Morphometric analysis to characterize the soil erosion susceptibility in the western part of lower Gangetic River basin, India

Abstract

The present study is an attempt to identify the morphometric parameters for determining the soil erosion susceptibility (SES) in upper plateau fringe catchment and lower undulating plain catchment of Kangsabati basin using of multi-criteria decision model as compound factor (CF) and hydrological model as revised universal soil loss equation (RUSLE) under GIS platform. In this research, twenty morphometric parameters of three aspects, namely, linear, areal, and relief were taken from twenty-seven sub-basins to compute the morphometric priority rank using CF model. In contrary, soil erosion factors like rainfall, soil character, slope, and land cover management were taken to measure the potential annual soil erosion using RUSLE. The result showed that fourteen sub-basins in upper catchment (UC) with low compound value represent high morphometric priority rank whereas thirteen sub-basins in lower catchment (LC) with high compound value indicate lower morphometric priority rank. RUSLE estimated that higher rate of mean soil erosion (225 t ha−1 year−1) occurred in UC, whereas low rate of mean soil erosion (74 t ha−1 year−1) occurred in LC. Stepwise least regression of Akaike information criteria (AIC) showed that mean bifurcation ratio (p > 0.013), ruggedness index (p < 0.0001), and form factor (p > 0.034) are the best positive coefficient of erosion susceptibility (ES), but gradient ratio (p > 0.003) and elongation ratio (p > 0.024) have perfect inverse coefficient of high ES in LC. Gradient ratio (p > 0.044) is the only best inverse coefficient parameter of ES in LC due to presence of several conservative practices. Moreover, proper management strategy, effective morphometric, and erosion factors play vital role to determine the ES.

Full text

ORIGINAL PAPER
Morphometric analysis to characterize the soil erosion susceptibility
in the western part of lower Gangetic River basin, India
Raj Kumar Bhattacharya
1
&Nilanjana Das Chatterjee
1
&Prasenjit Acharya
1
&Kousik Das
1
Received: 10 October 2018 / Accepted: 22 February 2021
#Saudi Society for Geosciences 2021
Abstract
The present study is an attempt to identify the morphometric parameters for determining the soil erosion susceptibility (SES) in
upper plateau fringe catchment and lower undulating plain catchment of Kangsabati basin using of multi-criteria decision model
as compound factor (CF) and hydrological model as revised universal soil loss equation (RUSLE) under GIS platform. In this
research, twenty morphometric parameters of three aspects, namely, linear, areal, and relief were taken from twenty-seven sub-
basins to compute the morphometric priority rank using CF model. In contrary, soil erosion factors like rainfall, soil character,
slope, and land cover management were taken to measure the potential annual soil erosion using RUSLE. The result showed that
fourteen sub-basins in upper catchment (UC) with low compound value represent high morphometric priority rank whereas
thirteen sub-basins in lower catchment (LC) with high compound value indicate lower morphometric priority rank. RUSLE
estimated that higher rate of mean soil erosion (225 t ha
−1
year
−1
) occurred in UC, whereas low rate of mean soilerosion (74 t ha
−1
year
−1
) occurred in LC. Stepwise least regression of Akaike information criteria (AIC) showed that mean bifurcation ratio (p>
0.013), ruggedness index (p< 0.0001), and form factor (p> 0.034) are the best positive coefficient of erosion susceptibility (ES),
but gradient ratio (p> 0.003) and elongation ratio (p> 0.024) have perfect inverse coefficient of high ES in LC. Gradient ratio (p
> 0.044) is the only best inverse coefficient parameter of ES in LC due to presence of several conservative practices. Moreover,
proper management strategy, effective morphometric, and erosion factors play vital role to determine the ES.
Keywords Soilerosion susceptibility (SES) .Hydro-geomorphic status .Morphometric priority .Upper catchment (UC) .Lower
catchment (LC)
Introduction
Soil erosion is a process of detachment or breakdown of par-
ticles that are transported or deposited in new area by water
action in response to several natural and anthropogenic
activities (Masselink et al. 2017; Ameri et al. 2017). Water
action in soil erosion causes several consequences like top soil
fertility deterioration (Chakraborty et al. 2020), depletion of
nutrient particles (Toubal et al. 2018), reduction in vegetation
growth (Biswas et al. 2015; Chakraborty et al. 2020), huge
reservoir sedimentation, and delta formation in the estuarine
sites (Biswas et al. 2015; Ameri et al. 2017). Moreover, soil
erosion capacity is controlled by various geo-environmental
features like land slope (Fang et al. 2019; Nehal and
Guettouche 2020), organic matter, and permeability rate
(Wischmeier and Smith 1978; Uddin et al. 2016), which plays
crucial role to create vulnerability in water erosion on terrain
surface than undulating plain surface (Karamage et al. 2016).
Several factors like geomorphic setup, lithology structure, soil
character, slope, and land use and land cover (LULC) lead to
accelerate the soil erosion that is further triggered by human
activities in different climatic region (Chuahan et al. 2016;
Fallah et al. 2016; Bhattacharya et al. 2019). In respect to
the significant role of controlling factors on erosion
Responsible Editor: Amjad Kallel
*Raj Kumar Bhattacharya
rajgeovu10@gmail.com
Nilanjana Das Chatterjee
nilanjana_vu@mail.vidyasagar.ac.in
Prasenjit Acharya
prasenjit.ac@mail.vidyasagar.ac.in
Kousik Das
kousikvugeo@gmail.com
1
Department of Geography, VidyasagarUniversity, WestBengal
Midnapore, India
https://doi.org/10.1007/s12517-021-06819-8
/ Published online: 13 March 2021
Arabian Journal of Geosciences (2021) 14: 501

susceptibility (ES), geo-environmental setting of those factors
is governed by morphometric properties including landform
formation, physical properties of soil, and consequences of
soil erosion (Patel et al. 2012; Rodrigo-Comino et al. 2016;
Bhattacharya et al. 2019; Malik et al. 2019). Morphometric
parameters in a drainage basin are useful criteria in under-
standing the numerical analysis of drainage network comput-
ing the three basic aspects such as linear, areal, and relief,
respectively (Patel et al. 2013; Gajbhiye et al. 2014). Sub-
basin prioritization in order to soil erosion susceptibility
(SES) and implementation of soil and water conservation
practices are entirely dependent on numerical analysis of
drainage network using GIS and RS techniques (Deepika
et al. 2013; Fallah et al. 2016; Ameri et al. 2017). Recently,
multi-criteria decision-making (MCDM) techniques under
GIS platform are employed to estimate the sub-basin priority
using decision accuracy. This accuracy is not considered only
one criterion but also taking others criterion (Mulliner et al.
2016; Ameri et al. 2017; Bhattacharya et al. 2019). In this
context, compound factor (CF) is one of the effective
MCDM techniques to use the sub-basin priority for assessing
the SES based on basin morphometric parameters (Patel et al.
2013; Altaf et al. 2014).
In contrary, many empirical models such as universal soil
loss equation model (USLE) (Wischmeier and Smith 1978;
Pham et al. 2018) and soil and water analysis tools (SWAT)
are used to estimate the soil erosion rate at sub-basin level
along with determining the significant role of erosion param-
eters using GIS techniques (Markhi et al. 2018; Sridhar et al.
2018). Revised version of USLE model and GIS techniques
used two important aspects i.e., to measure the potential mean
annual soil erosion distribution in one hand and to prepare a
sub-basin priority in order to integrated basin management
including soil and water conservation practices in other hand
(Fallah et al. 2016; Bhattacharya et al. 2020a). Several studies
have been successfully carried out using the revised universal
soil loss equation (RUSLE) model for the measuring of po-
tential annual soil erosion in different climatic region of India
(Kumar et al. 2014; Gansari and Ramesh 2016; Nasir and
Selvakumar 2018). Many researchers have been already ap-
plied the RUSLE model integrated with GIS technique to
estimate the soil loss for implementing of basin management
practices on Kangsabati river basin (Mahala 2018;
Bhattacharya et al. 2020a,b). This basin is a part of eastern
Chota Nagpur plateau under tropical climate, two topograph-
ical classes i.e., plateau fringe and undulating plain surface are
causes the huge acceleration of soil erosion throughout the
basin (Bhattacharya et al. 2020a). In spite of significant basin
morphometry, several soil erosion factors such as elevated
terrain surface, thick laterite upslope, single crop cultivation
(plateau fringe), high settlement density, huge deforestation,
double crop cultivation (undulating plain surface), and season-
al monsoon rainfall (July–September) are considered
responsible factors to create the most erosion-prone tropical
basin in the western part of lower Gangetic basin of India (Shit
et al. 2015; Bhattacharya et al. 2020b). In this context, RUSLE
and MCDM models have identified two research gaps at sub-
basin level i.e., all morphometric parameters are not equal
importance in order to ES, and all effective parameters have
no same significant role in every sub-basin as they have their
own characteristics. In this study, we have used one MCDM
model as CF to assign the sub-basin priority rank from mor-
phometric parameters in response to ES. On the other hand,
RUSLE model is used toquantify the erosion rate at sub-basin
level for assigning the actual soil erosion rank. In respect to
above views, this research is conducted with the three objec-
tives: (1) to select the best possible morphometric parameters
following the estimated erosion rate given by RUSLE model,
(2) to determine the different role of effective parameters at
sub-basin level, and (3) to identify the suitable sites for
implementing the soil and water erosion conservative prac-
tices and management.
Materials and methods
Geo-environmental description of the study area
Kangsabati River feeds its water to the river Ganges from the
right side in the lower Gangetic basin. This basin comprises
mainly three districts, namely, the entire Purulia district,
southwestern part of Bankura district, parts of the western,
and central and eastern sectors of Medinipur district in West
Bengal, as well as a small portion of southeastern Jharkhand,
India. Kangsabati basin extended from latitudinal 21° 45′Nto
23° 30′N and longitudinal 85° 45′Eto88°15′Ecoversan
area of about 9658 km
2
(Fig. 1a). Several tributaries, namely,
Kumari, Bhairabbanki, and Taraphini including their sub-
tributaries make dendritic to sub-dendritic drainage pattern
in Kangsabati basin. Two different confluence points of
Kansai-Kumari confluence near Mukutmanipur dam and
Bairabbanki-Taraphini confluence at Sijua near Lalgarh are
observed along the Kangsabati main channel, while one chan-
nel bifurcated point is observed at Kapastikri where one seg-
ment drains towards the Keleghai River and the other segment
drains towards Rupnarayan River (Fig. 1b). Tropical monsoon
climatic characteristic is mainly dominated in this basin where
most of the rainfall occurred in monsoon period ranges from
1077 to 1804 mm/year. Sixteen geological formations in this
basin comprise of fifteen structures, and in particular, the most
of the predominant structures are mica schist occasionally
garnetiferous and in situ caliche groups including sand, silt,
and clay formations (Mukhopadhaya 1992). The order form of
geomorphic features, buried pediments with lateritic capping
and deeply buried pediments, are covered in plateau fringe
site, while rocky outcrop and pediment with laterite capping,
501 Page 2 of 22
Arab J Geosci (2021) 14: 501

Fig. 1 Location map of Kangsabati Basin: sub-basin demarcation (a). Stream order (b)
Page 3 of 22 501Arab J Geosci (2021) 14: 501

valley fill deposits, and floodplain deposits are concentratedin
undulating plain surface (Bhattacharya et al. 2020c). Most of
the dominant soil groups are fine loamy type Haplustafs, fine
Aeric ochraquepts, and fine loamy ultipaleustalfs out of six-
teen groups. Several geo-environmental setups in Kangsabati
basin help to delineate the two different physiographic region
i.e., plateau fringe (upper catchment, UC) and undulating
plain surface (lower catchment, LC) (Mahala 2018;
Bhattacharya et al. 2020a,b).
Morphometric analysis
In present study, different data sets like SRTM DEM,
LANDSAT image, and topographical map were considered
for the delineation of sub-basins in Kangsabati basin.
Morphometric, stratigraphic, and landform characteristics in
the entire basin have been systematically analyzed using the
geospatial data sets under GIS platform (Table 1). Twenty
morphometric parameters of three aspects i.e., linear, areal,
and relief are considered for the analysis of ES (Strahler
1964). Linear aspect is determined by basin length (L), stream
order (u), stream length (Lu), stream length ratio (Rl), and
bifurcation ratio (Rb) whereas areal aspect is involved with
form factor (Ff), elongation ratio (Re), circularity ratio (Rc),
shape factor (Bs), drainage density (D), stream frequency (Fs),
drainage texture (T), compactness coefficient (Cc), length of
overland flow (Lo), and constant of channel maintenance (C).
On the other hand, relief aspect included basin relief (R)and
relief ratio (Rr), ruggedness number (Rn), and gradient ratio
(Gr). All parameters are computed using the following empir-
ical Eqs. (1–20) in Table 2. Theoretical explanations of twenty
morphometric parameters are well analyzed in Supplementary
Section.
Sub-basin prioritization
Compound factor (CF) is one of the effective MCDM tech-
niques, which is used for the assignment of fifteen morpho-
metric parameters from twenty-seven sub-basins, to prepare
priority rank orders at sub-basin level through the incorporat-
ed with ES. Therefore, CF is calculated after the assigned of
average rank value from selected morphometric parameters in
the following Eq. (21)(Altafetal.2014; Bhattacharya et al.
2019):
CF ¼1
n∑
pn
i¼1
Rð21Þ
where Rmeans parameters rank value, and pn indicates wa-
tershed rank from each parameter.
In terms of ES, linear aspects have a positive relationship
with the erosive capacity, but areal aspects have inverse rela-
tionship with the erosive capacity of the basin (Patel et al.
2012,2013; Bhattacharya et al. 2019). Therefore, rank 1 was
taken from the highest value of linear aspect, and the next rank
was rated by the second highest value of this aspect. In aerial
aspect, the lower compound value was assigned as rank 1, and
the second lowest as rank 2 and so on. While the largest
compound factor was taking as the lowest priority rank of
sub-watersheds and so forth following Ratnam et al. (2005),
Pateletal.(2012), and Altaf et al. (2014). Rank assignment in
relief aspect has same manner as like linear aspect (Pradhan
et al. 2018).
Delineation of hypsometric curves
Analysis of hypsometric curve helps to delineate the drainage
basin area at different elevation (Strahler 1952). In Kangsabati
basin, the hypsometric curve was obtained through the plot-
ting of relative area (a/A) along the horizontal cross-sectional
area across the drainage basin and relative elevation (h/H)
plotted along the vertical axis in the following Eq. (22):
Hc ¼a
A

=h
H
 ð22Þ
where a/Adenotes relative area, and h/Hdenotes relative
elevation.
Table 1 Details of geospatial data
Data type Source Resolution/
scale/type
Data uses
SRTM DEM Earth explorer
(2016)
https://www.usgs.
gov
30-m spatial
resolu-
tion
Extraction of basin
morphometric
parameters
LANDSAT 8
OLI/TM
Earth explorer
(2016)
http://landsat.usgs.
gov
30-m spatial
resolu-
tion
Extraction of C
(crop
management)
factor in RUSLE
model
Topographical
map
Survey of India
(73J/15, 9, I/2, 3,
4, 5, 7, 8, 9, 10)
http://www.
surveyofindia.
gov.in
1:50,000 Extraction drainage
network and LS
(slope length and
steepness) factor
in RUSLE
model
Soil sheet National Bureau of
Soil Survey and
Land Use
Planning (sheet
nos. 2, 3, 4)
http://www.
nbsslup.in
30-m spatial
resolu-
tion
Extraction of S (soil
erodibility)
factor in RUSLE
model
Rainfall Indian Metrological
Data
(1980–2016)
http://mausam.imd.
gov.in
Cumulative
daily data
type
Extraction of R
(rainfall
erosivity) factor
in RUSLE
model
501 Page 4 of 22
Arab J Geosci (2021) 14: 501

Estimation of HI
Hypsometric integral (H
si
) has an inverse relation with chan-
nel gradient, slope steepness, drainage density, and basin re-
lief, while it has a positive relation with rate of soil erosion in
homogeneous rock strata and more tectonic movements
(Meshram and Sharma 2017; Bhattacharya et al. 2019). H
si
was computed after the preparation of hypsometric curve fol-
lowing relief range and mean elevation of the basin (Strahler
1952).
HI ¼Emean−Emin
Emax−Emin ð23Þ
where E
mean
denotes the average elevation of a delineated
Table 2 Aspect wise mathematical formulae used based on several morphometric parameters
Aspects Formula Sources
Linear aspect
Basin L=length(L)L=1.312A
0.568
L= basin length (km),
A= area of the basin (km
2
)
Nookaratnam et al.
(2005)
Stream order (u) Hierarchical rank Strahler (1964)
Mean stream length
(Lsm)
Lsm = Lu/Nu
Lsm = mean stream length, Lu = total stream length of order u, Nu = total no. of stream segments of
order u
Strahler (1964)
Number of streams (Nu) Nu
Nu = number of streams in each order
Strahler (1964)
Stream length ratio (Rl) Rl = Lu/Lu−1
Rl = stream length ratio, Lu = total stream length of order u,Lu−1 = the total stream length of its next
lower order
Horton (1945)
Bifurcation ratio (Rb) Rb = Nu/Nu+1
Rb = bifurcation ratio,
Nu = total no. of stream segments of order u, Nu+1 = number of segments of the next higher order
Schumm (1956)
Mean bifurcation ratio
(Rbm)
Rbm = average of bifurcation ratios of all orders Strahler (1957)
Aerial aspect
Form factor (Ff) Ff = A/L
2
Ff = form factor, A= area of the Basin (km
2
), L= basin length (km)
Horton (1945,
1932)
Elongation ratio (Re) Re = 1.128 √A/L
Re = elongation ratio, A= area of the basin (km
2
), L= basin length (km)
Schumm (1956)
Circularity ratio (Rc) Rc = 4 πA/P
2
Rc = circularity ratio, π=3.14,A= area of the basin (km
2
), P= perimeter (km)
Miller (1953);
Strahler (1964)
Shape factor (Bs) Bs = L
2
/A
Bs = shape factor, L= basin length (km), A= area of the basin (km
2
)
Horton (1932)
Compactness coefficient
(Cc)
Cc = 0.282 1 P/A
0.5
Cc = compactness coefficient, P= perimeter (km), A= area of the basin (km
2
)
Gravelius (1914)
Drainage density (D)D=Lu/A
D
d
= drainage density, Lu = total stream length of all orders, A= area of the basin (km
2
)
Horton (1945,1932)
Stream frequency (Fs) Fs = ∑Nu/A
Fs = stream frequency, ∑Nu = total no. of streams of all orders, A= area of the basin (km
2
)
Horton (1945,1932)
Drainage texture (T)T=D×Fs
T= drainage texture, D
d
= drainage density, Fs = stream frequency
Horton (1945)
Constant of channel
maintenance (C)
C=1/D
C= constant of channel maintenance, D=drainagedensity
Schumm (1956)
Length of overland flow
(Lo)
Lo = 1/2D
Lo = length of overland flow, D=drainagedensity
Horton (1945)
Relief aspect
Basin relief (R)R=H−h
R=basinrelief,H= maximum elevation in meter, h= minimum elevation in meter
Hadley and
Schumm (1961)
Relief ratio (R
r
)R
r
=R/L
Rr = relief ratio, R= basin relief, L= longest axis in kilometer
Schumm (1956)
Ruggedness number
(Rn)
Rn = (R×D)/K
Rn = ruggedness number, R= basin relief, D= drainage density, K= a conversion constant 1000 when
relative relief is expressed in meter and drainage density in kilometer/square kilometer.
Schumm (1956)
Gradient ratio (Gr)Gr=(a−b)/L
Gr = gradient ratio, a= elevation at source, b= elevation at mouth, L= longest axis in kilometer
Sreedevi et al.
(2005)
Page 5 of 22 501Arab J Geosci (2021) 14: 501

watershed obtained from its contours, E
max
and E
min
refer
maximum and minimum elevation values extracted from de-
lineated watershed. With the help of convex and concave
curve shape, hypsometric integral values are determined the
several landform evolution stages i.e., convex curve (HI >
0.60) indicates to youth or equilibrium stage, medium convex
curve (HI 0.60-0.35) indicates to mature or equilibrium stage,
and concave curve (HI < 0.35) indicates old or monadnock
stage, respectively (Strahler 1952).
RUSLE parameter estimation
Revised universal soil loss equation (RUSLE) is applied to
estimate potential rate of soil erosion using parameters like
precipitation (rainfall erosivity—Rfactor), soil characteristics
(soil erodibility—Kfactor), elevation (slope length and
steepness—LS factor), land use and land cover patterns
(cropping management—Cfactor), and conservation prac-
tices (support practice—Pfactor) (Sahaar 2013; Nasir and
Selvakumar 2018; Bhattacharya et al. 2020a,b). RUSLE
model is developed by Renard et al (1997) with the modifica-
tion of USLE parameters as given by Wischmeier and Smith
(1978) in the following Eq. (24):
A¼RKLSCPð24Þ
where Ameans assessed the spatial distribution of potential
mean annual soil erosion. Selection of Kis expressed in every
unit of soil textural code whereas Ris selected to detect the
requiring period of rainfall intensity in each unit. All units
were generally considered based on practice patterns. So, A
represents in tons per hectare per year (t ha
−1
/year) while other
units are represented by per acre per year (t acre
−1
/year).
Rmeans rainfall-runoff erodibility factor: both raindrop
splash and runoff flow are reflected the significant role of
rainfall factor in SES on denuding surface in the following
Eqs. (25)and(26) (Renerd and Freimund 1994):
R¼0:04830 p1:610 p<850 mm ð25Þ
R¼587:7−1:219 pþ0:004 p>850 mm ð26Þ
where Rdenotes annual rainfall erodibility represent in MJ
mm ha
−1
year
−1
,Pdenotes annual precipitation (mm). In this
study, isoerodent map prepared from annual rainfall data of
sixteen rain gauge stations (1980–2016) on a raster surface,
which was set up in a cell size grid of 30-m spatial resolution
and adjusted with another thematic map.
Kmeans soil erodibility factor: soil erosion index unit is
measured as a standardized plot of erosion rate on a specific
soil (Bhattacharya et al. 2020b). Kfactor was computed based
on soil textures (% silt plus very fine sand, % sand, % organic
matter), soil structure, and permeability, and then, all values
were extracted from nomograph (Wischmeier and Smith
1978). Kfactor was computed through in the following Eq.
(27):
K¼2::1:10−412−0MðÞM1:14 þ3:25 S−2ðÞþ2:5P−3ðÞ
100

ð27Þ
where Mdenotes product of primary particle size fractions (%
of modified silt or the 0.002–0.1-mm-sized fraction), Kde-
notes tons acre per erosion index (t ha h ha
−1
MJ
−1
mm
−1
)in
SI units, Sdenotes structure soil class, and pdenotes perme-
abilityrateineachsoilstructure.
Lmeans slope length factor: ratio of soil erosion is entirely
dependent that on field slope length where 72.6 ft is identical
condition for significant increases of flow accumulation
(Sahaar 2013; Gansari and Ramesh 2016). Lfactor was com-
puted from in the following Eq. (28):
L¼λ
22:13

mð28Þ
where Ldenotes slope length; λdenotes horizontal slope
length unit in feet; mdenotes exponent of slope variable
length (Eq. 28.1) i.e., 0.5 taking in slopes steeper than 5%;
0.4 for slopes ranges 3–4%; 0.3 for slopes ranges 1–3%; and
0.2 for slopes less than 1%
m¼β
1þβðÞ ð28:1Þ
βmeans moderate susceptibility caused by rill and inter-rill
erosion in the following Eq. (28.2) (McCool et al. 1987)
β¼
sin∅
0:0896
3:0sin∅ðÞ
0:080:56
lm ð28:2Þ
where θdenotes angle of slope in degrees unit
Smeans slope steepness factor: soil erosion initiated when
the field of slope gradient falls under 9% or otherwise identical
conditions (Bhattacharya et al. 2020b). Flow accumulation
and percentage of slope steepness both factors are considered
for soil erosion with the following of Wischmeier and Smith
methods (1978). Sfactor was estimated by the application of
algorithm-based slope steepness value in the following Eqs.
(29)and(30) (McCool et al. 1987):
S¼10:8sin θþ0:03;where σ<0:09 ð29Þ
S¼16:8sin θ−0:5;where σ≥0:09 ð30Þ
where σdenotes gradient of slope in percentage.
Cmeans cover management factor: ratio of soil erosion de-
pends on specific cover and management types given in area
where ploughed land is continuously converted into the fallow
land (Nasir and Selvakumar 2018). Cfactor value in settlement
501 Page 6 of 22
Arab J Geosci (2021) 14: 501

was taking of 0.2 following of strategic environmental assess-
ment (SEA) given by UN-FAO (2001) whereas barren land with
laterite was considered 0.5 following Bakker et al. (2008)and
degraded forest value of 0.05 taking from Bakker et al. (2008)
and Jordan et al. of agriculture under ecosystems and environ-
ment (2005). Cfactor of dense forest considered the value of 0.01
with follow up UN-FAO (2001), Bakker et al. (2008), and Jordan
et al. (2005), while value of water body of 0 taking from Cox, C
and Madramootoo, and C of Computers and Environment in
Agriculture (1998). Single crop and multiple cropping systems
were considered the values of 0.2 and 0.31 following of
Wischmeir, in Soil Science Society Proceeding (1960).
Pmeans support practice factor: several types of support
practice such as strip cropping, contouring, and terracing were
taken to determine the soil erosion rate along the straight-row
farming up and downslope (Renard 1997). Pfactor values are
ranges from 0 to 1 following Cfactor values in respective
LULC patterns, which reveals that 0 represents good conser-
vation practices and 1 represents poor conservation practices,
respectively (Pradhan et al. 2018).
Determination of effective morphometric parameters
on ES
Morphometric parameters and their responses on soil erosion
were determined through the use of multiple regression analysis
of Akaike information criteria (AIC), introduced by Akaike
(1973). AIC was applied to find out the effective morphometric
parameters on potential annual soil erosion. The least value of
AIC helps to select the best fit parameter from all morphometric
Watersheds
Extraction
Study Area
Extraction
Toposheet Satellite Image
DEM
Study Area
Extraction
Slope
Fill & Flow
Direction
Flow
Accumulation
C Factor
LU/LC
Rainfall
Data
Land
Management
Soil Map
R Factor P Factor
K Factor L Factor S Factor
Overlay
A=R×K×L×S×C×P
Factor
M Factor
L Factor S Factor
Soil Erosion Map
Rating watersheds based on parameter
Compound Factor
Prioritization of watersheds
SOI Toposheet
1:50,000
Georeferencing DEM
Extraction of Watersheds
RPC and
SRTM Data
GIS
Software
Digitization
Drainage Network
Drainage
Flow Direction
Flow
Morphometric Analysis
Linear Areal Relief
Akaike Information Criterion
(AIC) Statistical analysis
Data Genera tion
Selection of morphometric parameters and modeling soil loss
Fig. 2 Methodological flow chart of morphometric analysis and RUSLE model for prediction of SES
Page 7 of 22 501Arab J Geosci (2021) 14: 501

aspects. AIC was computed in the following Eq. (31):
AIC ¼Log RSS
Nþ2kð31Þ
Nis the number of observation, RSS is the sum of the square of
error or the residual sum of square, and kis the number of
parameters to fit plus 1 (k=β+ 1). In this study, a linear rela-
tionship has been made to show an association between actual
soil loss from RUSLE and estimated soil loss using morphomet-
ric parameters (Fig. 2).
Results
Characteristics of linear morphometric aspects
Basin area in twenty-seven sub-basins of Kangsabati basin
ranges from 693.69 km
2
(Lalgarh 4SB) to 23.86 km
2
(Mohanpur 6SB) with an average value of 109.54 km
2
(Table 3). Amid, maximum basin length is observed in
Khatra 1SB (54.57 km), and minimum basin length is ob-
served in Mohanpur 6SB (7.95 km) with an average of
16.80 km among all the sub-basins. Moreover, basin perime-
ter ranges from 247 km (Khatra 1SB) to 31 km (Raipur 6SB)
along with average value of 56 km.
According to Strahler stream ordering system (1964), drain-
age system in Kangsabati basin is segregated into six stream
ordersi.e.,1st,2nd,3rd,4th,5th,and6th.Themaximum1st
order streams are observed in Khatra 1SB (1070) and Lalgarh
4SB (1227) out of a total number of stream orders (3004),
which radiated towards NE, E, and SE directions (Table 3;
Fig. 3a). Subsequently, 2nd order streams (743) are generally
radiated towards the same directions as 1st order streams, 3rd
order (188) streams are predominantly oriented towards NE and
SE directions, and 4th order (46) and 5th order (13) both are
propagated towards NE, S, E, and SE directions, respectively
(Table 3; Fig. 3b–f). However, all stream orders have not
Table 3 Sub-basin wise computation of linear aspect
Sub-basin no Sub-basin
name
Basin area (km
2
) Basin perimeter Basin length (L) km Stream order wise stream number (Nu)
1st 2nd 3rd 4th 5th 6th Total
1 Khatra-1 708 247.8 54.5733 1070 249 66 17 5 2 1409
2 Khatra-2 113 105.6 19.32571 29 6 1 36
3 Khatra-3 96 49.18 17.59845 115 29 6 2 1 153
4 Raipur-1 90.8 50.58 16.98842 69 22 7 1 99
5 Raipur-2 75.46 58.07 15.29322 128 24 6 1 159
6 Raipur-3 32.8 31.17 9.542322 30 9 3 1 43
7 Raipur-4 63.9 41.9 13.92132 89 21 4 2 1 117
8 Raipur-5 29.2 32.65 8.933478 8 1 9
9 Lalgarh-1 48.8 39.84 11.9502 14 3 1 18
10 Lalgarh-2 71.3 48.16 14.81849 9 3 1 13
11 Lalgarh-3 71.8 45.32 14.87388 12 4 1 17
12 Lalgarh-4 693.69 156.62 53.916 1227 319 76 20 6 1 1649
13 Lalgarh-5 51.6 42.24 12.32432 5 2 1 8
14 Lalgarh-6 43 35.5925 11.11266 12 4 1 17
15 Dherua-1 69.5 47.83 14.59507 50 10 3 1 64
16 Dherua-2 31.83 31.62 9.366611 4 1 5
17 Dherua-3 32.5 32.57 9.485608 25 5 1 31
18 Dherua-4 60.1 35.56 13.44754 16 4 1 21
19 Dherua-5 41.9 37.42 10.96142 16 7 2 1 26
20 Mohanpur-1 34.4 32.42 9.798067 2 2
21 Mohanpur-2 35.6 36.26 9.981079 7 2 1 10
22 Mohanpur-3 77 55.8 15.47777 7 3 1 11
23 Mohanpur-4 58.36 37 13.21572 5 1 6
24 Mohanpur-5 67.2 45 14.32609 7 3 1 11
25 Mohanpur-6 23.8 23 7.951925 27 8 3 38
26 Mohanpur-7 73.2 49 15.04085 5 1 6
27 Midnapore-1 159.75 74.46 23.41501 16 2 1 19
501 Page 8 of 22
Arab J Geosci (2021) 14: 501

appeared in every sub-basin except Lalgarh 4SB and Khatra
1SB. On the other hand, the total stream number among the
twenty-seven sub-basin ranges from 1649 (Lalgarh 4SB) to 2
(Mohanpur 1SB). Therefore, stream direction indicates that the
slope inclination of this basin propagated towards the stream
flow; furthermore, most of the ES occurs towards the first
stream order due to predominance of 1st order number than
others. In Kangsabati basin, maximum stream length is ob-
served in Lalgarh 4SB (1168.545 km) and Khatra 1SB
(1104.92 km), while minimum stream length is observed in
Raipur 5SB (9.96 km) and Mohanpur 1SB (12.80 km) with
an average value of 126.60 km (Table 4). On the other hand,
Lu is higher in 1st order (1763.98 km), but small Lu length
presence in 6th order (92.98 km) in the entire sub-basins. In
order of average Lu, maximum value is observed in Lalgarh
4SB (66.1 km), while minimum value is observed in Mohanpur
6SB (3.11 km). According to stream length ratio (Rl), 6th
stream order in Khatra 1SB and Lalgarh 4SB has the highest
Rl value as 4.05 in comparison with other stream orders in the
entire basin. In contrary, average value of Rl has more fluctu-
ated ranges from 0.95 (Lalgarh 1SB) to 0 (Mohanpur 1SB).
WiththepresenceofthelargestRlvalue,mostoftheESoc-
curred in UC sub-basins. Based on logarithm value, successful
geometric relationship has been established among the stream
order, stream length, and basin area. Geometric relation also
helps to access the ES in the entire basin. In respect of
Horton’s(1932) law, two negative relationships have been
established in this study i.e., average logarithm value of stream
number (Nu) vs. stream order (u)(R
2
= 0.99) and average
logarithms value of stream order number (u) and stream length
(Lu) (R
2
=0.971)(Fig.4a, b), while positive relationship has
been established between average logarithms value stream
number (Nu) and stream length (Lu) (R
2
=0.968),respectively
(Fig. 4c). It can be said that ES gradually increases in the entire
basin with the increasing of stream number and stream length,
and decreasing of stream order.
Bifurcation ratio (Rb) among the twenty-seven sub-basins
of Kangsabati basin varies from the maximum value of 4.24
(Lalgarh 4SB) to minimum value of 0.8 (Dherua 2SB), but the
average value is of 1.72 (Table 5). Ranges of Rb denote that
ES is more fluctuated in the entire basin.
Characteristics of aerial morphometric aspects
In the present study, value of form factor (Ff) varies from 0.38
(Mohanpur 6SB) to 0.24 (Khatra 1SB and Lalgarh 4SB) with
an average value of 0.32 (Table 5). In terms of Ff, Kangsabati
basin is elongated in shape that is characterized by flatter peak
flow for a longer duration. Elongation ratio (Re) denotes that
maximum Re values are concentrated in Mohanpur 6SB
(0.69), and minimum value is concentrated in Khatra 1SB
(0.55), although mean Re value throughout the channel is
0.64 (Table 5). Re result predicts that UC is considered a high
susceptible site corresponded with low elongation ratio, but
LC is considered low susceptible site corresponded with high
elongation ratio. With the following of circular ratio (Rc), the
lowest value is situated in Khatra 2SB (0.13) as revealed the
high ES due to the predominance of low infiltration capacity,
while the highest Rc is situated in Mohanpur 6SB (0.56) as
indicated the low ES due to the high infiltration capacity. Rc
values in all sub-basins denoted that maximum sub-basins fall
under medium to higher susceptibility with the presence of
more erosion quantity in all tributaries. Moreover, Ff, Re,
and Rc suggested that the shape of the studied basin is medi-
um to lower elongated in form and shape to associate the
undulating relief along with gentle to steep ground slopes.
Drainage density (D) in this basin ranges from 0.37 km/
km
2
(Mohanpur 1SB) to 1.79 km/km
2
(Raipur 2SB) (Table 5),
which indicates that most of the sub-basins have medium to
high ES caused by medium to sparse vegetation cover under
Table 4 Stream order wise stream length in different sub-basin along
the studied basin
Sub-
basin
Order wise stream length (km)
1st 2nd 3rd 4th 5th 6th Total
1 578.17 242.33 129.6 68.1 43.44 43.21 1104.92
2 30.13 18.7 4.64 0 0 0 53.48
3 62.75 29.87 18.65 19.68 0.9 0 131.88
4 47.59 20.67 8.85 10 0 0 87.18
5 82.63 24.28 18.9 8.88 0 0 134.72
6 26.43 8.76 3.23 1.9 0 0 40.34
7 37.86 22.19 6.44 15.89 0.15 0 82.53
8 5.072 4.88 0 0 0 0 9.96
9 12.69 2.93 13.25 0 0 0 28.89
10 18.78 3.36 7.39 0 0 0 29.53
11 20.038 18.58 10.53 0 0 0 49.15
12 578.25 255.52 144.98 87.28 52.72 49.77 1168.5
13 9.618 10.19 1.997 0 0 0 21.8
14 19.6 5.87 5.28 0 0 0 30.76
15 32.55 17.26 16.1 4.92 0 0 70.
16 5.74 8.85 0 0 0 0 14.59
17 10.45 5.28 13.19 0 0 0 28.93
18 19.11 9 10.2 0 0 0 38.34
19 14.182 14 9 0.1 0 0 37.59
20 12.8 0 0 0 0 0 12.8
21 12.87 11.9 0.61 0 0 0 25.39
22 19 12.19 2.42 0 0 0 33.66
23 11 4.56 0 0 0 0 16
24 17.43 10.42 5 0 0 0 32.86
25 20.47 5.58 4.97 0 0 0 31.03
26 22.75 5.77 0 0 0 0 28.52
27 35.34 10.18 28.24 0 0 0 73.77
Page 9 of 22 501Arab J Geosci (2021) 14: 501

ab
cd
ef
501 Page 10 of 22
Arab J Geosci (2021) 14: 501

impermeable rock strata. In particular, most of the higher D
value has situated in the upper part of this basin, which reveals
the higher ES due to the presence of low permeability. While
low Dvalue is situated in the lower part of this basin, it reveals
the low ES with the presence of permeable rock strata.
Drainage frequency (Fs) in the present study varies from 2
km
2
(Raipur 2SB) to 0.05 km
2
(all Mohanpur SB) with an
average value of 0.73. This result demonstrated that most of
the erosion-prone sub-basins are associated in the upper part
of Kangsabati basin caused by steep ground slopes under the
impermeable rock cover, but sub-basins in lower part fall un-
der low ES caused by gentle slope under permeable strata.
Drainage texture (T) in this basin are ranges between 3.76
(Raipur 2SB) and 0.02 (Mohanpur 1SB) with an average of
0.899 (Table 5). Khatra 1, 3SB; Raipur 1, 2SB; and Lalgarh
4SB have coarse texture, and Mohanpur 4SB and Midnapore
1SB have medium to finer texture. This result revealed that
finer to intermediate texture has predominated in Kangsabati
basin. On the other hand, maximum compact coefficient (Cc)
is found in Khatra 2SB (2.79), but minimum value is found in
Dherua 4SB (1.66). Therefore, most of the sub-basins in UC
fall under elongated shape and erosion-prone category. The
maximum length of overland flow (Lo) is concentrated in
Mohanpur 4SB (1.82), whereas minimum value is observed
in Lalgarh 4SB (0.29). So, ES is high in most of the UC sub-
basins due to the presence of steep slope. Average Lo value
(0.81) denoted that undulating surface is predominated in the
entire Kangsabati basin. Constant of channel maintenance (C)
in this basin varies from a maximum value of 3.63 (Mohanpur
4SB) to minimum value of 0.56 (Khatra 1SB). This result
demonstrated that entire basin is characterized by steep to
moderate slope.
Characteristics of relief aspects
Basin relief (R) in the present basin ranges from 57 m
(Mohanpur 7SB) to 397 m (Khatra 1SB) with an average of
340 m; therefore relative relief in the entire basin indicates that
basin area is extended from plateau fringe to flood-prone plain
surface. Relief ratio (Rr) varies from 0.017 (Raipur 3SB) to
0.002 (Midnapore 1SB) with an average value of 0.009
(Table 5). Ruggedness number (Rn) ranges from maximum
of 0.61 (Khatra 1SB) to minimum of 0.022 (Mohanpur 7SB)
with an average of 0.123. Gradient ratio (Gr) in this basin
ranges from 0.015 (Khatra 3SB) to 0.0016 (Midnapore 1SB)
with an average of 0.0077, which indicates that stream gauge
height always parallel to the bed slope that means discharge
volume could be drained out over the year.
Sub-basin prioritization based on morphometric
analysis
After the assignment of priority rank from all morphometric
aspects based on compound factor (CF), four priority classes
have been identified i.e., high (1–10), moderate (11–15) low
(16–20), and very low (21–25), respectively (Fig. 5). CF value
and its assigned priority rank in the entire sub-basin are
depicted in Table 5. The results of sub-basin prioritization
demonstrated that Khatra 1, 2SB; Raipur 1, 2SB; and
Fig. 3 1st stream order direction (a). 2nd stream order direction (b). 3rd
stream order direction (c). 4th stream order direction (d). 5th stream order
direction (e). 6th stream order direction (f)
y = 42.007x0.4512
R² = 0.9686
1
10
100
1000
10000
1 10 100 1000 10000
Log of Lu in km
Log of total stream number(Nu)
a
y = 11695e-1.374x
R² = 0.9998
1
10
100
1000
10000
02468
Log of Nu
Log of stream order(u)
b
y = 2886.9e-0.621x
R² = 0.9714
1
10
100
1000
10000
02468
Log of Lu in km
Log of stream order(u)
c
Fig. 4 Geometric relationship between a stream length (Lu) and stream
number (Nu) (a). Stream order (u) and stream number (Nu) (b). Stream
order (u) and stream length (Lu) in the Kangsabati basin (c)
Page 11 of 22 501Arab J Geosci (2021) 14: 501

Lalgarh 4SB with the lowest amount of compound values
were assigned as high priority ranks i.e., 1, 2, 3, 4, 5 whereas
Mohanpur 1, 7, 4, 3SB with the highest amount of compound
values were assigned as very lower priority ranks i.e., 27, 26,
25, and 24 including nine sub-basins. In addition, the rest of
the sub-basins were given medium (seven sub-basins) and low
priority rank (six sub-basins) following the compound values
in respective sub-basins. It is a point that low and very low ES
in fifteen sub-basins and medium and high ES in twelve sub-
basins have been identified. On the other hand, maximum
high erosion susceptible sub-basins have been identified in
UC, and maximum low erosion susceptible sub-basins are
found in LC.
Hypsometric characteristic
In respect to ten elevation class’s landmass distribution,
twenty-seven hypsometric curves were estimated using
Eq. (22). In UC, hypsometric curves were estimated from
78.75% spread area extending from low elevation point of
193 m where elevation ranges from 41 to 481 m. In contrary,
21.14% spreads area extending from high elevation points
of 193 m and 337 m where elevation ranges from 433 to 481
m as occupied of 0.25% curve area (Fig. 6a). In LC, hypso-
metric curves were estimated from 96.25% spreads area
extending from 11 to 83 m where elevation ranges from
11 to 131 m. In contrary, 0.2% curve area occupied from
high elevation ranges from 107 to 131 m (Fig. 6b). All
datasets showed that the UC faces more instability with
the presence of effective geomorphic processes while LC
has been belonging in stable geomorphic stage due to the
absence of geomorphic processes. Hypsometric integral
values (HI) in the UC range from 0.53 to 0.32 with the
average value of 0.44, which reveals that landforms reach
the instable stage or under processing of landforms evolu-
tion phase (Fig. 6c). In contrary, the integral value in LC
ranges from 0.48 to 0.37 with an average value of 0.40 that
means basin achieves the least erosion to stable geomorphic
phase following Strahler (1952) law. Therefore, it can be
said that UC has not reached in equilibrium stage.
Table 5 Calculation of compound factor and prioritized ranks based on morphometric parameters in twenty-seven sub-basins
SB Rl Rbm Rn Gr Ff Re Rc Bs Cc DFs TCLo CF Pr
1 0.62 3.57 0.619 0.006 0.24 0.55 0.14 4.2 2.63 1.56 2 3.1 0.6 0.78 6.79 1
2 0.17 2.17 0.069 0.01 0.31 0.62 0.13 3.28 2.79 0.47 0.3 0.15 2.1 0.23 12.5 12
3 0.44 2.76 0.311 0.015 0.31 0.63 0.5 3.21 1.41 1.36 1.6 2.16 0.7 0.68 7.21 2
4 0.4 2.66 0.158 0.011 0.31 0.63 0.45 3.18 1.5 0.96 1.1 1.05 1 0.48 8.93 6
5 0.31 3.07 0.311 0.01 0.32 0.64 0.28 3.1 1.89 1.79 2.1 3.76 0.6 0.89 7.29 3
6 0.26 1.87 0.12 0.015 0.36 0.68 0.43 2.77 1.53 1.23 1.3 1.6 0.8 0.61 10.2 7
7 0.67 2.7 0.222 0.014 0.33 0.65 0.46 3.03 1.48 1.29 1.8 2.36 0.8 0.65 7.79 5
8 0.19 1.6 0.027 0.01 0.37 0.68 0.34 2.72 1.7 0.34 0.3 0.1 2.9 0.17 16.4 20
9 0.95 1.53 0.083 0.007 0.34 0.66 0.39 2.92 1.61 0.59 0.4 0.22 1.7 0.3 12.3 10
10 0.48 1.2 0.041 0.006 0.33 0.64 0.39 3.08 1.61 0.41 0.2 0.08 2.4 0.21 15.6 18
11 0.3 1.4 0.065 0.005 0.32 0.64 0.44 3.08 1.51 0.68 0.2 0.16 1.5 0.34 14.6 15
12 0.63 4.24 0.462 0.004 0.24 0.55 0.36 4.19 1.68 1.68 2.4 4 0.6 0.84 7.57 4
13 0.25 0.9 0.04 0.006 0.34 0.66 0.36 2.94 1.66 0.42 0.2 0.07 2.4 0.21 17.3 21
14 0.24 1.4 0.079 0.007 0.35 0.67 0.43 2.87 1.53 0.72 0.4 0.28 1.4 0.36 13.4 14
15 0.35 2.27 0.099 0.007 0.33 0.64 0.38 3.06 1.62 1.02 0.9 0.94 1 0.51 10.9 8
16 0.31 0.8 0.037 0.006 0.36 0.68 0.4 2.76 1.58 0.46 0.2 0.07 2.2 0.23 17.4 23
17 0.6 2 0.083 0.006 0.36 0.68 0.39 2.76 1.61 0.89 1 0.85 1.1 0.44 12.4 11
18 0.32 1.6 0.066 0.005 0.33 0.65 0.6 3.01 1.29 0.64 0.3 0.22 1.6 0.32 14.6 16
19 0.33 1.56 0.083 0.007 0.35 0.67 0.38 2.86 1.63 0.9 0.6 0.55 1.1 0.45 12.6 13
20 0 0 0.031 0.006 0.36 0.68 0.41 2.79 1.56 0.37 0.1 0.02 2.7 0.19 20.4 27
21 0.2 1.1 0.068 0.01 0.36 0.67 0.34 2.8 1.71 0.71 0.3 0.2 1.4 0.36 15.1 17
22 0.17 1.07 0.037 0.006 0.32 0.64 0.31 3.11 1.79 0.44 0.1 0.06 2.3 0.22 17.4 24
23 0.081 0.0230.0060.330.650.542.991.370.280.10.033.60.1419.625
24 0.22 1.07 0.037 0.005 0.33 0.65 0.42 3.05 1.55 0.49 0.2 0.08 2 0.24 17.3 22
25 0.23 1.21 0.086 0.014 0.38 0.69 0.57 2.65 1.33 1.3 1.6 2.07 0.8 0.65 12.1 9
26 0.051 0.0220.0040.320.640.373.091.640.390.10.032.60.1920.126
27 0.612 0.0450.0020.290.610.363.431.660.460.10.052.20.2316.219
501 Page 12 of 22
Arab J Geosci (2021) 14: 501

Estimated RUSLE parameters and potential soil
erosion
As per RUSLE results, Rfactor in Kangsabati basin
varies from 8075 MJ mm ha
−1
year
−1
(Dherua 4SB) to
4955 MJ mm ha
−1
year
−1
(Khatra 1SB) with an average
value of 6118 MJ mm ha
−1
year
−1
(Fig. 7a). In terms of
average Rvalue, maximum rainfall erosivity is concen-
trated in LC (6564 MJ mm ha
−1
year
−1
)thanUC
(5702 MJ mm ha
−1
year
−1
). Kfactor ranges from 0.37
(Mohanpur 4SB) to 0.1 (Raipur 1SB), and in particular,
maximum average value is concentrated in LC (0.33)
than UC (0.27) (Fig. 7b). In the present study, LS
values range from 1.49 (Khatra 1SB) to 0.47
(Midnapore 1SB); however, maximum average value is
concentrated in UC (0.87) than LC (0.57) (Fig. 7c). On
the other hand, Cvalue ranges from 0.3 (Raipur 3SB)
to of 0.2 (Mohanpur 2SB), while maximum average C
value was found in UC (0.25) than LC (0.24) (Fig. 7d).
In this study, Pfactor (1) presence in the entire UC that
means lack of conservation practices is predominated.
While Pfactor is absent in some sub-basins of LC,
meanwhile, basin management practices are found.
Spatial distribution of average annual soil erosion in the
entire Kangsabati basin is estimated by RUSLE model as
computed from pixel cell size of raster (30 m × 30 m) surface.
In terms of mean soil erosion measure as 300 t ha
−1
year
−1
,the
total amount of soil erosion was estimated as 8425 t ha
−1
year
−1
in Kangsabati basin during 2016–2017. Soil erosion
potentiality in present study has been categorized into seven
types i.e., very low (0–50 t ha
−1
year
−1
), low (51–100 t ha
−1
year
−1
), low to medium (101–150 t ha
−1
year
−1
), medium
(151–200 t ha
−1
year
−1
), medium to high (201–250 t ha
−1
year
−1
), high (251–300 t ha
−1
year
−1
), and very high soil ero-
sion zone (301-350 t ha
−1
year
−1
), respectively (Fig. 8). On the
other hand, mean soil erosion rate in twenty-seven sub-basins
has been delineated into four classes i.e., very low (<200 t ha
−1
year
−1
), low (200–300 t ha
−1
year
−1
), medium (300–400 t ha
−1
year
−1
), and high erosion class ( > 400 t ha
−1
year
−1
), respec-
tively (Fig. 9). In particular, high mean soil erosion rates are
found in Khatra 1SB and Lalgarh 4SB including six sub-ba-
sins, while very low mean erosion rates are found in
Mohanpur 7SB and Midnapore 1SB including five sub-ba-
sins. Moreover, medium mean erosion rates are found in
Khatra 2SB and Lalgarh 1, 5SB including seven sub-basins,
but low mean erosion rates are found in Dherua 1, 2SB
Fig. 5 Morphometric prioritization of Kangsabati basin
Page 13 of 22 501Arab J Geosci (2021) 14: 501

501 Page 14 of 22
Arab J Geosci (2021) 14: 501

including nine sub-basins. It is observed that fourteen sub-
basins in UC are produced medium to a higher amount of
erosion as 75% (225.66 t ha
−1
year
−1
), whereas the rest of
thirteen sub-basins in LC are produced low to medium amount
of erosion as 25% (74.39 t ha
−1
year
−1
). It is found that most of
the UC sub-basins face high ES, while maximum sub-basins
of LC face low ES.
Selection of effective morphometric parameters for
SES using AIC
Pairwise bivariate correlation showed that Rn (r=0.88),Rbm
(r=0.71),Bs(r=0.93),Fs(r=0.60),T(r=0.63),andD(r=
0.58) have a significant positive correlation, whereas Ff (r=
−0.89), Re (r=−0.90), and Rc (r=−0.48) showed significant
negative correlation with soil erosion (Fig. 10a). Such a pat-
tern of the relationship was also found by Ratnam et al. (2005)
under similar morphometric characteristics. The rest of the
morphometric parameters are also shown some degree of cor-
relation with soil erosion in the UC; however, these correla-
tions are insignificant. On the other hand, the correlation pat-
tern for the LC shows that Ff (r=−0.79), Re (r=−0.79), Gr (r
=−0.63), Fs (r=−0.60), T(r=−0.63), and D(r=−0.44) have
a significant negative correlation with soil erosion (Fig. 10b),
while Bs (r= 0.78) is the only morphometric parameters to
show a significant positive correlation in this context.
The stepwise regression was carried out with AIC values
according to Eq. (31). The least AIC value from linear, areal,
and relief aspects was adopted to build the multiple regression
models. The smallest AIC value of 65.35 is found through the
adding of Rn (p<0.0001)andGr(p<0.003)inreliefaspect
whereas AIC for the linear aspect would be 86.88 due to
adding of Rbm (p< 0.013), but AIC value decrease of 64.93
in aerial aspect with the adding of Ff (p< 0.034) and Re (p<
0.024) and drop the rest of the morphometric parameters from
three aspect (Table 6). Likewise, this model determines the
best possible combination i.e., Rn, Gr of relief aspect (R
2
=
0.90); Rbm of linear aspect (R
2
= 0.54); and Ff and Re of aerial
Fig. 6 Hypsometric characteristic: distribution of land mass with respect
to 10 classes of elevation in upper (a). Lower catchment (b). Hypsometric
curve with integral in upper catchment and lower catchment (c)
Fig. 7 RUSLE parameters: Rfactor (a). Kfactor (b). LS factor (c). Cfactor (d)
Page 15 of 22 501Arab J Geosci (2021) 14: 501

aspect (R
2
= 0.959) are the best fittest parameter from others
morphometric parameters for soil erosion in UC (Fig. 11a–c).
The predicted soil erosion while computed at the sub-basin
level shows a negative occurrence in its estimated value. A
detailed investigation about such pattern of occurrence reveals
that sub-basin nos. 2, 6, and 8 (for relief aspect) and 8, 10, 11,
13, and 14 (for linear aspect) in upper part of the study area are
having such pattern, and that might happen due to the pres-
ence of an inverse relationship of selected explanatory mor-
phometric variables for modeling the soil erosion. Although
mean bifurcation ratio, ruggedness index, and form factor play
a crucial role to increase the soil erosion, gradient ratio and
elongated ratio decreased the rate of soil erosion.
In LC, the least AIC value of 24.95 from relief aspect
denotes only Gr (p< 0.044) as a significant parameter in
explaining the quantum of soil erosion obtained by RUSLE
model (Table 7). But the least value of AIC from linear
(30.95) and aerial (10.15) aspects does not predict any signif-
icant parameter on soil erosion. Their estimated coefficient
values are quite different from each and other. Although, ae-
rial aspect has a significant positive relationship (R
2
=0.92)
with the soil erosion than linear (R
2
= 0.047) and relief aspect
(R
2
=0.323)(Fig.11d–f). The coefficient value of all mor-
phometric parameters reveals that gradient ratio has inverse
significant relationship with soil erosion, but the rest of pa-
rameters such as stream length ratio, mean bifurcation ratio,
drainage density, drainage frequency, and drainage texture
and constant channel maintenance ruggedness index, form
factor, elongation ratio, circular ratio, basin shape, and com-
pact coefficient have no significant relation with modeled soil
erosion.
Discussion
As per CF and RUSLE models, the entire basin has been clas-
sified into four ES classes as high, medium, low, and very low.
ES classification is prepared with the assigning of morphometric
parameters, soil structure, LULC patterns, conservative prac-
tices, and rainfall intensity, respectively. The result of predicted
model as CF showed that 18.51% sub-basins in high ES class,
25.92% sub-basins in medium ES class, 22.22% sub-basins in
low ES class, and 33.33% sub-basins in very low ES class have
been identified. On the other hand, estimated model as RUSLE
result showed that 22.22% sub-basins in high ES class, 25.92%
sub-basins in medium ES class, 29.62% sub-basins in low ES
class, and 22.22% sub-basins in very low ES class have been
identified in Kangsabati basin. Both model results demonstrated
that high ES class has the highest spatial extent than other clas-
ses. This result demonstrated that the Kangsabati basin area has
more sensible to erosion. In terms of study area demarcation,
maximum erosion susceptible sub-basins of UC are situated in
Fig. 8 Potential soil erosion zone
of Kangsabati basin
501 Page 16 of 22
Arab J Geosci (2021) 14: 501

plateau fringe area. Therefore, maximum erosion has been oc-
curringinUCsub-basinsthanLCduetopresenceofmorpho-
metric variability and responsible erosion factors.
According to multiple correlations among all morphomet-
ric variables, Rbm of linear aspect, Ff and Re of areal aspect,
and Rn and Gr of relief aspect have more effective key roles
on ES in UC sub-basins of tropical plateau fringe basin. Rbm
is the prime indicator of stream number and stream length
ratio including hierarchical setup of all stream orders, which
reveals the highest ES and sedimentation occurring in plateau
fringe sub-basins caused by maximum number of first stream
order and maximum stream length ratio (Biswas and Pani
2015; Bhattacharya et al. 2020a). In general, first stream or-
ders in plateau fringe basin are identified as ephemeral rill
channels that have great role to accumulation of sediment at
sub-basin outlets (Gayen et al. 2019; Bhattacharya et al.
2020a). In areal aspect, both Ff and Re are reflected the sig-
nificant role of all areal parameters on ES. Areal parameters
represent the elongated basin shape in plateau fringe site of
Kangsabati basin. The maximum value of D, Fs, Lo, Cc, and T
and minimum value of Care the causes of high ES in Khatra
1, 3SB; Raipur 2, 4SB; and Lalgarh 4SB of UC. In addition,
low infiltration capacity in impermeable resistance rock strata,
high runoff volume, and sparse vegetation cover with undu-
lating terrain are the causes of high ES in plateau fringe basin
(Ameri et al. 2018; Bhattacharya et al. 2019). In relief aspect,
higher value of Rn and Gr denotes maximum value of R,Rl,
Rbm, D, Fs, Cc, and Lo and minimum value of Ff, Re, and Rc,
which are the causes of high ES in UC. Two guiding param-
eters of Rn and Gr not only reflected the role of relief param-
eters role but also signified the role of linear and areal param-
eters on ES in response to different flow direction in each
stream orders, maximum flow length, higher stream gradient,
and undulating terrain surface in UC (Ameri et al. 2018;
Pradhan et al. 2018). Furthermore, in respect to five effective
parameters role, Raipur 5SB and Lalgarh 2, 3, 6SB are situat-
ed in UC, but CF predicted ES has been identified in medium
class due to presence of minimum Rbm, Rn, Gr values, and
maximum Ff and Re values, respectively. On the other hand,
hypsometric characteristics in all sub-basins of UC clarify the
Fig. 9 Mean annual soil erosion entire twenty-seven sub-basin in Kangsabati basin
Page 17 of 22 501Arab J Geosci (2021) 14: 501

Fig. 10 Correlation between the variables of morphometric parameters and soil erosion: upper catchment (a). Lower catchment (b)
501 Page 18 of 22
Arab J Geosci (2021) 14: 501

predicted ES as given by CF models. In RUSLE model, result
reveals that variability of low monsoon rainfall (low Rfactor),
low water holding capacity in soil with scanty organic con-
tents (minimum K factor), high elevated terrain and greater
slope length with maximum steepness value (high LS factor),
sparse vegetation cover with dominant barren land (maximum
Cfactor), and lack of conservative practices (presence of P
factor) are triggered the high ES in UC. These results are
validated with the results of Shit et al. (2015), Mahala
(2018), and Bhattacharya et al. (2020a,b).
Based on multiple correlations of morphometric variables,
Gr in relief aspect is an important key parameter to determine
the ES in response to other dominant erosion factors in undu-
lating plain surface of LC sub-basins likeMidnapore, 1SB and
Mohanpur 1, 2SB. Due to flat surface in circular basin shape,
lack of stream number of all orders, permeable non-resistant
rock strata, and dense vegetation, Gr play crucial role to in-
crease the infiltration rate and water holding capacity in soil
but decrease the runoff volume and sediment concentration,
which is the cause of low ES in LC (Altaf et al. 2014;Pradhan
R² = 0.9003
-50
0
50
100
150
200
-50 0 50 100 150 200
Soil Erosion (t/ha/yr)
Predicted Soil Erosion (t/ha/yr)
a
R² = 0.5355
-100
-50
0
50
100
150
-100 -50 0 50 100 150
Soil Erosion (t/ha/yr)
Predicted Soil Erosion (t/ha/yr)
b
R² = 0.9589
-50
0
50
100
150
200
250
-50 0 50 100 150 200 250
Soil Erosion (t/ha/yr)
Predicted Soil Erosion (t/ha/yr)
c
R² = 0.3993
-10
-5
0
5
10
15
20
-10 -5 0 5 10 15 20
Soil Loss (t/ha/yr)
Predicted Soil Erosion (t/ha/yr)
d
R² = 0.047
-4
-2
0
2
4
6
8
10
12
14
16
-4 1 6 11 16
Soil Erosion (t/ha/yr)
Predicted Soil Erosion(t/ha/yr)
e
R² = 0.9235
-5
0
5
10
15
20
25
-5 0 5 10 15 20 25
Soil Erosion (t/ha/yr)
Predicted Soil Erosion (t/ha/yr))
f
Fig. 11 Relationships between RUSLE modified actual soil erosion and significant morphometric estimated soil erosion: relief aspect (a). Linear aspect
(b). Aerial aspect in upper catchment (c). Relief aspect (d). Linear aspect (e). Aerial aspect in lower catchment (f)
Page 19 of 22 501Arab J Geosci (2021) 14: 501

et al. 2018). Despite the location in LC, CF predicted ES has
been identified as medium category in Dherua 1, 3SB and
Mohanpur 6SB for the presence of high gradient undulating
surface to increase the Rbm value for accelerating the erosion
rate. Moreover, Gr directly related to slope steepness and its
length (Schumm 1956; Fallah et al. 2016) that gradually de-
creases towards the floodplain sites of LC. On the other hand,
hypsometric integral values in thirteen sub-basins of LC de-
note that stable geomorphic processes reduce soil erosion rate
then try to reach the mature stage of landform evolution.
Result of RUSLE model demonstrated that intensive rainfall
during monsoon (high Rfactor), dense organic content and
high clay percentage with maximum water holding capacity
(maximum K factor), flat plain surface with lower slope steep-
ness and length (low LS factor), dense vegetation with double
crop cultivation (minimum Cfactor), and presence of conser-
vative practices (absence of Pfactor) are governed to low ES
as validated by the results of Mahala (2018) and Bhattacharya
et al. (2020a,b). In order of RUSLE estimated mean soil
erosion, UC sub-basins of Khatra 2SB; Raipur 2, 3, 5SB;
and Lalgarh 2, 3, 6SB fall under low and very low erosion
categories caused by lower amount of slope steepness and
length, whereas LC sub-basins of Dherua 4SB and
Mohanpur 1SB fall under high erosion category caused by
Table 6 Coefficients of selected morphometric parameters on soil erosion entire upper catchment
Aspect Source Value Standard error tPr > |t|AIC
Relief Intercept 15.18 7.42 2.04 0.065 65.34
Relief Rn 129.96 14.55 8.93 <0.0001 65.34
Relief Gr −2593.13 694.65 −3.73 0.003 65.345
Linear Intercept −33.47 15.32 −2.18 0.052 86.88
Linear RI 21.88 26.58 0.82 0.428 86.88
Linear Rbm 18.19 6.12 2.96 0.013 86.88
Areal Intercept 2597.66 671.26 3.87 0.012 64.92
Areal Ff 7305.87 2525.98 2.89 0.034 64.92
Areal Re −7686.13 2390.84 −3.21 0.024 64.92
Areal Rc −104.03 205.55 −0.5 0.634 64.92
Areal Cc −23.8 51.88 −0.45 0.666 64.92
Areal D64.36 76.51 0.84 0.439 64.92
Areal Fs 8.92 31.64 0.28 0.789 64.92
Areal T−22.81 32.19 −0.71 0.510 64.92
Areal C10.84 20.42 0.53 0.618 64.92
Table 7 Coefficients of selected morphometric parameters on soil erosion entire lower catchment
Aspect Source Value Standard error tPr > |t|AIC
Relief Intercept 9.35 1.78 5.24 0.000 24.95
Relief Rn 5.01 29.79 0.16 0.870 24.95
Relief Gr −610.27 265.47 −2.29 0.044 24.95
Linear Intercept 4.396 2.05 2.14 0.058 30.94
Linear RI −1.568 7.83 −0.2 0.845 30.94
Linear Rbm 1.356 2.43 0.55 0.590 30.94
Areal Intercept −117.95 353.84 −0.33 0.756 10.15
Areal Ff −291.52 1097.41 −0.26 0.804 10.15
Areal Re 212.46 1120.85 0.19 0.859 10.15
Areal Rc 81.52 35.36 2.3 0.082 10.15
Areal Cc 40.28 21.24 1.89 0.131 10.15
Areal D−13.93 11.17 −1.24 0.281 10.15
Areal Fs 3.28 6.98 0.47 0.663 10.15
Areal T−0.55 5.78 −0.09 0.928 10.15
Areal C−3.35 2.19 −1.52 0.201 10.15
501 Page 20 of 22
Arab J Geosci (2021) 14: 501

higher amount of slope steepness and length, respectively. In
spite of R,K,C,andPconsiderable roles, LS play a signified
role in these two different physiographic regions (plateau
fringe and undulating surface).
The analysis, however, shows morphometric aspect of
the study area of Kangsabati basin, and in particular,
the plateau fringe sub-basins are more suitable for
modeling soil erosion result than plain surface sub-
basins due to presence of several human activities such
as strip cropping, terracing intensification, and
contouring in the LC. The results, however, indicate
that morphometric variables can also be used to esti-
mate soil erosion parallel to the use of RUSLE since
they are having a significant level of correlation be-
tween them (R
2
= 0.992). Therefore, predicted soil ero-
sion perfectly correlated with modeled soil erosion in
the plateau fringe basin than LC. Moreover, these re-
sults help to take planning of soil and water conserva-
tion measures in high erosion susceptible sub-basins,
especially plateau fringe sites to ensure the extensive
ecological adaptation.
Conclusion
Evaluation of ES in Kangsabati basin using the CF and
RUSLE prediction models under GIS platform has reflected
the significant role of effective morphometric parameters and
other responsible factors that create two different susceptible
sites i.e., the most exposed to erosion in plateau fringe site of
UC and the least affected to erosion in undulating plain sur-
face of LC. In line with obtaining results, morphometric pa-
rameters have an effective role to create many hot spots of ES
in plateau fringe sites with the presence of elongated basin
shape and high elevated undulating terrain surface including
steep slope, while other factors like cropping management and
intense conservative practices with maximum water holding
capacity in soil profile have a greater role to create many cold
spots of ES in LC in respect to low gradient ratio or small
slope steepness and its length. The result reveals that conser-
vation practices are essentially needed to protect the soil ero-
sion especially in the plateau fringe site of Kangsabati basin.
In spite of morphometric properties in a drainage basin, sev-
eral conservative measures have a wider implication to check
the soil erosion.
The present study showed that both CF predicted ES and
RUSLE estimated soil erosion are facilitated to realistic mea-
sure of sub-basins priority. According to CF and RUSLE re-
sults, pragmatic method of AIC is applied to select the best
effective morphometric parameters that have great influence
in the basin management as well as water resource
conservation.
Supplementary Information The online version contains supplementary
material available at https://doi.org/10.1007/s12517-021-06819-8.
Acknowledgements The authors are thankful to the Survey of India
(SOI), Irrigation Office of Paschim Medinipur, and Bankura for provid-
ing required data. We would like to thank the Editor-in-chief Abdullah M.
Al-Amri and four anonymous reviewers for their valuable comments and
suggestions on previous manuscript version that have great role to im-
prove this revised manuscript.
Declarations
Conflict of interest The authors declare that they have no competing
interests.
References
Akaike H (1973) Problems of control and information. In 2nd
International Symposium on Information Theory, Budapest:
Akademiai Kaido (pp. 267-281)
Altaf S, Meraj G, Romshoo SA (2014) Morphometry and land cover
based multi-criteria analysis for assessing the soil erosion suscepti-
bility of the western Himalayan watershed. Environ Monit Assess
186(12):8391–8412
Ameri AA, Pourghasemi HR, Cerda A (2018) Erodibility prioritizationof
sub-watersheds using morphometric parameters analysis and its
mapping: a comparison among TOPSIS, VIKOR, SAW, and CF
multi-criteria decision making models. Sci Total Environ 613:
1385–1400. https://doi.org/10.1016/j.scitotenv.2017.09.210
Bakker MM, Govers G, van Doorn A, Quetier F, Chouvardas D,
Rounsevell M (2008) The response of soil erosion and sediment
export to land-use change in four areas of Europe: the importance
of landscape pattern. Geomorphology 98(3):213–226
Bhattacharya RK, Chatterjee ND, Das K (2019) Multi-criteria-based sub-
basin prioritization and its risk assessment of erosion susceptibility
in Kansai–Kumari catchment area, India. Appl Water Sci 9(4):76.
https://doi.org/10.1007/s13201-019-0954-4
Bhattacharya RK, Chatterjee ND, Das K (2020a) An integrated GIS ap-
proach to analyze the impact of land use change and land cover
alteration on ground water potential level: a study in Kangsabati
Basin, India. Groundwater for Sustainable Development, 100399
Bhattacharya RK, Chatterjee ND, Das K (2020b) Estimation of erosion
susceptibility and sediment yield in ephemeral channel using
RUSLE and SDR model: tropical plateau Fringe Region, India gully
erosion studies from India and surrounding regions, Advances in
Science, Technology & Innovation, P. K. Shit et al. (eds.), https://
doi.org/10.1007/978-3-030-23243-6_10
Bhattacharya RK, Chatterjee ND, Das K (2020c) Land use and land cover
change and its resultant erosion susceptible level: an appraisal using
RUSLE and logistic regression in a tropical plateau basin of West
Bengal, India. Environ, Dev Sustain, 1-36. https://doi.org/10.1007/
s10668-020-00628-x
Biswas SS, Pani P (2015) Estimation of soil erosion using RUSLE and
GIS techniques: a case study of Barakar River basin, Jharkhand,
India. Model Earth Syst Environ 1(4):42
Biswas H, Raizada A, Mandal D, Kumar S, Srinivas S, Mishra PK (2015)
Identification of areas vulnerable to soil erosion risk in India using
GIS methods. Solid Earth 6:1247–1257
Deepika B, Avinash K, Jayappa KS (2013) Integration of hydrological
factors and demarcation of groundwater prospect zones: insights
from remote sensing and GIS techniques. Environ Earth Sci 70(3):
1319–1338
Page 21 of 22 501Arab J Geosci (2021) 14: 501

Fallah M, Kavian A, Omidvar E (2016) Watershed prioritization in order
to implement soil and water conservation practices. Environ Earth
Sci 75(18):1248
Fang G, Yuan T, Zhang Y, Wen X, Lin R (2019) Integrated study on soil
erosion using RUSLE and GIS in Yangtze River Basin of Jiangsu
Province (China). Arab J Geosci 12(5):173
Gajbhiye S, Sharma SK, Meshram C (2014) Prioritization of watershed
through sediment yield index using RS and GIS approach. Int J Serv
Sci Technol 7(6):47–60
Gayen A, Pourghasemi HR, Saha S, Keesstra S, Bai S (2019) Gully
erosion susceptibility assessment and management of hazard-prone
areas in India using different machine learning algorithms. Sci Total
Environ 668:124–138
Horton RE (1932) Drainage-basin characteristics. EOS Trans Am
Geophys Union 13(1):350–361
Jordan G, Van Rompaey A, Szilassi P, Csillag G, Mannaerts C, Woldai T
(2005) Historical land use changes and their impact on sediment
fluxes in the Balaton basin (Hungary). Agric Ecosyst Environ
108(2):119–113
Karamage F, Zhang C, Ndayisaba F, Shao H, Kayiranga A, Fang X,
Nahayo L, Nyesheja EM, Tian G (2016) Extent of cropland and
related soil Erosion risk in Rwanda. Sustainability 2016(8):609
Kumar A, Devi M, Deshmukh B (2014) Integrated remote sensing and
geographic information system based RUSLE modelling for estima-
tion of soil loss in western Himalaya, India. Water Resour Manag
28(10):3307–3317
Mahala A (2018) Soil erosion estimation using RUSLE and GIS
techniques—a study of a plateau fringe region of tropical environ-
ment. Arab J Geosci 11(13):335
Malik A, Kumar A, Kandpal H (2019) Morphometric analysis and prior-
itization of sub-watersheds in a hilly watershed using weighted sum
approach. Arab J Geosci 12(4):118
Masselink RH, Temme AJAM, Giménez Díaz R, Casalí Sarasíbar J,
Keesstra S D (2017) Assessing hillslope-channel connectivity in
an agricultural catchment using rare-earth oxide tracers and random
forests models. Cuadernos de Investigación Geográfica n 43 (1), pp.
19-39. https://doi.org/10.18172/cig.3169
McCool DK, Brown LC, Foster GR, Mutchler CK, Meyer LD (1987)
Revised slope steepness factor for the universal soil loss equation.
Trans ASAE 30(5):1387–1396
Meshram SG, Sharma SK (2017) Prioritization of watershed through
morphometric parameters: a PCA-based approach. Appl Water Sci
7(3):1505–1519
Mulliner E, Malys N, Maliene V (2016) Comparative analysis of MCDM
methods for the assessment of sustainable housing affordability.
Omega 59:146–156
Nasir N, Selvakumar R (2018) Influence of land use changes on spatial
erosion pattern, a time series analysis using RUSLE and GIS: the
cases of Ambuliyar sub-basin, India. Acta Geophysica 66(5):1121–
1130
Patel DP, Dholakia MB, Naresh N, Srivastava PK (2012) Water harvest-
ing structure positioning by using geo-visualization concept and
prioritization of mini-watersheds through morphometric analysis in
the Lower Tapi Basin. J of Indian Soc of Rem Sensing 40(2):299–
312
Patel DP, Gajjar CA, Srivastava PK (2013) Prioritization of Malesari
mini-watersheds through morphometric analysis: a remote sensing
and GIS perspective. Environ Earth Sci 69(8):2643–2656
Pham TG, Degener J, Kappas M (2018) Integrated universal soil loss
equation (USLE) and geographical information system (GIS) for
soil erosion estimation in a sap basin: Central Vietnam. Int Soil
Water Conserv Res 6(2):99–110
Pradhan R, Srivastava PK, Maurya S,Singh S, Patel DP (2018) Integrated
framework for soil and water conservation in Kosi River Basin.
Geocarto International, 1-20
Ratnam KN, Srivastava YK, Rao VV, Amminedu E, Murthy KSR (2005)
Check dam positioning by prioritization of micro-watersheds using
SYI model and morphometric analysis—remote sensing and GIS
perspective. J Indian Soc of Rem Sensing 33(1):25–38
Renard KG (1997) Predicting soil erosion by water: a guide to conserva-
tion planning with the Revised Universal Soil Loss Equation
(RUSLE). United States Government Printing, Washington D.C
Rodrigo-Comino J, Iserloh T, Lassu T, Cerdà A, Keestra SD, Prosdocimi
M, Brings C, Marzen M, Ramos MC, Senciales JM, Ruiz Sinoga
JD, Seeger M, Ries JB(2016) Quantitative comparison of initial soil
erosion processes and runoff generation in Spanish and German
vineyards. Sci Total Environ 565:1165–1174. https://doi.org/10.
1016/j.scitotenv.2016.05.163
Sahaar AS (2013) Erosion mapping and sediment yield of the Kabul
River Basin, Afghanistan (Doctoral dissertation, Colorado State
University)
Schumm SA (1956) Evolution of drainage systems and slopes in bad-
lands at Perth Amboy, New Jersey. Geol Soc Am Bull 67(5):597–
646
Shit PK, Nandi AS, Bhunia GS (2015) Soil erosion risk mapping using
RUSLE model on Jhargram sub-division at West Bengal in India.
Model Earth Syst Environ 1(3):28
Sridhar P, Kumari P, Ganapuram S (2018) Assessment and planning for
integrated river basin management using remote sensing, SWAT
model and morphometric analysis (case study: Kaddam River
Basin, India). Geocarto Int, pp. -1-31 https://doi.org/10.1080/
10106049.2018.1489420
Strahler AN (1952) Dynamic basis of geomorphology. Geol Soc Am Bull
63(9):923–938
Strahler AN (1964) Quantitative geomorphology of drainage basin and
channel networks. Handbook of applied hydrology
Toubal AK, Achite M, Ouillon S, Dehni A (2018) Soil erodibility map-
ping using the RUSLE model to prioritize erosion control in the
Wadi Sahouat basin, North-West of Algeria. Environ Monit
Assess 190(4):210
Uddin K, Murthy MSR, Wahid SM, Matin MA (2016) Estimation of soil
erosion dynamics in the Koshi Basin using GIS and remote sensing
to assess priority areas for conservation. PLoS One 11(3):e0150494
UN-FAO. (2001). Strategic environmental assessment: an assessment of
the impact of cassava production and processing on the environment
and biodiversity (Vol. 5). Rome.
Wischmeier WH, Smith DD (1978) Predicting rainfall erosion losses-a
guide to conservation planning. Predicting rainfall erosion losses-a
guide to conservation planning
501 Page 22 of 22
Arab J Geosci (2021) 14: 501