB) (~72$USD, 1USD = 56 birr), this amount of fertilizer
is applied. A historical annual fertilizer price for the year’s consecutive years (1987, 2000,
2013, and 2023) was obtained from Ethiopian Statistic Services
(http://www.statsethiopia.gov.et/) to analyze trends in soil fertilizer restoration costs over
time.
The secondary data used to evaluate economic valuation of erosion control were collected
from global Economics of Ecosystems and Biodiversity (TEEB) database
(http://www.teebweb.org) (McVittie & Hussain, 2013), which provides valuations for
various ecosystem services within typical biomes. The nearest equivalent biome from the
TEEB database was assigned to each LU/LC type within the watershed to estimate their
respective erosion control values. Based on this approach, the erosion control services were
valued at an estimated cost US$360/ha/year forest land, US$55.89/ha/year for grazing land
and US$30/ha/year for bush land. The total cost of soil erosion treatment was calculated
based on the model developed by based on the model developed by (Telles et al. (2013) , as
outlined equation 11 below.
m
Ct= ∑ (Pi × Qi) Eq. (11)
i=1

When: Ct= the required Cost to treat soil erosion; C i= Prices of treatment cost (Data of
TEEB valuation of erosion control and prices of fertilizer for agricultural land in Ethiopia);
Q= quantity of soil erosion estimated by the InVEST model.
 2.5. Statistical analysis
Geospatial analysis was carried out using GIS software (version 10.8.1) to generate maps
representing biophysical parameters and land use dynamics. Statistical regression analysis
was used and checked to correlate the observed and predicted soil loss within the study
watershed. Soil loss and sediment export were quantified using InVEST-SDR model
software (version 3.14.2) (Natural Capital Project, 2025;
https://naturalcapitalproject.stanford.edu/software/invest). The influence of key variables
including altitude, area, slope, soil types, and rainfall on soil loss and sediment export was
assessed using one way Analysis of Variance (ANOVA). Statistical software analyses were
executed using R software (version 4.4.2) (Team, 2020).
3. Results
3.1. Model validation
The high coefficient of determination result reveals (R² = 0.85; p > 0.05) a strong positive
correlation between the predicted and observed soil loss values (Fig.5), indicated the
model's performance in simulating soil loss within the watershed. Quality and reliability of
the model was further confirmed by a low root mean square error (RMSE) of 5.30 t/ha and
a negligible percent bias (PBIAS) of 0.29 (Fig.5). To validate the reliability of the model,
registered sediment data was taken from five local gauging stations found within and near
study watershed (Agaro, Bedelle, Gimbe, Toba, and Yebu) (Supplementary table.4) further
validate the model's predictive accuracy.

Figure 5. The variation of modelled versus observed soil loss trends (2000-2023) in the
 Didessa watershed.
3.2. Soil loss and sediment export rates in Didesa watershed
The spatiotemporal analysis of soil loss and sediment export exhibited increase in
degradation in Didessa watershed between 1987 and 2023 (Fig. 6 & 7). In the past 36-year
period, the mean soil loss by land use was increased by 350% (from 9.85 in 1987 to 44.38
in 2023 t/ha), and sediment export rose by 309.5% (from 2.1 to 8.6 t/ha) (Fig, 6;
Supplementary table. 5). Spatially, the problem was more pronounced; in the western parts
of the watershed associated with high human activity and steep terrain (Fig. 6 A-B).
Moreover, our analysis showed a clear erosion pattern across land use/cover type
(Supplementary table.4). The rain fed agriculture dominant land use types, agricultural
lands within the watershed had the greatest soil loss rates, followed by settlement and
grazing lands, demonstrating the effects of cultivation and human use of land on soil
detachment. On the other hand, sediment export was higher in agricultural land, followed
by wetlands and grazing lands. The relationship of soil loss and sediment export illustrates
how land use affects the different aspects of the erosion process. Over the past 36 year’s
period, rate of soil loss in agricultural land was raised nearly 200% (form 22.67 t/ha in 1987
to 67.78 t/ha in 2023), while forest land experienced 387% increase (from 2.51 t/ha to 12.22
t/ha) (Supplementary table.5). Maximum values doubled from 24.0 to 48.1 t/pixel (soil loss)
and 10.2 to 21.8 t/pixel (sediment export), showing severe degradation over 36 years.
Figure 6. Spatiotemporal dynamics of soil loss (A-B) and sediment export (C-D), and (D)
in Didesa watershed.
Figure 7. Mean soil loss (A) and sediment export (B) rates by land use/cover (t/ha/yr)
ANOVA results revealed a significant differences in soil loss across the years (F(1.54,
38.59) = 226.79, p< 0.0001, η²g = 0.57), and in sediment export (F(1.45, 36.18) = 213.62,
p< 0.0001, η²g = 0.57). Post hoc Bonferroni tests showed significant difference between all
years (1987, 2000, 2023 and 2023), confirming a steady rise in both soil loss and sediment
export (Supplementary figure.3). ANOVA results showed significant effects of topographic
and climatic factors on the soil loss and sediment export within the watershed . The altitude
(F = 8.815, p = 0.004), area size (F = 7.379, p = 0.008), slope gradient (F = 7.410, p =
0.008) and mean annual rainfall (F=118.281; p =<2e-16) were statistically significant
predictor of erosion dynamics.
The temporal analysis of soil loss and sediment export by land use types across the study
period (1987-2023), showed important patterns that remained constant throughout the study
period (Fig. 8). Agricultural lands were persistently the highest contributors to soil loss
within the watershed (Agricultural > Settlement > Bush land > Wetland > Grazing >
Forest), underscoring the extent that cultivation impacts soil detachment (Supplementary
table 4). At the same time, agriculture lands were consistently the dominating by the
sediment exporter in watershed (Agricultural > Wetland > Settlement > Grazing > Bush
land > Forest), indicating that wetlands often act as natural transport pathways for
sediments. The forest had the lowest soil loss and sediment export across all years of study,
illustrating their role as protective systems. The overall soil loss in the studied watershed
was increased form 2.40 million tons in 1987 to 7.01 million tons in 2023. This shows that
the area has affected by soil erosion even if, it has relatively good vegetative covers.
 Figure 8. Temporal changes in soil loss (A) and sediment export (B) percentages by land
use/land cover types across the four study years
3.3. Predictor of Soil loss and Sediment Export in the watershed
The correlation analysis showed strong positive associations between agriculture use and
both soil loss (R² = 0.82, p< 0.01) and sediment export (R² = 0.72, p< 0.01) (Fig.9),
exhibited agriculture accounts for 82% and 72% of the variance in various erosion
processes. Oppositely, forest land showed importantly inverse relationship with both soil
loss (R² = –0.69; p< 0.01) and sediment export (R² = –0.56; p< 0.01). Similarly, wetlands
(R² = –0.87; p< 0.01), bush-land (R² = –0.78; p< 0.01), and grazing land (R² = –0.68; p<
0.01) all showed statistically significant (p< 0.01) negative relationships with soil loss,
demonstrating their protective roles against erosion processes.
 Figure. 9. Relationship bbetween Land Use/Land Cover Change and Soil Loss
3.4. Erosion severity and management priority hotspots within the watershed
The geospatial analysis showed severe soil erosion risk throughout the study watershed,
more noticeable in the western part followed by the northern section of the watershed with
extreme soil loss rates of 48.1 t/pixel on mid-slope (>15° slope) deforested areas and
maximum sediment exportation peaks at 21.8 t/pixel in downstream in the watershed
(Fig.10). These hot spots occupied relatively small area of the total watershed, contributing
to higher of amount of soil loss and sediment export.
Eventually, in the low erosion severity category (<10 t/ha), soil loss exhibited escalated by
500% (from 2.88 to 17.33 t/ha), indicating that even traditionally stable areas (i.e., low
erosion rate) are degrading. Moderate severity zone (10-30 t/ha) experienced about 140%
rise (from 22.77 to 55.44 t/ha), while high-severity zones (30–50 t/ha) increased by 150%
(47 to 117.44 t/ha), indicating extreme pressure on erosion prone landscapes. The
progression raised the watershed-wide average from 24.21 to 63.40 t/ha (an increase of
160%), with seven sub-watersheds now classified as high-severity class (Supplementary
table 6). However, the majority of the sub-watersheds continued to be classifying as low-
severity class because of vegetation cover (Figs. 10-11; Supplementary table.6).
A spatiotemporal analysis of soil erosion among the sub-watersheds (SWs) revealed both
soil loss and sediment export with three distinct phases of degradation (Fig. 11). In the first
phase (1987-2000) soil loss and sediment export rates were generally stable, with inter-
annual variability within and between sub-watersheds and increased by 26% (1987-2000).
The second phase revealed a rapid increase in soil loss rates between 2000-2013, with the
overall increase of average soil loss 38%. Although, no significant increases in soil loss
rates were evident in the subsequent third period post-2013, soil loss was increased by 54%
(Supplementary table.7). Higher than the baseline (1987) in most severely affected sub-
watersheds. Of the SWs, SW-1 was the most extreme, with soil loss and sediment export
rates from 14.2- 62.68 t/ha (341% gain between 1987 - 2023); SW-20 accelerated from
26.48-60.68 t/ha (129% gain).
Figure 10. Soil loss in the sub-watersheds (A), sediment export in the sub-watersheds (B),
soil loss severity level in the sub-watersheds (C) of the year 2023
Topographic based analysis indicated that both elevation and slope gradients demonstrate a
geospatial response to the escalating soil loss across the watershed. Relatively lower
elevation sites, experience a 71% increase over time (27.22 to 46.67 t/ha) whereas mid-
elevation sites show cyclical behavior improving temporarily to 19 t/ha in 2000, then
degenerate to 41.11 t/ha in 2023. Higher elevation sites showed a 42% decrease over time
(46.67 to 27.22 t/ha) due to better vegetation cover, and the system stabilizing at 38.33 t/ha
overall (Supplementary table. 8)
Slope gradient based soil loss analysis showed widespread intensification of erosion with
gentle slope raised + 156% (7.56 - 19.33 t/ha), in the moderate increased +178% ((18.11 -
50.44 t/ha), and in the steep slopes increased +65% (20.89 - 76.33 t/ha).The net effect of
309% increase in mean soil loss (11.90 - 48.7 t/ha) showed that degradation occurred across
the watershed with varying degree (Supplementary table 9). These results show: 1. widened
non linear elevation dependent erosion responses; erosion vulnerability for slopes generally
increased and; the need for gradient specific conservation strategies, particularly for mid
elevation zones and steep slopes where the greatest rates of change were noted.

Figure 11. Temporal trends in mean annual soil loss and sediment export (t/ha/yr) across the
sub-watersheds (1987, 2020, 2013, 2023).
3.5. Economic cost of soil erosion within the watershed
The analysis of the 36-year erosion reclamation cost revealed variations over time across
major land use/cover types associated with soil loss rate (Fig. 12 and Supplementary
table.10). The entire watershed reclamation cost was increased by 476% (from 554 USD/ha
in 1987 to 3,193 USD/ha in 2023). The reclamation costs for agricultural and grazing lands
have considerable increases throughout time. Agricultural land showed the most dramatic
increase with over six-fold, ranging from $770.8/ha in 1987 to $4,880.2/ha, with an
intermediate peaks reached $1,108.5/ha (2000) and $1,940.8/ha (2013). Likewise, grazing
land experienced overall price inflation, ranging from $325.8/ha (1987) to $2,177.5/ha
(2023) with intermediate highs of $852.3/ha (2000) and $1,404.0/ha (2013). These
increases are consistent with greater agricultural intensification and overgrazing, suggesting
the restoration of human-managed systems is becoming increasingly costly due to
cumulative degradation. Forest and bush lands had different trajectories in the coast, and
both were now alarming. The forest land reclamation costs increased dramatically, from
$903.6/ha in 1987 to $4,399.2/ha in 2023, with intermediate values of $1,778.4/ha in 2000
and $1,321.2/ha in 2013. For the bush land, the variability was considerable, increasing
from $217.2/ha in 1987 to $1,313.4/ha in 2023, with an intermediate dip to $496.5/ha in
2013 and $509.4/ha in 2000 (Supplementary table.8). It was evident that the sharp increase
in costs in natural ecosystems in the last 10 years suggests the increasing challenge
associated with restoring ecosystems due to climate change, biodiversity loss, and
ecosystem fragmentation
Figure 12. Temporal dynamics of reclamation cost (USD/ha/yr) for different land use/cover
types
4. Discussions
The spatio temporal dynamics of soil loss and sediment export in the rain fed agriculture
dominated landscape Didessa watershed aligned with global and regional trends determined
in area undergoing fast land-use transitions and high rain fed agricultural intensification
(Biggelaar et al., 2004; Lal, 2009; Pimentel & Burgess, 2013). In Ethiopia, soil erosion
remains a serious environmental challenge, specially, in the Northern highlands where steep
terrain, high rainfall erosivity, intensive cultivation, and population pressure have led to
terrible degradation, with annual losses ranging from 5 to 42 t/ha (Fenta et al., 2021;
Haregeweyn et al., 2015; Hurni et al., 2010b). Rain fed agricultural practices in these
highlands lose approximately 58 t/ha/yr (Hurni et al., 2015), exceeding both national and
tolerable loss thresholds 5–11 t/ha (Hurni, 1983).
In contrast, the current study (Didessa watershed) in Southwestern Ethiopia show lower
erosion rates (23 t/ha/yr) than the Northern Highlands (42 t/ha/yr), mostly to its dense
vegetation cover, forested areas, relatively lower population density, and gentler slopes
(Fenta et al., 2021). However, due to different anthropogenic factors, soil loss in the
Didessa watershed has enhanced significantly, rising from 9.85 t/ha in 1987 to 44.38 t/ha in
2023far surpassing the tolerable threshold of 5-11 t/ha (Hurni, 1983). Similarly, sediment
export increased from 2.1 t/ha to 8.6 t/ha over the same period. These trends align with
findings from comparable watersheds, including Fakisi (35.0 t/ha (Tsegaye et al., 2024),
Koga (27.3 t/ha (Gashaw et al., 2020), and Gumara (42.67 t/ha; (Belayneh et al., 2019).
Even higher erosion rates (47-107 t/ha) have been reported in sub-basins such as Koga,
Erer, and Muger, reflecting variations in land use, topography, soil characteristics, and
methodological approaches (Bewket & Teferi, 2009; Gelagay & Minale, 2016; Tamene,
Adimassu, Ellison, et al., 2017).
Soil erosion in Ethiopia is mainly dominated by the interaction of soil erodibilty (K- factor)
and rainfall erosivity (R- factor), aggravated by steep slopes,, intense seasonal rainfall,
deforestation, and unsustainable land degradation practices (Haregeweyn et al., 2015).
Nitisols, vertisoils, and cambisols, are some of the dominant soil types found in the country
and that have various degrees of erodibility. Southwestern highlands dominant soil types,
Nitisols have low K values because of high structural stability and its organic matter
content, but sandy and silty soils in arid regions are very erodible because of their low
cohesiveness (Fenta et al., 2021). Shallow and erodible soils (Cambisols), are particularly
sensitive to erosion when subjected to aggravated agriculture and overgrazing on steep
slopes. This condition is exacerbated by decreasing soil organic matter, which
encouragement K values (Tsegaye et al., 2024). This concept highly agreed with the current
study result in Didessa watershed dominated by Nitisols and whereas Cambisols relatively
low in coverage which clearly explain the status soil erosion result.
High rainfall with unpredictable disturbance promote to high unpredictable erosivity on
bare or disturbed land, while rising rainfall intensities associated with climate change are
increasing R values overtime. Areas with high rain fall erosivity (R-factor) and soil
erodibility (K -factor) values coincide with steep slopes (high LS factor) and low vegetation
cover (high C -factor), erosion rates often surpass 100 t/ha/yr (Haregeweyn et al., 2015).
The Blue Nile Basin one of the Ethiopia’s most studied regions, frequently records R values
above 1000 MJ/mm/ha/yr and moderate to high K values, resulting in severe soil loss on
unprotected croplands. In contrast, areas such as Southwestern Ethiopia (current study area)
in the moist Afromontane forest zone, despite high rainfall, exhibit lower erosion risk due to
the dominance of less erodible Nitisols and protective forest cover, though deforestation
and land conversion rapidly undermine this natural resistance whereas, shallow, erodible
soils and high rainfall erosivity has caused gully formation and land degradation in northern
Ethiopia.
The effect of land use types and the cover management importantly causing erosion
intensity, mostly in agricultural and settlement land areas in Didessa showing the highest
soil loss rates 67.87 and 43 t/ha/yr, respectively, In contrast, forest and bush lands remained
stable. These study are consistent with studies from Gilgel Gibe, where agricultural lands
recorded soil loss of 43.48 t/ha (Aneseyee et al., 2020; Demissie et al., 2013; Nut et al.,
2021). Still, the determined rates was lower than extreme cases in recently deforested areas,
while losses can exceed 130 t/ha (Alemu & Melesse, 2020; Aneseyee et al., 2020; Demissie
et al., 2013). Land use plays a critical role in determining erosion severity. The severity
level of soil loss varies across various regions, as proved by multiple studies. However,
Didessa watershed severity report indicated about 97.4% of the area was categorized as low
to high erosion risk (4.15% low, 76.02% medium, and 19.82% high), while only 3.2% fell
under extreme erosion categories coupled to slopes and altitudes. But, a study report from
Guda Watershed where reported 41.29% of the study area practiced intense erosion results,
(8.39% and 9.42% categorized as high and very high severity, respectively, and only
35.29% as low risk (Kerbe et al., 2023). Study from Dembecha district of Northwestern
Ethiopia, reported 84% of the area proven low to medium erosion risks (24% very slight,
49% slight, and 11% moderate) (Zerihun et al., 2018). The severity of erosion across these
regions determines the critical role of land cover management. Higher erosion rates
correlate strongly with land use, especially in settlements (C-factor: 0.42) and agricultural
lands (C-factor: 0.40), which showing the lowest vegetation cover, as indicated by NDVI
trends (Aneseyee et al., 2020; Zhang et al., 2021). The aggressive and unsustainable
agricultural practices with highly modified land use are dramatically increased sediment
export from watersheds, posing important risks to aquatic ecosystems, reservoir capacity,
and hydro power infrastructure. A prominent example is the Gilgel Gibe River, which
contributes an annual siltation load of 277,000 tonnes to the Gibe I Dam (Aneseyee et al.,
2020). Sediment export from this watershed has increased by 309%, raised from 2.1 t/ha in
1987 to 8.6 t/ha in 2023. Anthropogenic pressure like land use and land cover dynamics and
poor agricultural practices, endanger ecological capacity, socioeconomic constancy, and
downstream ecosystems. Unreasonable sedimentation leads to water contamination,
reservoir siltation, and shortened river life, finally reducing aquatic biodiversity and water
accessibility. This study result agreed with the report from Winike, the Gilgel Gibe River,
partially contributes significant higher siltation due to factors such as soil properties,
vegetation cover, rainfall intensity, topography, and inadequate land management
(Aneseyee et al., 2020).
Environmental and economic challenges caused by high rates of LU/LC modifications, with
soil erosion estimated to USD$ 24 million in Didessa watershed annually over 36 years for
reclamation cost of soil degradation. The improvement to overcome the soil fertility
degradation, sedimentation, and other economic losses that imposes heavy socioeconomic
and biophysical burdens important (Lal, 2004; Pimentel, 2006) with costs projected to
increase further (Nkonya, 2019). In agricultural practice dominant ecosystem, soil erosion
decreases productivity, necessitating high priced amendments and fertilizers while
diminishing arable land, exacerbating food insecurity, and inflating prices (Foley et al.,
2005; Lal, 2004). In addition to unbalance agricultural practices, sedimentation affects
infrastructure, decreases water quality, biodiversity loss, desertification and disrupts
ecosystems services (Pimentel, 2006), (Abebe & Woldemariam, 2024; Pachauri et al.,
2014). Hence, quantifying erosion in economic terms can strengthen policy engagement and
attract investment in conservation efforts. Sustainable land management via afforestation,
SWCs, and climate-smart agriculture is-useful to mitigate erosion, preserve watershed
integrity, and ensure long-term food security (Foley et al., 2005; Pachauri et al., 2014).
4. Conclusion
This study assessed the impact of soil erosion and sediment export under rainfed
agricultural dominated landscapes in the Didessa Watershed, from 1987 to 2023 using
InVEST Sediment Delivery Ratio (SDR) model in conjunction with ArcGIS. The current
findings revealed the significant rise in both soil loss (SL) and sediment export (SE) with
their average rates rising by approximately more than 100%, throughout the study years.
The primary reasons were the high conversion of forest, bush land, wetland, and grazing
land cover types to the dominant of high agricultural practice and settlement land cover
types, which led to reduced cover of vegetation and organic matter and increased
vulnerability to erosion. Spatial analysis revealed clear spatio temporal disparities: Soil loss
and sediment export were lowest in 1987 and highest in 2023, with the most terrible effects
clustered in the western, eastern and northern watershed but southern and central parts
relatively lowest rates. Agricultural, settlement, and grazing land were positively correlated
with erosion, but forest and wetland land played a key buffering role.
The overall soil loss in the studied watershed was increased form 2.40 million tonnes in
1987 to 7.01 million tonnes in 2023. This shows that the area has affected by rainfed
agriculture dominated with by 192% of soil erosion even if, it has relatively good vegetative
covers. Sediment export increased from a 324.00 tonnes to 433.00 tonnes thousands
annually between 1987 and 2023. The cumulative effect of soil erodibility (K-factor) and
rainfall erosivity (R-factor), slopes (5°-64°), and land use/land cover (LULC) modification
highlight the importance of site specific SWC interventions. Attention has to be directed
 towards degraded and intensively cultivated areas to control erosion impact problems for
sustainable land management. Downstream sediment accumulation presents the dominant
risk to dam structures, for example, the Grand Ethiopian Renaissance Dam (GERD). Active
watershed management by physical and biological SWC interventions, local level
institutionalized land use planning, intensive agriculture, advanced and mechanized
irrigation and sustainable land management strategies are therefore urgently called for to
enhance ecological stability and safeguard local livelihoods. Sustainable land management
with intention soil and water conservation, and climate-smart agriculture is essential to
mitigate erosion, preserve watershed integrity, and ensure long-term food security,
demanding urgent adoption to safeguard ecological and socioeconomic stability.
References
Abebe, T. G., & Woldemariam, A. (2024). Erosion spatial distribution mapping and sediment yield estimation using
RUSLE and Arc GIS of Ayigebire watershed, North Shewa zone of Amhara region, Ethiopia. Water-Energy
Nexus, 7, 124-134.
Alemu, W. G., & Melesse, A. M. (2020). Impacts of longterm conservation measures on ecosystem services in
Northwest Ethiopia. International Soil and Water Conservation Research, 8(1), 47-55.
Aneseyee, A. B., Elias, E., Soromessa, T., & Feyisa, G. L. (2020). Land use/land cover change effect on soil erosion
and sediment delivery in the Winike watershed, Omo Gibe Basin, Ethiopia. Science of the Total Environment,
728, 138776.
Assessment, E. P. (2020). Harnessing Continued Growth for Accelerated Poverty Reduction. World Bank, 36-61.
Atieku, J., & Segbefia, S. (2024). Land Degradation: A Global Challenge and its Effects on Humankind and the
Environment, with a Special Focus on Ghana. Indiana Journal of Humanities and Social Sciences, 5(7), 58-
64.
Attia, M., & Nasr, M. (2024). Impacts of Upper Nile Mega Dams on Agricultural Environment in Downstream
Country. In Environmental Consequences of International Conflicts: The MENA Region (pp. 209-238).
Springer.
Bayissa, B. T. (2020). Assessment of Woody Species Diversity, Structure and Regeneration Status of Setema Natural
Forest, Setema District, Southwest Ethiopia.
Belayneh, M., Yirgu, T., & Tsegaye, D. (2019). Potential soil erosion estimation and area prioritization for better
conservation planning in Gumara watershed using RUSLE and GIS techniques’. Environmental Systems
Research, 8(1), 1-17.
Bewket, W., & Teferi, E. (2009). Assessment of soil erosion hazard and prioritization for treatment at the watershed
level: case study in the Chemoga watershed, Blue Nile basin, Ethiopia. Land degradation & development,
20(6), 609-622.
Biggelaar, C. d., Lal, R., Wiebe, K., Eswaran, H., Breneman, V., & Reich, P. (2004). The global impact of soil erosion
on productivity. II. Effects on crop yields and production over time.
Bishaw, B. (2001). Deforestation and land degradation in the Ethiopian highlands: a strategy for physical recovery.
Northeast African Studies, 7-25.
Boerger, V., Bojic, D., Bosc, P., Clark, M., Dale, D., England, M., Hoogeveen, J., Koo-Oshima, S., Mejias Moreno, P.,
& Muchoney, D. (2021). The State of the World’s Land and Water Resources for Food and Agriculture–
Systems at breaking point. Synthesis report 2021.
Borselli, L., Cassi, P., & Torri, D. (2008). Prolegomena to sediment and flow connectivity in the landscape: A GIS and
field numerical assessment. Catena, 75(3), 268-277.
Borselli, L., Torri, D., Poesen, J., & Iaquinta, P. (2012). A robust algorithm for estimating soil erodibility in different
climates. Catena, 97, 85-94.
Caro, C., Marques, J. C., Cunha, P. P., & Teixeira, Z. (2020). Ecosystem services as a resilience descriptor in habitat
risk assessment using the InVEST model. Ecological Indicators, 115, 106426.
Cavalli, M., Trevisani, S., Comiti, F., & Marchi, L. (2013). Geomorphometric assessment of spatial sediment
connectivity in small Alpine catchments. Geomorphology, 188, 31-41.
Cong, W., Sun, X., Guo, H., & Shan, R. (2020). Comparison of the SWAT and InVEST models to determine
hydrological ecosystem service spatial patterns, priorities and trade-offs in a complex basin. Ecological
Indicators, 112, 106089.
de Magalhães, S. F. C., de Moura Barboza, C. A., Maia, M. B., & Molisani, M. M. (2022). Influence of land cover,
catchment morphometry and rainfall on water quality and material transport of headwaters and low-order
streams of a tropical mountainous watershed. Catena, 213, 106137.
Demissie, T. A., Saathoff, F., Seleshi, Y., & Gebissa, A. (2013). Evaluating the effectiveness of best management
practices in Gilgel Gibe Basin Watershed-Ethiopia. Journal of Civil Engineering and Architecture, 7(10),
1240.
Durigon, V., Carvalho, D., Antunes, M., Oliveira, P., & Fernandes, M. (2014). NDVI time series for monitoring
RUSLE cover management factor in a tropical watershed. International journal of remote sensing, 35(2), 441-
453.
Ethiopia, U. (2018). Ethiopia’s progress towards eradicating poverty. Implementation of the Third United Nations
Decade for the Eradication of Poverty (2018–2027) Discussion Paper.
FAO, I. (2015). Status of the World’s Soil Resources (SWSR)–technical summary. Food and Agriculture Organization
of the United Nations and Intergovernmental Technical Panel on Soils, Rome, Italy.
Fenta, A. A., Tsunekawa, A., Haregeweyn, N., Tsubo, M., Yasuda, H., Kawai, T., Ebabu, K., Berihun, M. L., Belay, A.
S., & Sultan, D. (2021). Agroecology-based soil erosion assessment for better conservation planning in
Ethiopian river basins. Environmental research, 195, 110786.
Foley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., Chapin, F. S., Coe, M. T., Daily, G. C.,
& Gibbs, H. K. (2005). Global consequences of land use. science, 309(5734), 570-574.
Ganasri, B., & Ramesh, H. (2016). Assessment of soil erosion by RUSLE model using remote sensing and GIS-A case
study of Nethravathi Basin. Geoscience Frontiers, 7(6), 953-961.
Gashaw, T., Bantider, A., Zeleke, G., Alamirew, T., Jemberu, W., Worqlul, A. W., Dile, Y. T., Bewket, W., Meshesha,
D. T., & Adem, A. A. (2021). Evaluating InVEST model for estimating soil loss and sediment export in data
scarce regions of the Abbay (Upper Blue Nile) Basin: Implications for land managers. Environmental
Challenges, 5, 100381.
Gashaw, T., Tulu, T., Argaw, M., & Worqlul, A. W. (2018). Modeling the hydrological impacts of land use/land cover
changes in the Andassa watershed, Blue Nile Basin, Ethiopia. Science of the Total Environment, 619, 1394-
 1408.
Gashaw, T., Worqlul, A. W., Dile, Y. T., Addisu, S., Bantider, A., & Zeleke, G. (2020). Evaluating potential impacts of
land management practices on soil erosion in the Gilgel Abay watershed, upper Blue Nile basin. Heliyon, 6(8).
Gelagay, H. S., & Minale, A. S. (2016). Soil loss estimation using GIS and Remote sensing techniques: A case of Koga
watershed, Northwestern Ethiopia. International Soil and Water Conservation Research, 4(2), 126-136.
Gupta, G. S. (2019). Land degradation and challenges of food security. Rev. Eur. Stud., 11, 63.
Haile, S. (2004). Population, development, and environment in Ethiopia. Environmental change and security project
report, 10, 43-51.
Hamel, P., Bosch, M., Tardieu, L., Lemonsu, A., de Munck, C., Nootenboom, C., Viguié, V., Lonsdorf, E., Douglass,
J. A., & Sharp, R. P. (2024). Calibrating and validating the Integrated Valuation of Ecosystem Services and
Tradeoffs (InVEST) urban cooling model: case studies in France and the United States. Geoscientific Model
Development, 17(12), 4755-4771.
Hamel, P., Bremer, L. L., Ponette-González, A. G., Acosta, E., Fisher, J. R., Steele, B., Cavassani, A. T., Klemz, C.,
Blainski, E., & Brauman, K. A. (2020). The value of hydrologic information for watershed management
programs: The case of Camboriú, Brazil. Science of the Total Environment, 705, 135871.
Hamel, P., Guerry, A., Polasky, S., Han, B., Douglass, J., Hamann, M., Janke, B., Kuiper, J., Levrel, H., & Liu, H.
(2021). Mapping the benefits of nature in cities with the InVEST software. Npj Urban Sustainability, 1(1), 25.
Haregeweyn, N., Tsunekawa, A., Nyssen, J., Poesen, J., Tsubo, M., Tsegaye Meshesha, D., Schütt, B., Adgo, E., &
Tegegne, F. (2015). Soil erosion and conservation in Ethiopia: a review. Progress in Physical Geography,
39(6), 750-774.
He, N., Guo, W., Wang, H., Yu, L., Cheng, S., Huang, L., Jiao, X., Chen, W., & Zhou, H. (2023). Temporal and spatial
variations in landscape habitat quality under multiple land-use/land-cover scenarios based on the PLUS-
InVEST model in the Yangtze River Basin, China. Land, 12(7), 1338.
Hossain, A., Krupnik, T. J., Timsina, J., Mahboob, M. G., Chaki, A. K., Farooq, M., Bhatt, R., Fahad, S., &
Hasanuzzaman, M. (2020). Agricultural land degradation: processes and problems undermining future food
security. In Environment, climate, plant and vegetation growth (pp. 17-61). Springer.
Hurni, H. (1983). Soil erosion and soil formation in agricultural ecosystems: Ethiopia and Northern Thailand.
Mountain research and development, 131-142.
Hurni, H. (1985). Erosion-productivity-conservation systems in Ethiopia.
Hurni, H., Abate, S., Bantider, A., Debele, B., Ludi, E., Portner, B., Yitaferu, B., & Zeleke, G. (2010a). 12 Land
Degradation and Sustainable Land Management in the Highlands of Ethiopia. Global change and sustainable
development, 187.
Hurni, H., Abate, S., Bantider, A., Debele, B., Ludi, E., Portner, B., Yitaferu, B., & Zeleke, G. (2010b). Land
degradation and sustainable land management in the highlands of Ethiopia.
Hurni, H., Giger, M., Liniger, H., Studer, R. M., Messerli, P., Portner, B., Schwilch, G., Wolfgramm, B., & Breu, T.
(2015). Soils, agriculture and food security: the interplay between ecosystem functioning and human well-
being. Current Opinion in Environmental Sustainability, 15, 25-34.
Hurni, H., Tato, K., & Zeleke, G. (2005). The implications of changes in population, land use, and land management
for surface runoff in the upper Nile basin area of Ethiopia. Mountain research and development, 25(2), 147-
154.
Initiative, E. (2015). The value of land: Prosperous lands and positive rewards through sustainable land management.
Available from www. eld-initiative. org.
Jiru, T. L. (2021). Rate of Land use land cover Changes of Setema District Jimma Zone South west, Ethiopia.
JIRU, T. L., & MOLLA, M. B. (2022). FOREST COVER CHANGES, AND COMMUNITY PERSPECTIVE ON ITS
DRIVING FORCES IN SETEMA DISTRICT JIMMA ZONE, SOUTHWEST ETHIOPIA. Journal of Global
Ecology and Environment, 15(4), 1-16.
Kaiser, M. S. (2021). Land degradation: causes, impacts, and interlinks with the sustainable development goals. In
Responsible consumption and production (pp. 1-13). Springer.
Kefyalew, E. (2011). Fertilizer Consumption and Agricultural Productivity in Ethiopia.
Kerbe, T. A., Mosissa Ejeta, T., & Gemeda, D. O. (2023). Soil erosion risk assessment using geospatial technologies
and RUSLE model in Bore Guda Watershed, Southwestern Ethiopia. Geology, Ecology, and Landscapes, 1-
13.
Kidane, D., & Alemu, B. (2015). The effect of upstream land use practices on soil erosion and sedimentation in the
Upper Blue Nile Basin, Ethiopia. Research Journal of Agriculture and Environmental Management, 4(2), 55-
68.
Lal, R. (2004). Soil carbon sequestration to mitigate climate change. Geoderma, 123(1-2), 1-22.
Lal, R. (2009). Soil degradation as a reason for inadequate human nutrition. Food Security, 1, 45-57.
Lillesand, T., Kiefer, R. W., & Chipman, J. (2015). Remote sensing and image interpretation. John Wiley & Sons.
McVittie, A., & Hussain, S. J. E. S. E. T. a. t. S. s. R. C. (2013). The economics of ecosystems and biodiversity-
valuation database manual. 5, 133-142.
Moriasi, D. N., Arnold, J. G., Van Liew, M. W., Bingner, R. L., Harmel, R. D., & Veith, T. L. (2007a). Model
evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the
ASABE, 50(3), 885-900.
Moriasi, D. N., Arnold, J. G., Van Liew, M. W., Bingner, R. L., Harmel, R. D., & Veith, T. L. J. T. o. t. A. (2007b).
Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. 50(3), 885-
900.
Munoth, P., & Goyal, R. (2019). Effects of DEM source, spatial resolution and drainage area threshold values on
hydrological modeling. Water Resources Management, 33, 3303-3319.
Negese, A. (2021). Impacts of land use and land cover change on soil erosion and hydrological responses in Ethiopia.
Applied and Environmental Soil Science, 2021, 1-10.
Nkonya, E. M. (2019). Climate Change and Land: an IPCC special report on climate change, desertification, land
degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems.
Nut, N., Mihara, M., Jeong, J., Ngo, B., Sigua, G., Prasad, P. V., & Reyes, M. R. (2021). Land use and land cover
changes and its impact on soil erosion in Stung Sangkae catchment of Cambodia. Sustainability, 13(16), 9276.
Nyssen, J., Poesen, J., Moeyersons, J., Deckers, J., Haile, M., & Lang, A. (2004). Human impact on the environment in
the Ethiopian and Eritrean highlands—a state of the art. Earth-science reviews, 64(3-4), 273-320.
Pachauri, R. K., Allen, M. R., Barros, V. R., Broome, J., Cramer, W., Christ, R., Church, J. A., Clarke, L., Dahe, Q., &
Dasgupta, P. (2014). Climate change 2014: synthesis report. Contribution of Working Groups I, II and III to
the fifth assessment report of the Intergovernmental Panel on Climate Change. Ipcc.
Panagos, P., Borrelli, P., Meusburger, K., Alewell, C., Lugato, E., & Montanarella, L. (2015). Estimating the soil
 erosion cover-management factor at the European scale. Land use policy, 48, 38-50.
Panagos, P., Borrelli, P., Poesen, J., Meusburger, K., Ballabio, C., Lugato, E., Montanarella, L., & Alewell, C. (2016).
Reply to the comment on “The new assessment of soil loss by water erosion in Europe” by Fiener &
Auerswald. Environmental science & policy, 57, 143-150.
Pandit, R., Scholes, R., Montanarella, L., Brainich, A., Barger, N., ten Brink, B., Cantele, M., Erasmus, B., Fisher, J., &
Gardner, T. (2018). Summary for policymakers of the assessment report on land degradation and restoration
of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services.
Pimentel, D. (2006). Soil erosion: a food and environmental threat. Environment, Development and Sustainability, 8,
119-137.
Pimentel, D., & Burgess, M. (2013). Soil erosion threatens food production. Agriculture, 3(3), 443-463.
Posner, S., Verutes, G., Koh, I., Denu, D., & Ricketts, T. (2016). Global use of ecosystem service models. Ecosystem
services, 17, 131-141.
Renard, K. G. (1997). Predicting soil erosion by water: a guide to conservation planning with the Revised Universal
Soil Loss Equation (RUSLE). US Department of Agriculture, Agricultural Research Service.
Sadeghi, S., Mizuyama, T., Miyata, S., Gomi, T., Kosugi, K., Fukushima, T., Mizugaki, S., & Onda, Y. J. G. (2008).
Development, evaluation and interpretation of sediment rating curves for a Japanese small mountainous
reforested watershed. 144(1-2), 198-211.
Selassie, Y. G., & Belay, Y. J. J. o. A. S. (2013). Costs of nutrient losses in priceless soils eroded from the highlands of
Northwestern Ethiopia. 5(7), 227.
Sharp, R., Douglass, J., Wolny, S., Arkema, K., Bernhardt, J., Bierbower, W., Chaumont, N., Denu, D., Fisher, D., &
Glowinski, K. (2020). InVEST 3.8. 7. user’s guide. The Natural Capital Project.
Shukla, P. R., Skeg, J., Buendia, E. C., Masson-Delmotte, V., Pörtner, H.-O., Roberts, D., Zhai, P., Slade, R., Connors,
S., & Van Diemen, S. (2019). Climate Change and Land: an IPCC special report on climate change,
desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in
terrestrial ecosystems.
Tamene, L., Adimassu, Z., Aynekulu, E., & Yaekob, T. (2017). Estimating landscape susceptibility to soil erosion
using a GIS-based approach in Northern Ethiopia. International Soil and Water Conservation Research, 5(3),
221-230.
Tamene, L., Adimassu, Z., Ellison, J., Yaekob, T., Woldearegay, K., Mekonnen, K., Thorne, P., & Le, Q. B. (2017).
Mapping soil erosion hotspots and assessing the potential impacts of land management practices in the
highlands of Ethiopia. Geomorphology, 292, 153-163.
Tamene, L., & Vlek, P. L. (2008). Soil erosion studies in northern Ethiopia. In Land use and soil resources (pp. 73-
100). Springer.
Team, R. C. (2020). RA language and environment for statistical computing, R Foundation for Statistical. Computing.
Telles, T. S., Dechen, S. C. F., Souza, L. G. A. d., & Guimarães, M. d. F. J. S. A. (2013). Valuation and assessment of
soil erosion costs. 70, 209-216.
Teshome, A., de Graaff, J., Ritsema, C., & Kassie, M. (2016). Farmers' perceptions about the influence of land quality,
land fragmentation and tenure systems on sustainable land management in the north western Ethiopian
highlands. Land degradation & development, 27(4), 884-898.
Tsegaye, L., Degu, M., Mekonnen, M., & Gashaw, T. (2024). Soil erosion and sediment export analysis for watershed
 management options in Fakisi watershed of the Abbay basin of Ethiopia. Environmental Challenges, 15,
100948.
Tully, K., Sullivan, C., Weil, R., & Sanchez, P. (2015). The state of soil degradation in Sub-Saharan Africa: Baselines,
trajectories, and solutions. Sustainability, 7(6), 6523-6552.
UNICEF. (2023). The state of food security and nutrition in the world 2023.
Viera, A. J., & Garrett, J. M. (2005). Understanding interobserver agreement: the kappa statistic. Fam med, 37(5), 360-
363.
Vigiak, O., Borselli, L., Newham, L., McInnes, J., & Roberts, A. (2012). Comparison of conceptual landscape metrics
to define hillslope-scale sediment delivery ratio. Geomorphology, 138(1), 74-88.
Wassie, S. B. (2020). Natural resource degradation tendencies in Ethiopia: a review. Environmental systems research,
9(1), 1-29.
Wischmeier, W. H., & Smith, D. D. (1978). Predicting rainfall erosion losses: a guide to conservation planning.
Department of Agriculture, Science and Education Administration.
Wu, C., Qiu, D., Gao, P., Mu, X., & Zhao, G. (2022). Application of the InVEST model for assessing water yield and
its response to precipitation and land use in the Weihe River Basin, China. Journal of Arid Land, 14(4), 426-
440.
Xiao, P., Zhou, Y., Li, M., & Xu, J. (2023). Spatiotemporal patterns of habitat quality and its topographic gradient
effects of Hubei Province based on the InVEST model. Environment, Development and Sustainability, 25(7),
6419-6448.
Yengoh, G. T. (2020). Impact of Land Degradation on Rural Livelihoods in Sub-Saharan Africa. In Life on Land (pp.
537-551). Springer.
Yohannes, H., Argaw, M., & Seifu, W. (2024). Impact of land use/land cover change on surface water hydrology in
Akaki river catchment, Awash basin, Ethiopia. Physics and Chemistry of the Earth, Parts A/B/C, 135, 103690.
Zerihun, M., Mohammedyasin, M. S., Sewnet, D., Adem, A. A., & Lakew, M. (2018). Assessment of soil erosion using
RUSLE, GIS and remote sensing in NW Ethiopia. Geoderma regional, 12, 83-90.
Zhang, J., Wang, N., Wang, Y., Wang, L., Hu, A., Zhang, D., Su, X., & Chen, J. (2021). Responses of soil erosion to
land‐use changes in the largest tableland of the Loess Plateau. Land degradation & development, 32(13),
3598-3613.
CRediT authorship contribution statement
Deresa Abetu Gadisa: Conceptualization, methodology, writing-review and editing,
original draft, Data collection and analysis, Software. Abreham Berta Aneseyee:
Supervision, Validation, Software, Review and editing. Hirpa Abduro Ogo: Review and
editing,Validation and Methodology, Habtamu Eticha Olana: Review and editing,
Validation, Methodology, Jiacheng Huang: Validation, Software, Review and editing Sisi
YU: Supervision, Software, Methodology, Comments, Review & editing. Junyao Sun:
Methodology, Comments, Review & editing Xue Yan: Supervision, Conceptualization,
Visualization and validation, Investigation, Review & editing and Funding acquisition.

Acknowledgements

This study was supported by the Sino-Africa Joint Research Center (SAJOREC). We thank
Jimma University College of Agriculture and Veterinary Medicine (JUCAVM) soil
laboratory for conducting the soil analysis. Finally, the authors appreciate the helpful
comments from the anonymous reviewers and the editor for their comments and
suggestions, which have greatly improved the scientific quality of this work.