See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/345310143

Soil erosion susceptibility mapping using a GIS-based multi-criteria decision

approach: Case of district Chitral, Pakistan

Article in Ain Shams Engineering Journal · November 2020

DOI: 10.1016/j.asej.2020.09.015

CITATIONS READS

60 1,599

6 authors, including:

Bilal Aslam Ahsen Maqsoom

Northern Arizona University University Mohammed VI Polytechnic
56 PUBLICATIONS 1,010 CITATIONS 144 PUBLICATIONS 2,488 CITATIONS

SEE PROFILE SEE PROFILE

Wesam Alaloul Muhammad Ali Musarat

Universiti Teknologi PETRONAS University of Malta
257 PUBLICATIONS 4,592 CITATIONS 178 PUBLICATIONS 2,842 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Muhammad Ali Musarat on 04 November 2020.

The user has requested enhancement of the downloaded file.

 Ain Shams Engineering Journal xxx (xxxx) xxx

Contents lists available at ScienceDirect

Ain Shams Engineering Journal

journal homepage: www.sciencedirect.com

Electrical Engineering

Soil erosion susceptibility mapping using a GIS-based multi-criteria

decision approach: Case of district Chitral, Pakistan
Bilal Aslam a, Ahsen Maqsoom b, Wesam Salah Alaloul c,⇑, Muhammad Ali Musarat c, Talha Jabbar b,
Ahmed Zafar b
a
Department of Earth Sciences, Quaid-i-Azam University, Islamabad, Pakistan
b
Department of Civil Engineering, COMSATS University Islamabad Wah Campus, Wah Cantt, Pakistan
c
Department of Civil and Environmental Engineering, Universiti Teknologi PETRONAS, Perak, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: Soil erosion has serious threats to agricultural production, hydraulic structures, and the world’s ecosys-
Received 4 May 2020 tem. The objective of this study is the delineation of soil erosion susceptibility zones in the Chitral district
Revised 3 September 2020 using spatial analyst tool in conjunction with Analytical Hierarchy Process (AHP). This district is highly
Accepted 5 September 2020
vulnerable to soil erosion due to the mountainous topography. For the sustainability of agricultural land
Available online xxxx
use as well as regional and local development, it is necessary to find out soil erosion probability zones and
soil loss at the watershed scale. Eleven different factors; lithology, slope, elevation, plain curvature, lin-
Keywords:
eaments, land cover, aspect, rainfall, drainage density, NDVI (Normalized Difference Vegetation Index)
Soil erosion
Multi decision making criteria
and NDWI (Normalized Difference Water Index) are considered in this study. Weights have been assigned
Analytical hierarchy process to each factor, and maps have been generated through GIS (Geographic Information System) tools. The
Soil loss final map from the combination of all maps shows intensities of soil erosion in five different classes
including very high, high, medium, low and very low. Very high and high erosion is observed in 13%
and 18% of the total study area respectively, which shows that area is under serious danger of soil erosion.
Elevation, slope, curvature, NDWI, and rainfall are found to be the dominant factors influencing soil ero-
sion process. This study highlighted areas at risk of severe erosion which will be helpful for researchers
and planners to plan for control of soil erosion.
Ó 2020 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Ain Shams Uni-
versity. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

1. Introduction the soil. It causes a reduction in the productivity of soil, pollutes

rivers and fills water reservoirs [34]. Many activities enhance rates
Loss of soil from the top surface by agents like air or water is of erosion and transportation and increase the process of sedimen-
called soil erosion [37]. The most prevailing erosion causing oper- tation. These activities involve the development of urbanization,
ator is water which erodes soil by raindrop impact and flowing roads, highways, control works on rivers and streams and mining
effect and phenomena involve steps like disengagement of soil par- [53]. The concentrated water formed by gullies and rills becomes
ticles, transport of particles and settlement of those particles which a source of stream channel erosion and washes away sediments
are called sediments [32]. Erosion due to the flow of water is a nat- from stream banks and streambeds. Erosion from banks in streams
ural environment-related process that has an unfriendly effect on causes the formation of meandering in channels that becomes a
environment, organic, physical and chemical characteristics of source of excessive erosion and it deposits soil particles in flood-
plain areas [42,53]. If transport capacity is less than the quantity
of detached soil, the only transportable quantity will be passed
⇑ Corresponding author.
downslope and remaining eroded soil particles will be dropped
E-mail address: [email protected] (W. Salah Alaloul).
on the section.
Peer review under responsibility of Ain Shams University.
If the soil formation rate is greater than the rate of soil erosion,
then it is threatening to sustainability, and it negatively affects the
productivity of soil [31]. Soil erosion is a serious natural issue
nowadays, and it subverts the world’s population by diminishing
Production and hosting by Elsevier
production in agriculture due to the lost nutrients and supple-

https://doi.org/10.1016/j.asej.2020.09.015
2090-4479/Ó 2020 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Ain Shams University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article as: B. Aslam, A. Maqsoom, W. Salah Alaloul et al., Soil erosion susceptibility mapping using a GIS-based multi-criteria decision
approach: Case of district Chitral, Pakistan, Ain Shams Engineering Journal, https://doi.org/10.1016/j.asej.2020.09.015
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Table 1 ferent erosion properties, as well as different sector requirements

Fundamental scale for pairwise comparison [66]. [55].
Rating Scale Numerical Reciprocal The Erosion Potential Method (EPM) is considered appropriate
Extremely preferred 9 1/9 for watershed administration on catchment level. Tazioli [63] and
Very strongly to extremely preferred 8 1/8 Memarian et al. [39] compared erosion by direct sediment trans-
Very strongly preferred 7 1/7 port measurements and the Gavrilovic model’s (1972) by EPM.
Strongly to very strongly preferred 6 1/6 The outcomes were adapted for various furnished basins in Africa
Strongly preferred 5 1/5
Moderately to strongly preferred 4 ¼
and Italy. Based on numerical results attained for some basins of
Moderately preferred 3 1/3 the Marche area in Italy, a comparison was made with EPM’s
Equally to moderately preferred 2 ½ empirical formula for erosion evaluation. It was concluded from
Equally preferred 1 1 Tazioli’s research that EPM is beneficial for small and medium
water paths that permitted to carry out erosion assessment in
the valley. This approach is also in practice in Croatia, Bosnia &
ments from topsoil [32]. The topsoil loses its fertility by this phe- Herzegovina, Italy, Macedonia, Montenegro, Slovenia, and Serbia.
nomenon which affects agricultural production and vegetation Transport of soil, reservoir sedimentation, as well as sediment
cover; therefore, it is classified as one of the serious natural issues. yields are calculated using EPM, and it is reputed because of its
According to a rough estimate, 10 million hectares of cultivated high authenticity [55]. The techniques based on AHP were applied
land are ruined because of soil erosion per year. Due to this reason, in many recent studies in geosciences especially in soil erosion and
the production of crops is reduced, leading to falling economies landslide susceptibility assessment, analysis of groundwater and
and increased undernourishment [29]. The quality of land-use is suitability of site [12,14,26,29,41]. However, this technique is still
deteriorated by soil erosion, and it is commonly found in agricul- novel to study the susceptibility of erosion. In the present study,
tural watersheds. Globally, rates of soil erosion are 10–40 times by use of Analytical Hierarchy Process (AHP), an endeavor is made
higher than rates of soil formation which is threatening to food to assess susceptibility of erosion in the Chitral River. The results of
security and quality of environment [54]. this attempt may bring value for decision-makers to form support-
Severe tillage, lack of vegetative-concealment and some charac- ive strategies for land efficiency and a supportive environment
teristics of soil lead to exposure of agricultural and cultivated [40,51].
regions to soil erosion. Harmful farming practices and irrational
land use decrease efficiency, sustainability and productivity of soils 2. Study area
and affects soil quality badly [14]. Soils have been manipulated
from centuries which have reduced nutrients and organic matter Chitral is the mountainous northern area of Pakistan. It is cov-
of soils in numerous regions of the world. Due to structural rear- ered with vegetation and tall trees due to excessive rainfall. Most
rangement of clays in soil, bulk densities of soils over the long term of the peaks of high mountains remain covered by snow in the
have been added by tillage [15]. Consequently, the rate of sediment winter season. The climate of this region is not too harsh for sum-
yield and runoff is increased. Researchers have proven that the use mer, but winters are severe here. This area is also known for its
of herbicide also increases susceptibility to soil erosion [21]. Soil natural beauty. River Indus has many indirect tributaries, and Chi-
erosion rate in Asia is 29.95 t/ha/y which indicates a hazard and tral River is among those subsidiary tributaries. Origin of the river
a need for management [50]. Asian rivers contribute about 80% is in far north from Baroghil pass in glaciated Hindu-
sedimentation production to the world oceans [43]. Hence, degra- Kush mountains of Chitral, Khyber Pakhtunkhwa, Pakistan. The
dation of soil and its supervision for sustainability is a serious and highest segment of this river is recognized by Yarkhum River. From
attention-seeking topic that needs to be addressed properly. There- the town of Mastuj to its confluence with Lotkoh River, down-
fore, soil erosion hazard assessment is necessary for detecting stream it is known as MastujRiver. Further downstream it is ter-
regions susceptible to intense erosion and to initiate appropriate med as Chitral River which is stationed at coordinates 35°50 and
land management campaigns. These sorts of evaluation in hydro- 71°48. It finally drains into Kabul River which is just to the east
geomorphologic qualities, geo-environmental and inadequate data of the city of Jalalabad in Afghanistan. Chitral river covers a drai-
situations relating to catchment approach or basin oriented are 2
nage area of 11.396 km . Last 34 years of record (1964–98) show
equipped to show great predictability [46]. The study of erosion
driven parameters and their logical and systematic assimilation that average annual runoff of river is 8670mm3 . The maximum
is a significant tool in this regard. The analysis can be carried out and minimum discharge recorded was 1586 m3 /s on July 16,
more appropriately through Geographic Information System (GIS) 1973, and 46 m3 /s on March 10, 1964 respectively. The melting
and Remote Sensing (RS). of glaciersand snow feeds the river, so it flows undeviatingly whole
The field-based techniques for quantification of soil loss and year. Some extra runoff can be recorded during the season of mon-
spatial diversity are cost and time-devouring, and deficiency in soon, i.e. Jul-Sept. River siphons a wide area of a steep slope and
inadequate sampling plots can reduce the consistency of actual widespread area covered with snow, and this contributes half of
spatial degree of the region suffering from soil erosion. In an out- the discharge to the Kabul River. A dam was constructed on the
dated field-based process, precise mapping and monitoring of spa- Chitral River for irrigation and hydropower generation having a
tial outline of soil loss in very large zones is certainly a difficult job maximum output of 250,000 k-Watts. This dam is located about
because it is costly and time consuming [48]. Numerous models 30 km west of Peshawar. The reservoir is almost full due to exten-
and methods can be used to outline zones susceptible to soil ero- sive silting. Now power generation capacity of the dam has
sion with an arrangement of various mathematical and geospatial dropped to 64,000–20,000 kW throughout summers and winters,
skills. Different strategies and procedures utilized for the evalua- correspondingly. Fig. 1 shows the study area of the District Chitral,
tion of vulnerability and soil loss quantification are specified in Pakistan.
previous studies [49]. Soil erosion and intensity of runoff can be
calculated by several empirical formulas. Steps involved in these 3. Methodology
methods are the acquisition of data, specification of model and
estimation [36]. Factors considered were dependent on scale, dif- The detailed methodology used in this study is provided in
Fig. 2.
2
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 1. Study area of District Chitral, Pakistan.

One of the widely used and potential approaches is called the logical and seismic maps helped in the digitization of faults, lithol-
Analytic Hierarchy Process (AHP). Based on the researches con- ogy and seismic zones. Around digitized faults, several ring buffer
ducted previously, critiques of professionals, nature and physical polygons of 500, 1090, 1500 and 2000 m were produced. Then vec-
traits of parameters helped in figuring out their encumbrance by tor layers of distance from faults and drainage, seismic intensities
use of AHP [25]. and lithology were rasterized. NDWI and NDVI are calculated from
There is a specific subdivided range of classes for each parame- the LANDSAT satellite while FAO published land cover data is used
ter having a rating value between 1 and 9, where 1 and 9 show for this study area which has very good accuracy. Data extraction
least and maximum susceptibility to erosion respectively. These techniques are mentioned in Table 2.
rating values show the possibility of a category to originate soil
instability. Each class has a rating value based on previously con-
ducted researches and personal evaluations of professionals 3.2. Analytical Hierarchy process (AHP) method
[20,62]. Satty [66] recommended numerical scale to be used as a
source for these values, which have been allocated analytically The procedure of AHP is classified into three steps which
using the AHP method as shown in Table 1. include identification of a hierarchy of objectives, criteria, and
alternatives; pairwise comparison of criteria; integration with the
result from pairwise comparison as relative importance in overall
3.1. Structuring of thematic maps levels of hierarchy. Initially, this approach assimilates the disinte-
gration of decision-making issue into a sequence of the importance
In GIS natural break method has been used for the arrangement of criteria and alternatives. According to the importance of each
of thematic layers of slope, curvature, elevation, and aspect. A the- factor, AHP appoints preference values to decide relative signifi-
matic layer of land cover was prepared using the land cover map. cance in association with objective [15].
Precipitation map was obtained via interpolation of annual rainfall Priority weights of each factor by calculation of normalized
records, and it was then blended with the earlier issued annual eigenvector are specified in the next step that’s comparatively a
precipitation map of the Pakistan Metrological Department better procedure for evaluating priorities from an unpredictable
(PMD). From Arc Hydro Tools, drainage was extracted by the usage pairwise evaluation matrix. The addition of values in every column
of STRM-based DEM. For the formation of a thematic layer of drai- of pairwise comparison matrix and value of every cell is divided by
nage distance, a 300 m buffer polygon was formed to get distance combined values of constant factors column assist in deriving
around streams [40]. This distance was taken by keeping Pavelsky weights. The primary eigenvector of the matrix is the mean value
and Smith [44] study into consideration. Previously published geo- of every row. A few degrees of irregularity may happen when hap-
3
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 2. Methodology adopted for this study.

Table 2
Techniques for production of various thematic data layers.

Parameter Source Techniques References

Elevation STRM-based DEM 30 m  30 m digital elevation model Saha et al. [52]
Slope STRM-based DEM Tanh ¼ 636:6
Ni
N = no of contour cutting; i = contour interval Wentworth [57]
p pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Aspect STRM-based DEM Rðp; qÞ ¼ 1  2ðb=kdÞ 1 þ q þ q2 =ð1  p þ q Horn [64]
Curvature STRM-based DEM K ¼ k dT k Zevenbergen & Thorne [65]
ds
Land use/land cover FAO Land cover Maximum likelihood Anderson [13]
Drainage density STRM-based DEM Proximity analysis Pavelsky and Smith [44]
P h   i
Rainfall PMD R ¼ 12 2 Arnoldus [16]
i¼1 1:73510 1:5log10 P i jP  0:08188

Lineament Geological Map of Pakistan Proximity analysis Pavelsky and Smith [44]
Lithology Geological Map of Pakistan Proximity analysis Pavelsky and Smith [44]
NDWI LANDSAT NDWI ¼ ððGreenþNIR
GreenNIRÞ
Þ
Ekumah et al. [28]
NDVI LANDSAT NDVI ¼ ðNIRBANDIRBAND
NIRBANDþIRBANDÞ
Carlson and Ripley [22]

hazardly arranged this network [18]. The consistency ratio is ðkmax  nÞ

CI ¼ ð2Þ
obtained using the following Eqn 1: n1
CI
CR ¼ ð1Þ where ‘‘n” represents the order of the matrix and "k.Max" is taken as
RI
maximum eigenvalue [40]. Identification of areas prone to soil ero-
where "CI" represents the Consistency Index and "RI" represents sion is considered as a decision-maker in this analysis. Significance
Random Index. The consistency ratio’s value decides the exclusion of 11 criteria of geo-environment, i.e. lithology, land cover, drai-
or inclusion of a variable within a study. The numerical value of nage, slope, lineament, aspect, elevation, NDVI (normalized differ-
CR should not be greater than 0.1. It should be less than or equal ence vegetation index) curvature, soil type, NDWI (normalized
to 0.1 [19]. The output parameter (CI) shows the consistency of difference water index) and rainfall and their sub-criteria (each cat-
one’s judgment. The following Eqn 2 can be used to calculate the egory of factors) is analyzed by this AHP method. The distribution of
value of CI: factors over catchment areas and decisions of experts for making
comparison matrices is supported by preference values of impor-
4
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

tance allotted to them and subclasses. CR of every factor was eval- 4.2. Slope
uated for exclusion or inclusion from or in the study.
The slope gradient plays a vital role in erosion control. It is a
well-understood fact that steeper slopes have high chances of soil
4. Results erosion [5]. It is suitable for the flow of surface and soil erosion [6].
It is calculated by using a number of contour cutting and contour
As mentioned earlier, in AHP pairwise comparison is drawn and interval. The steepness and length of the slope are important fac-
then relative weights for each factor has been calculated using tors for runoff and soil erosion. Another slope factor that
standard AHP methodology as shown in Table 3. All the pairwise affects erosion is the shape of the slope. Generally, different ero-
ranking were given purely on local knowledge based on experience sion and runoff characteristics exist in different slopes which can
and published literature. Multiple references studies be classified as a uniform, concave, convex and complex shape.
[47,26,33,46,47] have also been followed for ranking selection, The area is distributed into 6 divisions ranging from less than 12
and the CR value indicated that the ranking and weightage are degrees to greater than 47 degrees, as shown in Fig. 3b. Green color
unbiased and can be trusted. Each pixel will calculate soil hazard shows low slope areas while red color shows steeper slopes greater
from following derived Eqn (3), which is the dataset multiplies than 47 degrees. Soil erosion is high in the south-east part of Chi-
with its weight and the sum of all resultant datasets will be soil tral district, that is why the erosion rate is high in that area. The
hazard value for each pixel. effect of slope gradient on soil erosion of the study area is shown
in Table 3 using the AHP technique. The AHP pairwise comparison
SoilHazard ¼ Elev ation  0:091388 þ Slope  0:165733 shows that slope and elevation significantly increase erosion of
soil. Furthermore, slope and land use also increase soil erosion with
þ Aspect  0:026046 þ Cur v ature  0:119522 an index factor of 3.
þ Landuse  0:077331 þ Drainage  0:050972
þ Rainfall  0:163220 þ Lineaments  0:076627 4.3. Aspect
þ Lithology  0:022920 þ NDWI  0:165733
Aspect is another soil erosion factor that importantly affects the
þ NDVI  0:040552 ð3Þ type of greenness, which control period of sunlight, wetness, and
water loss and have influence action of erosion [2]. This shows
the erosion rate by considering discharge proportional to catch-
ment area and slope. The DEM map in Fig. 3c shows directional
4.1. Elevation
classes of study area concerning their influence on the erosion pro-
cess [7]. It is arranged in to classes as flat (1), north (0–22.5),
Elevation (from an ASTER digital elevation model (DEM) at a 30-
north-east (22.5–67.5), east (67.5–112.5), south-east (67.5–
m resolution) influences soil organic matter, and it is considered as
157.5), south (157.5–202.5), south-west (202.5–247.5), west
a significant factor of soil erosion [30]. The natural and anthro-
(247.5–292.5), north-west (292.5–337.5), north (337.5–360). The
pogenic conditions exert great influence on the rise or develop-
pairwise criteria in Table 3 show that it doesnot have a significant
ment of the erosion process. The previous researches also
effect when combined with other factors (0.2–1).
highlighted the impact of elevation upon soil erosion. The eleva-
tion was one of the most influential factors for soil erosion. Table 3
shows the effect of elevation on soil erosion based on AHP model. 4.4. Curvature
The AHP model pairwise comparison shows that slope and eleva-
tion significantly increase soil erosion process. The elevation map Curvature is the amount by which a curve deviates from its lin-
generated from DEM shows elevations of the area and their suscep- ear path. It influences the act of converging and diverging to
tibility to erosion. Chitral district is considered among the highest downward flow from the slope [9]. Fig. 3d shows the curvature
regions of the world, ranging from 1094 m at Arandu to 7726 m at of the study area, ranging from low (5.2) to high (6.5). In the area
Tirchmir packing over 40 peaks more than 6100 m in height, as with high curvature, water strikes the surface with high intensity,
shown in Fig. 3a. In DEM, the elevation is divided into five classes so the rate of erosion increases [35]. The study area includes low,
showing values of altitudes in different areas. Higher elevated medium and high curvatures which contribute to the rate of soil
areas contribute more to soil erosion [33]. erosion. It is evident from applied AHP technique in Table 3 which

Table 3
Pairwise comparison of dataset.

Pairwise Comparison Elevation Slope Aspect Curvature Landuse Drainage Rainfall Lineament Lithology NDWI NDVI
Elevation 1 0.33 3 3 0.33 3 0.33 1 3 0.33 3
Slope 3 1 5 1 3 5 1 3 5 1 5
Aspect 0.33 0.2 1 0.2 0.33 0.33 0.2 0.33 1 0.2 1
Curvature 0.33 1 5 1 1 3 1 1 5 0.33 1
Land use 3 3 0.33 3 1 1 0.2 1 5 0.33 1
Drainage 0.33 0.2 3 0.33 1 1 0.33 0.33 3 0.2 3
Rainfall 3 1 5 1 5 3 1 3 5 1 3
Lineament 1 0.33 3 1 1 3 0.33 1 3 0.33 3
Lithology 0.33 0.2 1 0.2 0.2 0.33 0.2 0.33 1 0.2 0.2
NDWI 3 1 5 1 3 5 1 3 5 1 5
NDVI 0.33 0.2 1 0.2 1 0.33 0.33 0.33 5 0.2 1
Sum 15.667 5.80 35 9.933 16.87 25 5.933 14.33 41 5.8 30.2
CR = 0.0705

5
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 3. Eleven parameters considered for the study area, (a) elevation, (b) slope, (c) aspect, (d) profile curvature, (e) land cover, (f) drainage density, (g) rainfall, (h) fault
density, (i) lithology, (j) NDWI, (k) NDVI.

shows that curvature when combined with Lithology, drainage or 4.5. Land cover
aspect significantly affects soil erosion. None of output AHP factors
were less than 1 which shows importance of this factor Land cover greatly affects the geological stability of the slope,
(curvature). which leads to the erosion process [10]. In terms of area size and
6
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 3 (continued)

pattern, the different land cover has different behavior toward ero- The pairwise comparison shows that drainage when combined
sion. Areas with smaller land cover have more risk of erosion [5]. with aspect, lithology or NDVI significantly affects soil erosion.
FAO-Landcover data is used in this study. That dataset is derived
from Landsat-8 OLI satellite data (from Earth Explorer at 30-m res-
4.7. Rainfall
olution) and field verification using a maximum likelihood classifi-
cation algorithm [31] in ArcGIS were used to produced land use
Rainfall transport and displaced soil from a specific area, which
land cover classification. Ranks are assigned according to the type
results in soil erosion. This is considered an important factor in soil
of each land use and its vulnerability to soil erosion. LC includes
erosion. Where the rate of rainfall is high, the rate of erosion also
agriculture in sloping valley, bare areas, shrubs, natural trees,
increases [5]. Type of rainfall, its duration, and extremeness in a
snow, ice and water bodies, as shown in Fig. 3e. There is a lot of
season and year affect erosion process [8,61]. A technique men-
portion, which is barren, so it greatly contributes to erosion. Land
tioned in Table 2 was used to extract the value of rainfall erosivity
use, when combined with lithology, highly affect soil erosion, as
index. Fig. 3g shows the rate of rainfall. Numerical value 583 shows
indicated in Table 3 with an output factor of 5. Furthermore, the
a greater rate of rainfall, while 1 shows light rainfall. In west and
combination of land use with elevation, slope, or curvature also
south of the Chitral district, rainfall rate is high, so it results in a
affects soil erosion as indicated by AHP technique.
greater rate of erosion. AHP pairwise comparison shows that rain-
fall highly affects soil erosion. Table 3 indicates the pairwise value
(5) for rainfall with aspect, land use or lithology. The lowest pair-
wise value of 1 was reported for combination of rainfall with other
4.6. Drainage
input factors.
The length of streams and channels in an area is an index to
explain soil erosion. The agreement is made that drainage in a 4.8. Lineaments
locality is considered as an index of soil erosion, although their
precise relationship has not been established [58,11]. Soil erosion Lineament is topographic features which show the characteris-
is caused by the critical value of drainage density per square kilo- tic of crust, which are responsible for a greater rate of infiltration.
metre by water is 0.9 km per square km of an area [5]. Pavelsky and This figure out fractured areas of the earth’s surface, contributing
Smith [44] suggested the drainage density method and similar to soil erosion. The study area includes both major and minor
method has been adopted for this study because several studies faults. Variation is observed at different distances and directions.
used this method which has a similar area. Drainage density higher Major significant structural faults are observed in the study area,
than this critical value increases surface runoff. In Fig. 3f, at a dis- including Ayun, Hispar, MKT, Reshun, ShahGhari and Trich Mir
tance of 6 km/sq km, there are more chances of erosion. In Chitral faults. Fault increases soil erosion process, as shown in Fig. 3h.
district, most of the portion has high and medium drainage in all The pairwise comparison using AHP method in Table 3 shows that
sides of the river, so the rate of erosion should also be higher. lineament along with aspect, drainage or lithology affects soil
7
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

erosion with pairwise factors of 3. Slope or rainfall along with lin- of the study area in Fig. 4 are also illustrated as 1, 2, 3, 4, 5 and
eament does not significantly affect soil erosion. 6, as mentioned in the hazard map.
Firstly, AHP method in integration with GIS is used to map the
4.9. Lithology areas susceptible to soil erosion. After this RUSLE model is used to
estimate the quantity of soil loss. Detailed methodology for RUSLE
Lithology shows the general physical characteristics of rocks. It modeling is taken from Maqsoom et al. [38] for the study area.
consists of mainly three rock types including igneous, sedimentary Fig. 5 shows the map of Chital district showing the loss of soil.
and metamorphic rocks. Each rock type has its specific behavior to Study area is divided into five different classes showing the inten-
soil erosion. Metamorphic rocks and limestone showed high infil- sity of soil erosion. Classes includes <50, 50–100, 100–150, 150–
tration rates, while marl areas have low infiltration rates because 200 and >200 tons/ha/year. Maximum erosion is observed in lower
they have high surface runoff [3]. In the study area of Chitral, area of Chitral. Bare areas and highlands with steep slopes are
details of following rocks are shown which contribute toward ero- more vulnerable to soil loss. Sediment yield is the quantity of soil
sion that includes granite, metamorphic, volcanic, dolomite and eroded from reservoir due to rainfall. Sediment yield is calculated
quartzite schist in Fig. 3i. Lithology does not significantly affect soil from the ‘‘weighted flow accumulation” tool in GIS to produce the
erosion, as shown in the AHP comparison in Table 3. total amounts of sediments that pass through each pixel. Highest
sediment yield is observed in sandy soil and steep slopes areas.
Fig. 6 shows the capacity of sediment yield in the study area. It is
4.10. NDWI classified into five different classes including <5, 5–10, 10–20,
20–50 and >50 in terms of tons/ha/year. Less than 5 tons/ha/year
Normalized difference water index (NDWI) is the quantity used is the lowest sediment yield and greater than 50 tons/ha/year is
to predict erosion rate; it predicts the depth of soil and wetness. It the maximum sediment yield observed in the research area.
is a proportion of erosion intensity of water flow, assuming that the
rate of flow is adequate to a specific area [24]. Areas with greater
water index have more influence on the process of soil erosion 5. Discussion
[1]. In our study area, south and west regions have a high value
of wetness index, so the rate of erosion is also high, as shown in Chitral River is prone to soil erosion for a decade. Chitral district
Fig. 3j. The pairwise comparison in Table 3 based on the AHP tech- consists of hilly areas, that is why the rate of erosion is high due to
nique shows that NDWI significantly affects soil erosion and is a many factors. Therefore, this studys access and explain two objec-
highly sensitive factor when combined with other input factors tives, profile of areas susceptible to erosion and rate of erosion by
affecting soil erosion. using ArcGIS. For sustainability of agricultural land use and regio-
nal and local development, it is necessary to find out areas vulner-
4.11. NDVI able to soil erosion. AHP sets criteria that give qualitative results in
the quantitative form [23]. According to their influence on soil ero-
The normalized difference vegetation index (NDVI) shows sion process, weights are assigned to all parameters including land
greenness and its health [4]. Bare areas have greater vulnerability cover, slope, elevation, aspect, drainage, rainfall, curvature, NDWI,
to erosion than forest green areas. Vegetation minimizes erosion NDVI, lineament, and lithology. Higher index values indicate
potential. Positive values (0.58) in NDVI show shrubs, grassland, higher prone areas, while low values indicate low-risk areas to soil
temperate and tropical rainforests while negative values (1) cor- loss. The prepared soil erosion susceptibility map shows a CR value
responds to water while values near zero (0.1 to 0.1) represent of 0.007 which is less than 0.1 [59]. This indicates that the assump-
barren areas and snow [46]. Most of the area of Chitral district tion made in all parameters against their contribution to soil ero-
includes rock and sand as shown in Fig. 3k. Table 3 shows that sion correct.
NDVI doesn’t have a significant effect on soil erosion. It’s highly Pradeep et al. [47] used AHP method in which it was found that
sensitive when combined with lithology with a pairwise factor of 5. more susceptibility to soil erosion is due to relative relief, slope and
land cover. The slope is the influencing factor for land degradation
4.12. Hazard due to the geomorphology of the Chitral region. Higher values of
drainage density show more susceptibility to soil erosion [26]. In
GIS-based proposed methods were used to prepare maps of rate this study, rainfall intensity is higher in lower areas of Chitral, that
and probability of erosion potential. The final map indicates the is why Das et al. [26] illustrated that areas with more rainfall have
soil erosion rate in the Chitral district. All selected factors which more chances of erosion. Many previous studies [33,46,47] showed
include elevation, slope, aspect, plane curvature, rainfall, linea- a similar geographical and climatic condition with the present
ments, lithology, NDWI, NDVI, land cover, and drainage were study area. The above comparison with previous studies shows
boarded up for mapping the area according to their susceptibility the concreteness of this study. It elaborates that the assumptions
to erosion. They are divided into five classes with percentages as for all factors are approximately correct because they all have
very high, high, medium, low, very low, as shown in Table 4. shown the same behavior about the influence of the soil erosion
In the areas of zhuil and totiraznokuispheri very low and low process. AHP method gives the ranking of each factor as for eleva-
susceptibility to erosion is observed while along the sides of the tion (0.091388), slope (0.165733), aspect (0.026046), curvature
path of river 21% erosion is seen in most of the areas in the study (0.119522), land use (0.077331), drainage (0.050972), rainfall
area [51]. River channel and some areas in Chitral are high and very (0.163220), lineaments (0.076627), lithology (0.022920), NDWI
highly susceptible to soil erosion, as shown in Fig. 4. Field images (0.165733) and NDVI (0.040552). The percentages of each factor
show the intensity of factors influence on erosion process.
This study gives a qualitative analysis of the trend of soil ero-
Table 4 sion. The area is divided into five classes based on their vulnerabil-
Classes according to susceptibility to erosion.
ity to soil loss. 28% of the area is under a very low susceptibility
13% 18% 21% 22% 28% zone which shows that this part of the area has no risk of soil ero-
Very high High Medium Low Very low sion. Areas under low, medium and high are 22%, 21%, and 18%
respectively. The lowermost portion of the Chitral district is prone
8
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 4. Soil erosion hazard map of the study area along with field images.

to very high soil erosion rate. A study by Maqsoom et al. [38], based It is observed in previous studies that only some factors are lar-
on the RUSLE method, in a similar study area, illustrated the same gely contributing to soil erosion rather than others, but for more
areas prone to high-risk erosion hazard. The study conducted by accuracy, eleven factors were considered in this study. Slope and
Yalew et al. [60] showed that 28.6% of the total study area is under elevation are the most important factors in soil erosion due to
very high susceptibility to soil erosion and in our study, 13% of the the hilly area. There are steep slopes and highly elevated regions
area is under the very high susceptible zone. This difference elab- which are more vulnerable to erosion. In a barren land, risks to ero-
orates that the Chitral district has fewer threats of soil loss than the sion are high while agricultural and vegetation-covered areas are
Abbay basin. Hence, the hazard map generated in the study gives less hit by the soil loss. An increase in rainfall in monsoon season
the profile of the Chitral district according to the rate of vulnerabil- increases the risk of soil loss because it increases the velocity of
ity to soil erosion. water which results in more soil erosion. Curvature and faults have

9
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 4 (continued)

Fig. 5. Spatially variable erosion rate (ton/hectare).

10
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Fig. 6. Sediment yield at each pixel.

a significant effect on erosion. Slope direction, drainage, moisture, soil erosion. High steep slopes on high elevations are the main
and lithology are also considered in this study and all the parame- cause of soil loss due to lack of knowledge about agricultural prac-
ter more or less significantly contributes to soil erosion. Some par- tices which comes in denudational slopes, side slope plateau, and
ticular class of factors may contribute more, but all are considered valleys present in this area. Curvature has a significant effect in
there for more accuracy in the results. AHP models gave highly combination with aspect, drainage density, and lithology on soil
accurate results of areas susceptible to erosion, and their results loss susceptibility. Areas with more rainfall and higher wetness
were well explained [12,14,27,29,41,45,51,56]. AHP soil modeling index values are more erosive. It is also found that barren lands
give the soil erosion probable zones and to quantify those zone have more risk of soil loss than agricultural areas. This factor
RUSLE model was used for the soil erosion quantification. Once increases the vulnerability of land degradation in the region. Qual-
the quantification is calculated then sediment yield is also calcu- itative results are generated in this study which would be helpful
lated to identify the potential for each place that how many sedi- to generate quantitative intensity and to control soil erosion
ments a specific point may produce. This will help in decision- process.
making for better land conversation and soil management. Similar
studies are also carried out in the Chital district showing the inten-
sity of soil loss [17,38]. It has been found that similar results are 6.1. Contribution/implications
generated which authenticate this study. The root purpose of this
study is to prepare a schematic diagram of Chitral River, that Results generated in this study are important tools for research-
how much erosion is susceptible in this area and to find out the ers and engineers to plan for control of soil erosion in the study
weight of each factor contributing to soil erosion. Proper remedies area. They should take active and meaningful decisions to avoid
should be taken to lower the impact of these influencing factors by this hazard in high and very high areas prone to soil erosion.
respective public agencies. Appropriate measures should be taken to avoid soil loss which
results in erosion of soil from the surface having nutrients in these
6. Conclusion agricultural areas and to maintain the design capacity of the dam.

In this study, the AHP method and GIS tools are integrated to
map the potential erosion watersheds, its spatial pattern, and the 6.2. Limitations and future studies
effect of different parameters (land use, lithology, lineament,
NDWI, NDVI, slope, elevation, aspect, curvature, and rainfall) in This can be done through other methods like fuzzy and TOPSIS
the Chitral district. Numerical weights are assigned to each param- methods for more perfection in results. After analyzing the results
eter according to the hierarchy of each factor. The results are gen- in AHP, to control soil erosion is a big challenge for researchers,
erated in the form of five different classes showing the severity of engineers, and respective agencies. For this exact quantity of ero-
erosion; very high (13%), high (18%), medium (21%), low (22%) and sion can also be calculated for risk assessment. Some structures
very low (28%). Due to the mountainous region, more or less all can be suggested in BIM to avoid soil erosion and which will also
considered parameters influence the soil erosion process. Slope, help to manage agricultural lands and the design capacity of the
elevation, moisture, and rainfall are major factors contributing to dam.
11
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

Declaration of Competing Interest [25] Dabral P, Baithuri N, Pandey A. Soil erosion assessment in a hilly catchment of
North Eastern India using USLE, GIS and remote sensing. Water Resour Manage
2008;22(12):1783–98.
The authors declare that they have no known competing finan- [26] Das B, Bordoloi R, Thungon LT, Paul A, Pandey PK, Mishra M, et al. An
cial interests or personal relationships that could have appeared integrated approach of GIS, RUSLE and AHP to model soil erosion in West
Kameng watershed, Arunachal Pradesh. J Earth Syst Sci 2020;129(1):1–18.
to influence the work reported in this paper.
[27] Dube F, Nhapi I, Murwira A, Gumindoga W, Goldin J, Mashauri D. Potential of
weight of evidence modelling for gully erosion hazard assessment in Mbire
Acknowledgment District-Zimbabwe. Phys Chem Earth, Parts A/B/C 2014;67:145–52.
[28] Ekumah B, Armah FA, Afrifa EK, Aheto DW, Odoi JO, Afitiri A-R. Geospatial
assessment of ecosystem health of coastal urban wetlands in Ghana. Ocean
The authors would like to appreciate the Shale Gas Research Coast Manag 2020;193. 105226.
Group (SGRG) in UTP and Shale PRF project (cost center # [29] Gayen A, Saha S, Pourghasemi HR. Soil erosion assessment using RUSLE model
and its validation by FR probability model. Geocarto Int 2019:1–19.
0153AB-A33) awarded to E. Padmanabhan for the support.
[30] Halefom A, Teshome A. Modelling and mapping of erosion potentiality
watersheds using AHP and GIS technique: a case study of Alamata Watershed,
References South Tigray, Ethiopia. Modeling Earth Syst Environ 2019;5(3):819–31.
[31] Hembram TK, Saha S. Prioritization of sub-watersheds for soil erosion based on
morphometric attributes using fuzzy AHP and compound factor in Jainti River
[1] Gómez-Gutiérrez Á et al. Using topographical attributes to evaluate gully
basin, Jharkhand, Eastern India. Environ Dev Sustain 2018:1–28.
erosion proneness (susceptibility) in two mediterranean basins: advantages
[32] Jebur MN, Pradhan B, Tehrany MS. Optimization of landslide conditioning
and limitations. Nat Hazards 2015;79(1):291–314.
factors using very high-resolution airborne laser scanning (LiDAR) data at
[2] Jaafari A et al. GIS-based frequency ratio and index of entropy models for
catchment scale. Remote Sens Environ 2014;152:150–65.
landslide susceptibility assessment in the Caspian forest, northern Iran. Int J
[33] Kayastha P, Dhital MR, De Smedt F. Application of the analytical hierarchy
Environ Sci Technol 2014;11(4):909–26.
process (AHP) for landslide susceptibility mapping: A case study from the
[3] Robledano-Aymerich F et al. Ecogeomorphological consequences of land
Tinau watershed, west Nepal. Comput Geosci 2013;52:398–408.
abandonment in semiarid Mediterranean areas: Integrated assessment of
[34] Khosrokhani M, Pradhan B. Spatio-temporal assessment of soil erosion at
physical evolution and biodiversity. Agriculture, Ecosyst Environ
Kuala Lumpur metropolitan city using remote sensing data and GIS.
2014;197:222–42.
Geomatics, Natl Hazards Risk 2014;5(3):252–70.
[4] Saha S. Geo-environmental evaluation for exploring potential soil erosion
[35] Lombardo L, Mai PM. Presenting logistic regression-based landslide
areas of jainti river basin using AHP model, Eastern India. Universal J Environ
susceptibility results. Eng Geol 2018;244:14–24.
Res & Technol 2018;7(1).
[36] Madureira L, Nunes LC, Borges JG, Falcão AO. Assessing forest management
[5] Saini SS et al. Vulnerability assessment of soil erosion using geospatial
strategies using a contingent valuation approach and advanced visualisation
techniques-A pilot study of upper catchment of Markanda river. Int J Adv
techniques: A Portuguese case study. J Forest Econ 2011;17(4):399–414.
Remote Sens, GIS and Geography 2015;2(1):9–21.
[37] Maity DK, Mandal S. Identification of groundwater potential zones of the
[6] Tahmassebipoor N et al. Spatial analysis of groundwater potential using
Kumari river basin, India: an RS & GIS based semi-quantitative approach.
weights-of-evidence and evidential belief function models and remote
Environ Dev Sustain 2019;21(2):1013–34.
sensing. Arabian J. Geosci. 2016;9(1):79.
[38] Maqsoom A, Aslam B, Hassan U, Kazmi ZA, Sodangi M, Tufail RF, et al.
[7] Vijith H et al. An assessment of soil erosion probability and erosion rate in a
Geospatial assessment of soil erosion intensity and sediment yield using the
tropical mountainous watershed using remote sensing and GIS. Arabian J.
revised universal soil loss equation (RUSLE) model. ISPRS Int J Geo-Inf 2020;9
Geosci. 2012;5(4):797–805.
(6):356.
[8] Wischmeier WH, Smith DD. Predicting rainfall erosion losses: a guide to
[39] Memarian H, Balasundram SK, Talib J, The Boon Sung C, Sood AM, Abbaspour
conservation planning. Department of Agriculture, Science and Education
KC, Haghizadeh A. Hydrologic analysis of a tropical watershed using
Administration; 1978.
KINEROS2. EnvironmentAsia 2012;5(1).
[9] Yilmaz C et al. GIS-based landslide susceptibility mapping using bivariate
[40] Mhazo N, Chivenge P, Chaplot V. Tillage impact on soil erosion by water:
statistical analysis in Devrek (Zonguldak-Turkey). Environ Earth Sci 2012;65
Discrepancies due to climate and soil characteristics. Agric Ecosyst Environ
(7):2161–78.
2016;230:231–41.
[10] Zakerinejad R, Maerker M. An integrated assessment of soil erosion dynamics
[41] Mihi A, Benarfa N, Arar A. Assessing and mapping water erosion-prone areas in
with special emphasis on gully erosion in the Mazayjan basin, southwestern
northeastern Algeria using analytic hierarchy process, USLE/RUSLE equation,
Iran. Nat. Hazards 2015;79(1):25–50.
GIS, and remote sensing. Appl Geomatics 2019:1–13.
[11] Zakrzewska B. Trends and methods in land form geography. Ann. Assoc. Am.
[42] Morgan RPC, Nearing M. Handbook of erosion modelling. John Wiley & Sons;
Geogr. 1967;57(1):128–65.
2016.
[12] Al-Rahbi AKH, Abushammala MF, Qazi WA. Application of the analytic
[43] Naqvi HR, Mallick J, Devi LM, Siddiqui MA. Multi-temporal annual soil loss risk
hierarchy process for management of soil erosion in Oman. Int J Analytic
mapping employing revised universal soil loss equation (RUSLE) model in Nun
Hierarchy Process 2020;12(1).
Nadi Watershed, Uttrakhand (India). Arabian J Geosci 2013;6(10):4045–56.
[13] Anderson JR. Land-use classification schemes. Photogrammetric Eng 1971.
[44] Pavelsky TM, Smith LC. RivWidth: A software tool for the calculation of river
[14] Arabameri A, Pradhan B, Rezaei K. Gully erosion zonation mapping using
widths from remotely sensed imagery. IEEE Geosci Remote Sens Lett 2008;5
integrated geographically weighted regression with certainty factor and
(1):70–3.
random forest models in GIS. J Environ Manage 2019;232:928–42.
[45] Pawluszek K, Borkowski A. Impact of DEM-derived factors and analytical
[15] Arekhi S, Niazi Y, Kalteh AM. Soil erosion and sediment yield modeling using
_
hierarchy process on landslide susceptibility mapping in the region of Roznów
RS and GIS techniques: a case study, Iran. Arabian J Geosci 2012;5(2):285–96.
Lake, Poland. Nat Hazards 2017;86(2):919–52.
[16] Arnoldus H. An approximation of the rainfall factor in the Universal Soil Loss
[46] Pourghasemi HR, Pradhan B, Gokceoglu C. Application of fuzzy logic and
Equation. An approximation of the rainfall factor in the Universal Soil Loss
analytical hierarchy process (AHP) to landslide susceptibility mapping at
Equation 1980:127–32.
Haraz watershed, Iran. Nat Hazards 2012;63(2):965–96.
[17] Aslam B, Maqsoom A, Kazmi ZA, Sodangi M, Anwar F, Bakri MH, et al. Effects of
[47] Pradeep G, Krishnan MN, Vijith H. Identification of critical soil erosion prone
landscape changes on soil erosion in the built environment: application of
areas and annual average soil loss in an upland agricultural watershed of
geospatial-based RUSLE technique. Sustainability 2020;12(15):5898.
Western Ghats, using analytical hierarchy process (AHP) and RUSLE
[18] Asteriou D, Hall SG. ARIMA models and the Box-Jenkins methodology. Appl
techniques. Arabian J Geosci 2015;8(6):3697–711.
Econometrics 2011;2(2):265–86.
[48] Prasannakumar V, Shiny R, Geetha N, Vijith H. Spatial prediction of soil erosion
[19] Bhunia GS, Samanta S, Pal B. Quantitative analysis of relief characteristics
risk by remote sensing, GIS and RUSLE approach: a case study of Siruvani river
using space technology. Int J Phys Social Sci 2012;2(8):350–65.
watershed in Attapady valley, Kerala, India. Environ Earth Sci 2011;64
[20] Biro K, Pradhan B, Buchroithner M, Makeschin F. Land use/land cover change
(4):965–72.
analysis and its impact on soil properties in the northern part of Gadarif
[49] Prasannakumar V, Vijith H, Geetha N, Shiny R. Regional scale erosion
region, Sudan. Land Degrad Dev 2013;24(1):90–102.
assessment of a sub-tropical highland segment in the Western Ghats of
[21] Camilo DC, Lombardo L, Mai PM, Dou J, Huser R. Handling high predictor
Kerala, South India. Water Resour Manage 2011;25(14):3715.
dimensionality in slope-unit-based landslide susceptibility models through
[50] Price G, Alam R, Hasan S, Humayun F, Kabir M, Karki C, et al. Attitudes to water
LASSO-penalized generalized linear model. Environ Modell Software
in South. Asia: Royal Institute of International Affairs; 2014.
2017;97:145–56.
[51] Rahmati O, Haghizadeh A, Pourghasemi HR, Noormohamadi F. Gully erosion
[22] Carlson TN, Ripley DA. On the relation between NDVI, fractional vegetation
susceptibility mapping: the role of GIS-based bivariate statistical models and
cover, and leaf area index. Remote Sens Environ 1997;62(3):241–52.
their comparison. Nat Hazards 2016;82(2):1231–58.
[23] Chowdary V, Chakraborthy D, Jeyaram A, Murthy YK, Sharma J, Dadhwal V.
[52] Saha S, Gayen A, Pourghasemi HR, Tiefenbacher JP. Identification of soil
Multi-criteria decision making approach for watershed prioritization using
erosion-susceptible areas using fuzzy logic and analytical hierarchy process
analytic hierarchy process technique and GIS. Water Resour Manage 2013;27
modeling in an agricultural watershed of Burdwan district, India. Environ
(10):3555–71.
Earth Sci 2019;78(23):649.
[24] Conforti M, Aucelli PP, Robustelli G, Scarciglia F. Geomorphology and GIS
[53] Sahaar, A.S., 2013. Erosion mapping and sediment yield of the Kabul River
analysis for mapping gully erosion susceptibility in the Turbolo stream
Basin, Afghanistan. Fort Collins, Colorado.
catchment (Northern Calabria, Italy). Nat Hazards 2011;56(3):881–98.

12
B. Aslam, A. Maqsoom, W. Salah Alaloul et al. Ain Shams Engineering Journal xxx (xxxx) xxx

[54] Spalevic V, Djurovic N, Mijovic S, Vukelic-Sutoska M, Curovic M. Soil erosion [60] Yalew SG, van Griensven A, Mul ML, van der Zaag P. Land suitability analysis
intensity and runoff on the Djuricka River Basin (North of Montenegro). for agriculture in the Abbay basin using remote sensing, GIS and AHP
Malaysian J Soil Sci 2013;17(1):49–68. techniques. Modeling Earth Syst Environ 2016;2(2):101.
[55] Spalević, V., Nyssen, J., Curovic, M., Lenaerts, T., Kerckhof, A., Annys, K., . . . [61] Yang Q, Xie Y, Li W, Jiang Z, Li H, Qin X. Assessing soil erosion risk in karst area
Frankl, A., 2013. The impact of land use on soil erosion in the River Basin using fuzzy modeling and method of the analytical hierarchy process. Environ
Boljanska Rijeka in Montenegro. Paper presented at the 4th International Earth Sci 2014;71(1):287–92.
symposium’Agrosym 2013’. [62] Zhang Y, Degroote J, Wolter C, Sugumaran R. Integration of modified universal
[56] Wang Q, Li W, Chen W, Bai H. GIS-based assessment of landslide susceptibility soil loss equation (MUSLE) into a GIS framework to assess soil erosion risk.
using certainty factor and index of entropy models for the Qianyang County of Land Degrad Dev 2009;20(1):84–91.
Baoji city, China. J Earth Syst Sci 2015;124(7):1399–415. [63] Tazioli A. Evaluation of erosion in equipped basins: preliminary results of a
[57] Wentworth CK. A simplified method of determining the average slope of land comparison between the Gavrilovic model and direct measurements of
surfaces. Am J Sci 1930(117):184–94. sediment transport. Environmental Geology 2009.
[58] Wu Q, Wang M. A framework for risk assessment on soil erosion by water [64] Horn BKP. Hill shading and the reflectance map. Proceedings of the IEEE 1981.
using an integrated and systematic approach. J Hydrol 2007;337(1–2):11–21. [65] Zevenbergen Lyle W, Thorne Colin R. Quantitative analysis of land surface
[59] Xi J-C, Kong Q-Q, Wang X-G. Spatial polarization of villages in tourist topography. Earth surface processes and landforms 1987;12(1):47–56.
destinations: A case study from Yesanpo, China. J Mountain Sci 2015;12 [66] Saaty RW. The analytic hierarchy process—what it is and how it is used.
(4):1038–50. Mathematical Modelling 1987;9(3–5):161–76.

13

View publication stats