MAPPING FLOOD VULNERABILITY BY APPLYING EBF AND
AHP METHODS, IN THE IRAQI MOUNTAIN REGION
Abdulrazzaq Q. Mikail
a, b*, Rahel Hamad a,c
a Scientific
Research Center, GIS & Remote Sensing Department, Delzyan
Campus, Soran University, Soran
44008, Iraq.
b Faculty
of Science, Delzyan Campus, Soran
University, Soran 44008, Iraq (abdulrazaq.mikail@soran.edu.iq)
c Faculty
of Science, Petroleum Geosciences Department, Delzyan
Campus, Soran University, Soran
44008, Iraq.
Received: 08 Oct., 2022 / Accepted: 09 Nov., 2022 / Published: 01
Jan., 2023 https://doi.org/10.25271/sjuoz.2022.10.4.1033
ABSTRACT
Flood hazards are a member of
the world's catastrophic events with a hydrological climate origin. They are
referred to as a situation in which the river flows and water level increases suddenly
and causes human and financial losses. This research aims to determine
flood-prone zones and evaluate the efficacy of RS and GIS-based evidence-based
belief function (EBF) and hierarchical analysis (AHP) models in flood-prone
area mapping. Using the Rezan River basin in the Mergasor area of Erbil governorate, Iraq, as an example, 11
factors such as slope, slope direction, land use, distance to the stream,
distance to the road, elevation, soil, rainfall, geology, NDVI, and drainage
density were utilized for flood moderation. The prediction rates of the EBF and
AHP models were also analyzed to be 0.869% and
0.836%, respectively, indicating that these two models are better predictors.
The findings of the study area revealed that 32% of the study area is under
very high to high flooding hazard zones for the EBF method and 22% for the AHP
method. This research’s conclusions are crucial for flood-prone region
management, decision-making, and local administrative planning.
KEYWORDS: Flood Vulnerability, Susceptibility,
Hazard, Rezan River, Mergasor.
Floods are one of nature's damaging natural
catastrophes (Tehrany &
Kumar, 2018). Accurately analyzing
their hazards is challenging owing to a lack of data and knowledge of flood
damage of various magnitudes (Grahn & Nyberg,
2017). Natural catastrophes inflict substantial economic
and human losses in human communities. They are caused by“natural disasters that provide a severe risk including flooding,
drought, earthquake, landslide, cyclones, and volcanos. These incidents do not
become natural disasters in regions without straight contact with persons or
impact mortal welfare (Fernández & Lutz, 2010).
A brief description
of flood events is that it is the rise of the level of water than the necessary
level generated by increased surface water in a stream succeeding severe rains,
covering the floodplain and surrounding grounds, damaging agricultural and
metropolitan areas, and resulting in fatalities (Chapi et al., 2017;
Huang et al., 2008). Maps
indicating the potential for floods are a valuable tool for determining the
future course of urban expansion and are often used to pinpoint these locations
(Büchele et al.,
2006). Maps of the flood zone and the hazard evaluations
for diverse regions comprise numerous criteria items that must be spatially
related (Booij, 2005). Floods are frequently caused by a combination of
severe meteorological, hydrological, and physical conditions (Chowdhuri et al.,
2020). One current and specific case of a flash flood
tragedy happened in Erbil, Iraq's Kurdistan region, destroying the city
financially and causing human casualties. Erbil city had a massive
floods in November and December 2021 as a consequence of severe monsoons
in a short time, unmethodical sewerage, and deluge rains,
killing 12 people, destroying over 1200 automobiles, and sinking more than 200
dwellings. Thus, flood vulnerability mapping is a critical step in flood relief
because it recognizes areas numerously vulnerable to floods and provides enough
time to prepare so that people can adapt to flooding predictably rather than
retroactively (Zhao et al., 2018) . Several authors investigated some approaches to
appreciate the possibilities of these solutions in light of the necessity for
an accurate and trustworthy technique to determine floodable places (Rahmati et al.,
2016).
Due to its ability
to handle enormous amounts of geographical data, the GIS has been proved as an
effective instrument for spatial analysis and information handling (Oh & Pradhan,
2011). Many investigators have extensively used a mix of
based on analytical models in conjunction with Geographic Information systems
and remote sensing (Tehrany &
Kumar, 2018). A range of arithmetical and probabilistic methods
have been experienced recently for creating flood vulnerability maps (Lee et al., 2012;
Levy et al., 2007). The evidential belief function (EBF) technique,
for instance, has been used to estimate potential groundwater zones (Althuwaynee et al.,
2014) and analyze the vulnerability of landslides (Tien Bui et al.,
2018), among other natural catastrophes. It is now
infrequently utilized for flood studies (Nampak et al.,
2014). In order to estimate
flood risk, Chen and Yeh (Chen et al., 2011) employed the AHP and GIS, producing valuable,
all-inclusive data for flood risk management in Taiwan. Additionally, other
researchers have studied and put into practice a variety of maps of flood
susceptibility strategies over the years. Tehrany and Kumar (2018) examined the potent EBF approach in the background
map of flood susceptibility. Additionally, when used with logistic regression
techniques, Tien Bui and Khosravi showed that the EBF
model had the most fantastic accuracy in forecasting flood vulnerability. On
the other hand, (Das, 2020) used the AHP approach to create maps of flood
susceptibility, sensitivity, and risk for the whole Western Ghat coastline area
in India by combining a sizable number of environmental flood conditioning
components from various sources.
As a result,
different locations need different aspects for flood mapping and risk
assessments (Poussin et al., 2014). As a result, multi-criteria decision analysis
(MCDA) methods have been successfully used in several research and are now
commonly regarded as a helpful tool for evaluating complex choice situations (Cinelli et al.,
2014; Hongoh et al., 2011). Additionally, MCDA considers various factors,
such as technical, environmental, and socioeconomic factors, to reach a
flawless conclusion (Mühlbacher &
Kaczynski, 2016). Furthermore, a GIS-based MCDA was utilized to map
flood regions (Dash & Sar,
2020), owing to the simplicity with which GIS methods
facilitate spatial data handling and analysis and the convenience with which
the MCDA result may be seen, interpreted, and evaluated (Hammami et al.,
2019).
Validation is
crucial when making maps of vulnerability to natural disasters for use in
planning (Khosravi et al.,
2016). In reality, a study of
flood vulnerability and verification of the resulting map for accessibility and
implementation mistakes has to be conducted. Furthermore, confirming the
prediction outcome is essential due to the significant interpretation that
researchers obtain from the prediction outcome (Shariat et al.,
2010).
The primary goals of
the study were to (1) assess the efficacy of EBF, and AHP approaches to
generate a map of flood susceptibility and (2) pinpoint the region's most
vulnerable to river flooding, a phenomenon that frequently causes significant
property and infrastructure damage, injuries, and fatalities in mountainous
regions. In order to do this, a number of GIS-based patterns
are used and assessed in the Mergasor district's Rezan river drainage area.
The research region is about 175 kilometres northeast of Erbil, Iraq,
between 36°41'6.53 “and
37° 2'7.13 N latitudes, 44°18'53.20 and 43°38'36.59 E longitudes. The drainage region covers
the Baze, Barzan, Rezan, Bele, and Mergasor districts bordered by Turkey. The “elevation” spans from 402 to 2292
meters higher than the Sea-level, and the climate is
described as having the Mediterranean climate with an average annual
precipitation of 1042 mm. The wettest months are November through April,
extending approximately from 250 to 400 mm. The rainy season, with its heavy
rainfall combined with topographical conditions, can cause flooding. The
moderate annual temperature in the basin varies from 22° to 26°, with the
lowest and highest temperatures meandering from 5° to 12° and 35° to 40°,
separately. The research area is 55 kilometres lengthy and 25 kilometres vast,
with a whole area of 1173 km2, and it has both flat and hilly topography
(Figure 1).
Figure 1. The investigation region location map.
Table 1.
Displays the multiple data sets that were applied to produce flood
vulnerability mapping Data used Data Source Resolution Data Types LULC and NDVI Sentinel 2A Multispectral Instrument (MSI) images (Accusation date: 16 July 2021 European Space Agency (ESA) earth online 10 m Grid / Polygon NDVI Sentinel 2A Multispectral Instrument (MSI) images (Accusation date: 16 July 2021 European Space Agency (ESA) earth online 10 m Grid / Polygon Basin boundary, DEM, Hill-shade, Drainage Density, Slope, Lineament Density distance to river and road Advanced Land Observing Satellite (ALOS) - Phased Array type L-band
Synthetic Aperture Radar (PALSAR) 12.5 m Grid / Polygon /point Geological map 26 geological units Geological survey (Baghdad) 1:1000000 Grid / Polygon Soil types Soil map Exploratory Soil Map of Iraq, Scale 1:1,000,00 1:1000000 Grid / Polygon Rainfall Annual rainfall data from 2000 to 2020 Erbil and Duhok Meteorological Office 12.5 m Grid / Polygon /point
Shafer first presented the Dempster–Shafer evidence theory, which
represents a mathematical model consisting of logical and belief-relevance
heuristics, which is perceived as spatial integration (Shafer, 1976). It is a reasonable or helpful concept or
method used to extract knowledge from data. The “Dempster-Schafer” method’s stability
is handling data that aren’t complete. There have shown that the conclusion of “belief, disbelief,
uncertainty,” and “plausibility” should be as
precisely specified as possible; in this way, a favourable outcome may be
achieved (Carranza, 2009). The EBF approach is the best modern flood
vulnerability assessment and mapping method (Tehrany & Kumar, 2018). Depending on the data source, an EBF
approach can be computed using the data-based or knowledge-based approach (Mondal & Mandal, 2020).
The data-driven EBF
model will be used in the flood vulnerability study and will be viewed as a
multivariate statistical investigation since it considers the geographical
relationship between each causative element and flood classes (Carranza et al., 2005). Because of its great accuracy and flexibility,
researchers are increasingly turning to the data-driven Dempster-Shafer EBF
method to generate a flood vulnerability map. The first step to using this
method is to convert it to data layers and then combine those layers, which,
after this method is completed, create a flood vulnerability map that can be
used to predict flooding (Park, 2011). The EBF method contains four fundamental purposes
with values ranging from 0 to 1: “Bel
(belief function)”,
“Dis (disbelief
function)”, Unci (uncertainty function)”, and “Pls (possibility
function) (Chowdhuri et al., 2020)”.
The pictorial representation of these combinations is depicted in Figure 2.
Figure 2:
Illustrates the links between the evidential belief functions.
The value of the Bel
method represents a “pessimistic” estimate, while the value of the Pls method
is represented an “optimistic” assessment of the spatial association during the
flood. However, it also discusses the factors that affect these results to illustrate
a complete understanding of this analysis (Awasthi & Chauhan, 2011; Pradhan &
Althuwaynee). Consequently, the
Bel value is either a lower or the same amount compared to Pls. The distinction
between the value of Pls and Bel methods is the Unci value, which shows that the theory is
founded on ignorance or unawareness of evidence. In contrast, the Dis value
refers to the idea that a theory is incorrect as a consequence of evidence (Awasthi & Chauhan, 2011; Azadi et al., 2020; Tien
Bui et al., 2019). Before
employing the model, collect flood conditioning variables and flood inventory
data (training dataset) to get the essential quantitative information. This
investigation used 33 validation data points, 77 training data points, and
eleven influencing factors. The following instructions provide a data-driven
estimate of "Bel,
Dis, Unci, and Pls": (Althuwaynee et al., 2012; Carranza & Hale, 2003).
where β, which ensures that Bel + Dis
+ Unci = 1, because this value must always be equal to 1.
The Analysis
Hierarchy Method (AHP)” technique
is a powerful and flexible multi-criteria decision-making method that can solve
complex problems at different levels. For this reason, it is called a hierarchy
model because it is entered in the form of a tree model and levels (Saaty, 1977). The AHP method combines both objective and
subjective evaluations in an integrated structure based on scales with paired
comparisons. It helps analysts to organize the essential aspects of a problem
in a hierarchical format (Thomas & Doherty, 1980). Using the AHP approach, users and planners
may quantify their preferred scale generated from various possibilities (Ayalew & Yamagishi, 2005). In this work, the weights of all input
items for flood risk mapping were determined using AHP. In terms of weight
importance, the pairwise comparison matrix was created initially with the help
of local experts. In order to create the "normalized matrix", each "pairwise comparison
matrix element was
divided by the total of each column. The moderate weight of each row was used
to generate the absolute weight value of the corresponding parameter. The
consistency ratio (CR) was determined to gauge the level of consistency between
the weight values of different factors in order to evaluate the validity and
applicability of the importance value calculation procedure, Equation (6). The “pairwise comparison
matrix” is
appropriate if “Consistency
Index (CI)” values
are comparable to (equal or smaller than) (0.1). However, the matrix should be
re-evaluated when the “Consistency
Index (CI)” is
greater than (0.1). Equation 7 is employed to obtain the “Consistency Index (CI).
Where
The final
Since the ROC curve is a
thorough, acceptable, and graphically displayed validation approach, it will be
employed to accurately evaluate the flood vulnerability map in this research.
Many authors have employed the "ROC curve" for precision
inspection and validation. The area beneath the curve is excellent when the AUC
value ranges between 0.9 and 1.0. Also, perfect when the value is between 0.8
and 0.9. The AUC value will be good when it is between 0.7 and 0.8 (Chowdhuri et al., 2020). Moreover, no globally acknowledged approach
exists for separating “inventory data into training and validation data (Yariyan et al.,
2020). As the methods for
organizing inventory data are inconsistent, there is no universally accepted
approach in natural hazard evaluations for describing the ratio of inventory
data used for training and validation. However, the general split ratio for
classifying inventory data in the literature concerning natural hazard
assessments is 70/30 (Li & Chen, 2019). The 70/30 ratio
will be implemented for splitting datasets in this investigation.
Since most floods are brought on by flooding the primary river trough,
distance from the river is one of the
essential elements in the flood sensitivity analysis. According to (Appendix,
Table A1), the 0-50 m class
in the present investigation had the greatest Bel of 0.495 and the lowest Dis
of 0.099. Nevertheless, the flood danger Bel values were 0.184, 0.196, 0.126,
and 0.000 for the other four classes. Additionally, it shows that lower
elevations, with a range of 402 to 800 meters, had bigger Bel values, with the
greatest Bel of 0.402 and the lowest Dis of 0.129, showing the highest flooding
vulnerability when water flows into and meets lower locations. The EBF data has
verified that most floods happen at lower elevations, which makes flooding at
higher altitudes very unlikely. Bel is highest, and dis is lowest, 0.343 and
0.150, indicating that the risk of flooding is greatest for slopes from 0° to
15°, followed by the range from 45° to 60°, and then the content from 30° to
45°. Nevertheless, the southwest in the aspect situation has the highest Bel
value of 0.244 and the smallest Dis value of 0.107, implying that this category
has positive spatial associations with floods. In contrast, the other Bel
values are significantly small, indicating a low likelihood of floods. This
indicates that floods are likely to occur since the earth rapidly gets
saturated due to heavy rain. Likewise, water and barren terrain are represented
by negative values and 0, grassland by 0.2-0.4 NDVI values, and forest by
values greater than 0.5 (Chowdhuri et al., 2020). The first rainfall class with 900–1000 mm
had the highest Bel and the lowest Dis values. As a result, vegetation catches
more precipitation. Less water is available to flow over the surface of the
ground.
Regarding geological factors, the river terraces had the greatest Bel
and Dis values of 0.219 and 0.052, correspondingly. As for soil factors, the
chestnut type includes the highest Bel value of 0.448 and Dis value of 0.192,
observed by the Lithosols-Chromic cambisols and Calcaric Regosols”. At the same time,
the Bel amount for the “Rough class was 0. Contrarily, land use is vital in
averting flooding that endanger human lives, homes, property, and ways of
making a living. So, depending on how land is used, the danger of flooding may
go down or up. The consequences of the present analysis showed that accounting
for 0.415 of the overall land use, the areas used for water sites in the region
had the highest Bel values. Similar to how they harm roadways, floods can
worsen dangerous flood situations. The most excellent Bel value 0.333, and the lower Dis valued 0.124, were produced by
separating the roadways, ranging from 0 to 25 m. The consequence was a greater
chance of flooding in the regions nearest to the highways. More extensive
basins usually get more precipitation than smaller basins, which causes higher
runoff. In areas with a larger drainage density (number of streams), rainfall
accumulation occurs more quickly, shortening the lag time. The highest Bel
value was 0.323,
and the lowest Dis value was 0.1450 for the
drainage density category. Figure 3 illustrates flood hazard maps for the
research area for the EBF method.
Figure 3. Flood
hazard map of the investigation region
employing the (a)
Bel (degree of belief); (b) Dis (degree of disbelief); (c) Unci (degree of uncertainty); “and (d)
Pls (degree of plausibility). |
After creating a pairwise comparison matrix, the AHP was utilized to
determine the comparative relevance of the relevant components. Weights were
assigned to each parameter once rated in order of importance. The comparative
importance rating hierarchy varies from 1 to 9, with lower scores meaning lower
priority and more elevated scores indicating higher priority. Table 2 displays
the pairwise comparison matrix as an 11 x 11 matrix with diagonal components
equal to 1. The comparative relevance of each row is calculated by comparing it
to each column. For instance, the slope obtains a 3 grade since it is much more
critical from the aspect. The row hast the reverse value off then pairwise comparisons (e.g., 1/3 ford aspect) since it
represents the importance of each element.
Table 2.
Pairwise comparison matrix by AHP Slope Aspect Elevation Rainfall D.Density D.River Land use INDVI D.Road Geology Soil Slope 1 3 2 1/2 1/2 1/2 1/3 1/3 3 1/3 1/4 Aspect 1/3 1 1/2 1/3 1/3 1/2 1/3 1/3 1/3 1/5 1/5 Elevation 1/2 2 1 1/3 1/2 1/3 1/2 1/2 2 1/5 1/4 Rainfall 2 3 3 1 3 2 1/2 1/3 3 1/2 1/2 D.Density 2 3 2 1/3 1 1/2 1/2 1/2 2 1/3 1/3 D.River 2 2 3 1/2 2 1 3 2 3 1 1/2 Land use 3 3 2 2 2 1/3 1 1/2 2 1/3 1/4 NDVI 3 3 2 3 2 1/2 2 1 4 2 1/2 D.Roadi 1/3 3 1/2 1/3 1/2 1/3 1/2 1/4 1 1/4 1/5 Geology 3 5 5 2 3 1 3 1/2 4 1 3 Soil 4 5 4 2 3 2 4 2 5 1/3 1 Table 3.
Normalized Pairwise comparison matrix by AHP for eleven factors weight Factors Slope Aspect Elevations Rainfalls D. Density D. River Land use NDVI D. Road Geology Soil Slope 1.00 3.00 2.00 0.50 0.50 0.50 0.33 0.33 3.00 0.33 0.25 Aspect 0.33 1.00 0.50 0.33 0.33 0.50 0.33 0.33 0.33 0.20 0.20 Elevation 0.50 2.00 1.00 0.33 0.50 0.33 0.50 0.50 2.00 0.20 0.25 Rainfall 2.00 3.00 3.00 1.00 3.00 2.00 0.50 0.33 3.00 0.50 0.50 D. Density 2.00 3.00 2.00 0.33 1.00 0.50 0.50 0.50 2.00 0.33 0.33 D.River 2.00 2.00 3.00 0.50 2.00 1.00 3.00 2.00 3.00 1.00 0.50 Land use 3.00 3.00 2.00 2.00 2.00 0.33 1.00 0.50 2.00 0.33 0.25 NDVI 3.00 3.00 2.00 3.00 2.00 0.50 2.00 1.00 4.00 2.00 0.50 D.Road 0.33 3.00 0.50 0.33 0.50 0.33 0.50 0.25 1.00 0.25 0.20 Geology 3.00 5.00 5.00 2.00 3.00 1.00 3.00 0.50 4.00 1.00 3.00 Soil 4.00 5.00 4.00 2.00 3.00 2.00 4.00 2.00 5.00 0.33 1.00
Flooding likelihood
and effect are also affected by the
hydrological action
of the soil. Flooding is less likely when the soil has a high-water
permeability. It is determined by the grain size of the clay and the diameter
of the pores in that clay. Thus, clay has poor permeability and a high-water
retention capacity.
The “AHP flood” hazard map
indicates that places cover 22.18% of the investigation region with a high or
very high risk of flooding. The ArcGIS 10.4 software's reclassification
function was used to pre-process all of the
requirements as raster datasets while outlining each need in detail. After
grade, established the authorities' belief in soil administration, Geological
composition, risk administration, and local presidency and experts, their weights were judged utilizing IAHP. Through multi-tests reasoning,
the weightages undeviating
consolidation technique of AHP created a last
flood vulnerability map, Figures4,
by calculating all raster’s
maps cautiously promoting the raster computer in the geographical
reasoning. TheiWGS84/UTM/Zonei38 Norths coordinates systems was employed in the
ArcGIS 10.4 software’s environments to study all the
significant sub criteria affecting flooding in the study area.
A final flood risk area map was created and classified into
five classes of flood risk
(very low, low, medium, high, and very high).
The study region's location and features were considered while selecting
the flood conditioning parameters. Flood vulnerability mapping was created
using EBF and AHP methods, which integrated the eleven causative factors:
elevation, soils type, drainages density, distanced tor the
driver, distanced tor the broad, NDVI,
rainfall, aspect, land use, slope, and geology. Weights were determined for
each category of flood conditioning factor. It was discovered that areas with
highest Bel values and lowest Dis values were the most susceptible to flooding
by the EBF approach and, at the same time, generated a map of flood
vulnerability map with the AHP approach to determine the high-risk zone of
flood in the investigation region.
The present work employed 77 flood and non-flood datasets for training
and 33 for validation to analyze the model's presentation and validate the flood vulnerability
map. The EBF model had the highest sensitivity value of the AUC (0.869) in the
training dataset, showing a high degree of classification flood pixel performance, followed by AHP (0.836). As a
result, the (AUC) values indicate the accuracy of the two methodologies in flood vulnerability
maps, as shown in Figure 5.
Figure 4. Flood vulnerability map employing AHP.
Figure 5: Receiver operating characteristic (ROC) curve
for EBF and AHP methods.
Flood vulnerability mapping was generated using EBF by incorporating
elevation, geology, soil type, slope, rainfall, distance to the river, distance
to the road, NDVI, land use, drainage density, and aspect. By assigning
relative importance to each category of flood conditioning factors, we found
that areas with high values for the belief function (Bel) and low values for
the disbelief function (Dis) were particularly at risk of flooding.
Furthermore, the EBF model's performance was evaluated using the AUC criterion,
which indicated
that booths training and validating points
could map flood vulnerability with an AUCI of 0.763i for achievement band i0.869 for forecast Frates separately.
Our investigation revealed that distanced from the driver was a
significant factor in determining vulnerability to flooding. Because of its
proximity to the main channel and rapid flood response, the area between 0 and
50 meters in altitude is particularly vulnerable to flooding on the Rezan River. Flooding is more common in riverside
communities because of the river's closeness and the fast reaction to flooding,
as confirmed by (Chapi et al., 2017; Pham et al., 2020).
Bel values were high (0.448) for both soil type and distance from the
river. Several soil properties can affect the volume of runoff in a catchment
area (Tehrany et al., 2019). So, the amount of water drains from the
ground and causes floods depends on the soil's texture (Fontanine & Costache, 2013). Based on the results of the present
investigation, the "Chestnut" soil class was
found (Fontanine & Costache, 2013) to be particularly vulnerable to floods.
With a Bel value of 0.297, this soil type suggests it limits precipitation
infiltration, leading to more runoff than Lithosols, Rendzinas, and Chromic Cambisols.
For the NDVI, the highest Bel value of 0.426 was found between and (-0.5
to -0.07). According to the results, the likelihood of flooding was the highest
in areas with the fewest number of plants. Given their more excellent Bel value
and more robust association with floods, these areas were more at risk of
flooding than other classes, which is in line with findings from (Chowdhuri et al., 2020).
With a Bel value of 0.415 for the land use component (one of the human disturbance variables), waters bodies showed the
uppermost chance of flood occurrence. with such an order, Bel values of 0.228,
0.158, 0.143, and 0.057 were recorded for farmland, cities, barren land, and forests.
Studies have shown time and time again that changes in land use and land cover
have direct or indirect effects on hydrologically functions like penetrability (Ossola et al., 2015), evapotranspiration’s (Wiles & Sharp Jr, 2008), and runoffs (Arabameri et al., 2019).
Following the distance of the river, soil type, land use, NDVI, and
elevation was the most crucial factor in determining the likelihood of
flooding. The category 402 to 800 m elevation range already had the highest
value of Bel (0.402) and the smallest value of Dis (0.129) of all the classes,
suggesting a greater than average flood risk. In contrast, locations above 2000
m had the smallest Bel value (0.0).
Theses outcomes
are in accord with an investigation by (Chowdhuri et al., 2020) in the catchment of eastern India, (Arabameri et al., 2019) they mapped the vulnerability to water risks
for northern Iran, and the (Al-Hinai & Abdalla, 2021) investigation in Muscat Governorates, Oman, which determined
that only locations at a lower elevation were affected by floods.
Regarding geology, the stream terraces layer had the highest Bel value,
0.219. This was preceded by the
“Bai Hassan (Upper Bakhtiari)” section, which produced i0.171, and the
sloped sediments
formations, which
had 0.127. On the other hand, the scores for the different classes show that
flooding is not likely to occur because they are lower in Bel and higher in
Dis. Slope deposition layers and the river terrace of the Bai Hassan are
primarily composed of clastic sedimentary rocks, including conglomerates,
sandstone, and claystone, and rock pieces with fine clastic. Generally, the
area's geology plays a crucial role in flood inundation mapping due to the
involvement of multiple geological formations in hydrological processes.
According to (Regmi & Poudel, 2016), they can significantly alter the
conductivity and penetrability of water flow.
The soil rapidly gets saturated when rain falls in large amounts, leading
to flooding. The rain maps gained weight from EBF verified this. The heaviest
rainfall category (900-1000 mm) was assigned: the uppermost Bel value (0.283) and the
lowermost Dis value (0.181), indicating that
this category experienced the most favourable conditions. So, the more plants
there are, the more rain is collected, while less water is left to run off.
According to EBF's slope-flood study, the slope percentage class 00-150
produced the uppermost Bel value
(0.343) and the lowermost Dis value (0.150). Because of the inverse link between slope
and Bel value, the locations with the smallest slope earned the greatest Bel
value and the smallest Dis value. Devkota et al., (2013) have drawn connections between land usage,
slope, and rainfall depending on their observations in the fields. They
explained that land use significantly affected stability of the slope. Forested
areas control the flow of water and allow water to seep into the soil at
periodic intervals, while agricultural land might compromise stability of the
slope owing to much saturated soil.
When considering another human perturbation factor, distance to the
road, it is expected that the Bel class has a bigger influence on flooding
occurrences the closer the location is near the road. In addition, in regards
of the map of distance from the road, the 0-25 m class offered the uppermost
Bel value of 0.333
and the lowermost Dis value of 0.124. Furthermore, the impermeable barriers
that prevent runoff from percolating into the earth increase the frequency and
severity of floods. As a result, roadways in the area under investigation may
be harmed, and hazardous flood conditions may arise during a major storm.
The 0.86-1.43 density group was related to the drainage density values,
with the greatest Bel value
0.323 band them smallest value of Dis being i0.145. According to
the results, this possibility has the best increase in the spring season. (Pourghasemi & Beheshtirad, 2015) confirmed this by noting that the incidence
of springs increased with drainage intensity, hence establishing a direct
relationship between drainage density and groundwater spring possibility maps.
Since rainfall comes sooner in the basin with greater drainage density (number
of tributary rivers), leading to a shorter lag time, larger basins receive more
rain on the median, resulting in much more discharge.
By contrast, the southeast had the lowest value of 0.107 for the aspect
element, and the northwest obtained the greatest value of 0.244. This factor,
which also affects the frequency of floods, is vital for wet retention and
plant densities. Despite being connected to physiographic elements that,
according to (Rahmati et al., 2016), may affect hydrological conditions and soil
moisture regimes, the aspect component in the present research had a minor
effect on floods. In conclusion, the EBF method found that distance from the river is the essential
primary variable contributing equally to flood incidence, then a layer of soil,
NDVI, elevation, and land use. But rainfall, aspect, and geology have the most
negligible impact on flood danger, followed by slope and distance from the
road.
The hierarchical analysis process (AHP) model has been used in this
research to evaluate the determined criteria. By combining GIS facilities and
various data, this model is considered a powerful tool in the micro zoning of
environmental risks. It weights the criteria based on their importance and
impact on creating flood risk by comparing pairs between the requirements. For
this purpose, the determining criteria were prepared in pairs and
hierarchically in the form of a questionnaire. Finally, after calculating the
average of the questionnaires, the data was entered into the special software
of the AHP model (Experts choices software), and the
relative importance of each criterion was calculated.
According to Tehrany and Kumar (2018) Table 4. Estimated and actual flooded
areas using EBF and AHP approaches. EBF method AHP method Area km2 Ratio of area (%) Area km2 Ratio of area (%) Very
low 211.14 18.00 98.88 8.43 Low 332.89 28.38 335.71 28.62 Moderate 253.48 21.61 478.11 40.76 High 241.98 20.63 196.59 16.76 Very
high 133.48 11.38 63.57 5.42
Finally, flood
vulnerability mapping was created using the EBF and AHP methods. The risk of
flooding has been mapped, and the findings are broken into five categories. The
ranges covered here are very low, low, moderate, high, and very thigh. Table 4 shows
that the EBF approach covered a total area of i211.14 (18.00%), i332.89 (28.38%), 253.48 (21.61%), i241.98 (20.63%), and
133.48 (11.38%) square kilometres, respectively, whereas the AHP approach
covered a total of 98.88 (8.43%), 335.71 (28.62%), 478.11 (40.76%),196.59
(16.76%), and 63.57 (5.42%) square kilometres, respectively. Accordingly, the
research region faces a significant probability of flooding, with over 31% for
the EBF model and over 21% for the AHP model. There were high flood susceptibility
zones from the junction of the two rivers down to the Rezan
River's mouth. The transition between the high and highly high zones constituted
the moderate zone. On the other hand, most of the research area is located in
areas with a low chance of flooding.
Maps of flood vulnerability might be valuable tools for decision-makers
trying to lessen the impact of floods. Here's a rundown of the most important
findings:
1.
The maps demonstrated that both flood vulnerability models were suitable
for making flood hazard classifications. Nevertheless, the EBF model was more
effective than the AHP one. Areas with a highest Bel and a lower Dis for flood events
are particularly vulnerable, as shown by an examination of flood vulnerability
maps.
2.
Flood risk may be predicted using slope direction, slope percentage,
elevation, soil, proximity to rivers and roads, river density, geology,
rainfall, NDVI, and land use. With much of the area under study situated on a
mountain plain and steep mountain on all sides, the rapid formation of runoff
results in floods in low-slope regions.
This study divided the distance from the river into five categories; the
first category (0-25 m) was more vulnerable to flooding, while the chance of
flooding decreased with further distance from the river. Also, studies found
that the first class, with the lowest elevation, is more likely to experience
floods than the other two. Conversely, the first class of NDVI variables
includes the increased chance of floods caused by vegetation's absence and
vegetation degradation owing to uncontrolled livestock grazing in various
portions of the study region.
1. AHP: This model uses precise characteristics for judgements. i.e., in
practical situations, human emotions are murky, and the leaders may not be able
to connect the careful numerical attributes to the examination assessments. AHP
is not significant in this case. For creating pairwise correlations, the AHP
can only accept free criteria. Because nature is inherently contradictory and
decision-making is based only on the situation at hand and the leader's
intuition, the AHP cannot take uncertainties and threats into account while a
chief is making a decision.
When two criteria or options are examined pairwise,
input data are obtained. In any case, the excessive repetition in the
correlations is regarded as the reason the pairwise analysis is flawed. Due to
the lack of information on the criteria and options and the lack of focus
during pairing testing and speaking, AHP allows for irregularity.
2. EBF: Lack of required resources specific to the EBF model in flood bora. Most
of the studies that have used the model have been conducted on soil erosion.
3.
Difficulties in obtaining points in different parts of the study area
with GPS due to the high altitudes and difficulties of much of the area to get
the necessary data for the study.
Conflicts of Interest: The authors announce no struggle of attention.
Funding: There was no external support for this research.
Author contributions: All of the authors
made substantial contributions. They have seen and approved the final draft.
Althuwaynee, O. F., Pradhan, B.,
& Lee, S. (2012). Application of an evidential belief function model in
landslide susceptibility mapping. Computers
& Geosciences, 44, 120-135.
Althuwaynee, O. F., Pradhan, B.,
Park, H.-J., & Lee, J. H. (2014). A novel ensemble bivariate statistical
evidential belief function with knowledge-based analytical hierarchy process
and multivariate statistical logistic regression for landslide susceptibility
mapping. Catena, 114, 21-36.
Arabameri, A., Rezaei, K., Cerdà,
A., Conoscenti, C., & Kalantari, Z. (2019). A comparison of statistical
methods and multi-criteria decision making to map flood hazard susceptibility
in Northern Iran. Science of the Total
Environment, 660, 443-458.
Awasthi, A., & Chauhan, S. S.
(2011). Using AHP and Dempster–Shafer theory for evaluating sustainable
transport solutions. Environmental
Modelling & Software, 26(6),
787-796.
Ayalew, L., & Yamagishi, H.
(2005). The application of GIS-based logistic regression for landslide
susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan. Geomorphology, 65(1-2), 15-31.
Azadi, F., Sadough, S. H.,
Ghahroudi, M., & Shahabi, H. (2020). Zoning of Flood Risk in Kashkan River
basin using Two Models WOE and EBF. Journal
of Geography and Environmental Hazards,
9(1), 45-60.
Booij, M. J. (2005). Impact of climate
change on river flooding assessed with different spatial model resolutions. Journal of hydrology, 303(1-4), 176-198.
Büchele, B., Kreibich, H., Kron, A.,
Thieken, A., Ihringer, J., Oberle, P., Merz, B., & Nestmann, F. (2006).
Flood-risk mapping: contributions towards an enhanced assessment of extreme
events and associated risks. Natural
Hazards and Earth System Sciences, 6(4),
485-503.
Carranza, E., Woldai, T., &
Chikambwe, E. (2005). Application of data-driven evidential belief functions to
prospectivity mapping for aquamarine-bearing pegmatites, Lundazi district,
Zambia. Natural Resources Research, 14(1), 47-63.
Carranza, E. J. M. (2009). Controls
on mineral deposit occurrence inferred from analysis of their spatial pattern
and spatial association with geological features. Ore Geology Reviews, 35(3-4),
383-400.
Carranza, E. J. M., & Hale, M.
(2003). Evidential belief functions for data-driven geologically constrained
mapping of gold potential, Baguio district, Philippines. Ore Geology Reviews, 22(1-2),
117-132.
Chapi, K., Singh, V. P., Shirzadi,
A., Shahabi, H., Bui, D. T., Pham, B. T., & Khosravi, K. (2017). A novel hybrid
artificial intelligence approach for flood susceptibility assessment. Environmental modelling & software, 95, 229-245.
Chen, Y.-R., Yeh, C.-H., & Yu,
B. (2011). Integrated application of the analytic hierarchy process and the
geographic information system for flood risk assessment and flood plain
management in Taiwan. Natural hazards, 59(3), 1261-1276.
Chowdhuri, I., Pal, S. C., &
Chakrabortty, R. (2020). Flood susceptibility mapping by ensemble evidential
belief function and binomial logistic regression model on river basin of
eastern India. Advances in Space Research, 65(5), 1466-1489.
Cinelli, M., Coles, S. R., &
Kirwan, K. (2014). Analysis of the potentials of multi criteria decision
analysis methods to conduct sustainability assessment. Ecological indicators, 46,
138-148.
Das, S. (2020). Flood susceptibility
mapping of the Western Ghat coastal belt using multi-source geospatial data and
analytical hierarchy process (AHP). Remote
Sensing Applications: Society and Environment, 20, 100379.
Dash, P., & Sar, J. (2020).
Identification and validation of potential flood hazard area using GIS‐based
multi‐criteria analysis and satellite data‐derived water index. Journal of Flood Risk Management, 13(3), e12620.
Devkota, K. C., Regmi, A. D.,
Pourghasemi, H. R., Yoshida, K., Pradhan, B., Ryu, I. C., Dhital, M. R., &
Althuwaynee, O. F. (2013). Landslide susceptibility mapping using certainty
factor, index of entropy and logistic regression models in GIS and their
comparison at Mugling–Narayanghat road section in Nepal Himalaya. Natural hazards, 65(1), 135-165.
Fernández, D., & Lutz, M. A.
(2010). Urban flood hazard zoning in Tucumán Province, Argentina, using GIS and
multicriteria decision analysis. Engineering
Geology, 111(1-4), 90-98.
Fontanine, I., & Costache, R.
(2013). Using GIS techniques for surface runoff potential analysis in the
Subcarpathian area between Buzãu and Slãnic rivers, in Romania. Cinq Continents, 3(7), 47-57.
Grahn, T., & Nyberg, L. (2017).
Assessment of pluvial flood exposure and vulnerability of residential areas. International Journal of Disaster Risk
Reduction, 21, 367-375.
Hammami, S., Zouhri, L., Souissi,
D., Souei, A., Zghibi, A., Marzougui, A., & Dlala, M. (2019). Application
of the GIS based multi-criteria decision analysis and analytical hierarchy
process (AHP) in the flood susceptibility mapping (Tunisia). Arabian Journal of Geosciences, 12(21), 1-16.
Hongoh, V., Hoen, A. G.,
Aenishaenslin, C., Waaub, J.-P., Bélanger, D., & Michel, P. (2011).
Spatially explicit multi-criteria decision analysis for managing vector-borne
diseases. International Journal of Health
Geographics, 10(1), 1-9.
Huang, X., Tan, H., Zhou, J., Yang,
T., Benjamin, A., Wen, S. W., Li, S., Liu, A., Li, X., & Fen, S. (2008).
Flood hazard in Hunan province of China: an economic loss analysis. Natural Hazards, 47(1), 65-73.
Khosravi, K., Nohani, E.,
Maroufinia, E., & Pourghasemi, H. R. (2016). A GIS-based flood
susceptibility assessment and its mapping in Iran: a comparison between
frequency ratio and weights-of-evidence bivariate statistical models with
multi-criteria decision-making technique. Natural
Hazards, 83(2), 947-987.
Lee, J.-m., Hyun, K.-h., Choi,
J.-s., Yoon, Y.-j., & Geronimo, F. K. F. (2012). Flood reduction analysis
on watershed of LID design demonstration district using SWMM5. Desalination and Water Treatment, 38(1-3), 255-261.
Levy, J. K., Hartmann, J., Li, K.
W., An, Y., & Asgary, A. (2007). Multi‐criteria decision support systems
for flood hazard mitigation and emergency response in urban watersheds 1. JAWRA Journal of the American Water
Resources Association, 43(2),
346-358.
Li, Y., & Chen, W. (2019).
Landslide susceptibility evaluation using hybrid integration of evidential
belief function and machine learning techniques. Water, 12(1), 113.
Mondal, S., & Mandal, S. (2020).
Data-driven evidential belief function (EBF) model in exploring landslide
susceptibility zones for the Darjeeling Himalaya, India. Geocarto International, 35(8),
818-856.
Mühlbacher, A. C., & Kaczynski,
A. (2016). Making good decisions in healthcare with multi-criteria decision
analysis: the use, current research and future development of MCDA. Applied health economics and health policy, 14(1), 29-40.
Nampak, H., Pradhan, B., & Abd
Manap, M. (2014). Application of GIS based data driven evidential belief
function model to predict groundwater potential zonation. Journal of Hydrology, 513,
283-300.
Oh, H.-J., & Pradhan, B. (2011).
Application of a neuro-fuzzy model to landslide-susceptibility mapping for
shallow landslides in a tropical hilly area. Computers & Geosciences,
37(9), 1264-1276.
Ossola, A., Hahs, A. K., &
Livesley, S. J. (2015). Habitat complexity influences fine scale hydrological
processes and the incidence of stormwater runoff in managed urban ecosystems. Journal of Environmental Management, 159, 1-10.
Park, N.-W. (2011). Application of
Dempster-Shafer theory of evidence to GIS-based landslide susceptibility
analysis. Environmental earth sciences, 62(2), 367-376.
Pham, B. T., Avand, M., Janizadeh,
S., Phong, T. V., Al-Ansari, N., Ho, L. S., Das, S., Le, H. V., Amini, A.,
& Bozchaloei, S. K. (2020). GIS based hybrid computational approaches for
flash flood susceptibility assessment. Water, 12(3), 683.
Pourghasemi, H. R., &
Beheshtirad, M. (2015). Assessment of a data-driven evidential belief function
model and GIS for groundwater potential mapping in the Koohrang Watershed,
Iran. Geocarto International, 30(6), 662-685.
Poussin, J. K., Botzen, W. W., &
Aerts, J. C. (2014). Factors of influence on flood damage mitigation behaviour
by households. Environmental Science
& Policy, 40, 69-77.
Pradhan, B., & Althuwaynee, O.
Ensemble of data-driven EBF model with knowledge based AHP model for slope
failure assessment in GIS using cluster pattern.
Rahmati, O., Pourghasemi, H. R.,
& Zeinivand, H. (2016). Flood susceptibility mapping using frequency ratio
and weights-of-evidence models in the Golastan Province, Iran. Geocarto International, 31(1), 42-70.
Regmi, A. D., & Poudel, K.
(2016). Assessment of landslide susceptibility using GIS-based evidential
belief function in Patu Khola watershed, Dang, Nepal. Environmental Earth Sciences,
75(9), 1-20.
Saaty, T. L. (1977). A scaling
method for priorities in hierarchical structures. Journal of mathematical psychology, 15(3), 234-281.
Shafer, G. (1976). A mathematical theory of evidence (Vol.
42). Princeton university press.
Shariat, S. F., Lotan, Y., Vickers,
A., Karakiewicz, P. I., Schmitz-Dräger, B. J., Goebell, P. J., & Malats, N.
(2010). Statistical consideration for clinical biomarker research in bladder
cancer. Urologic Oncology: Seminars and Original Investigations,
Tehrany, M. S., & Kumar, L.
(2018). The application of a Dempster–Shafer-based evidential belief function
in flood susceptibility mapping and comparison with frequency ratio and
logistic regression methods. Environmental
Earth Sciences, 77(13), 1-24.
Tehrany, M. S., Kumar, L., &
Shabani, F. (2019). A novel GIS-based ensemble technique for flood
susceptibility mapping using evidential belief function and support vector
machine: Brisbane, Australia. PeerJ, 7, e7653.
Thomas, P. G., & Doherty, P. C.
(1980). The Analytic Hierarchy. Process: Planning, Priority Setting, Resource
Allocation, McGraw-Hill,
Tien Bui, D., Khosravi, K., Shahabi,
H., Daggupati, P., Adamowski, J. F., Melesse, A. M., Thai Pham, B.,
Pourghasemi, H. R., Mahmoudi, M., & Bahrami, S. (2019). Flood spatial
modeling in northern Iran using remote sensing and gis: A comparison between
evidential belief functions and its ensemble with a multivariate logistic
regression model. Remote Sensing, 11(13), 1589.
Tien Bui, D., Shahabi, H., Shirzadi,
A., Chapi, K., Alizadeh, M., Chen, W., Mohammadi, A., Ahmad, B. B., Panahi, M.,
& Hong, H. (2018). Landslide detection and susceptibility mapping by airsar
data using support vector machine and index of entropy models in cameron
highlands, malaysia. Remote Sensing, 10(10), 1527.
Wiles, T. J., & Sharp Jr, J. M.
(2008). The secondary permeability of impervious cover. Environmental & engineering geoscience, 14(4), 251-265.
Yariyan, P., Avand, M., Abbaspour,
R. A., Torabi Haghighi, A., Costache, R., Ghorbanzadeh, O., Janizadeh, S.,
& Blaschke, T. (2020). Flood susceptibility mapping using an improved
analytic network process with statistical models. Geomatics, Natural Hazards and Risk, 11(1), 2282-2314.
Appendix
Table A1. The
regions, amount of flood locations, and variable value classes in the
EBF approach
|
Layer |
Classes |
Area (%) |
Bel |
Dis |
Unc |
Pls |
|
Elevation (Appendix, Figure C1) |
402 - 800 |
39% |
0.402 |
0.129 |
0.469 |
0.871 |
|
800 - 1300 |
37% |
0.148 |
0.251 |
0.601 |
0.749 |
|
|
1300 - 1600 |
19% |
0.171 |
0.217 |
0.613 |
0.783 |
|
|
1600 - 2000 |
5% |
0.279 |
0.200 |
0.520 |
0.800 |
|
|
>2000 |
1% |
0.000 |
0.203 |
0.797 |
0.797 |
|
|
Slope (Appendix, Figure C2) |
0º – 15º |
38% |
0.343 |
0.150 |
0.507 |
0.850 |
|
15º – 30º |
38% |
0.160 |
0.242 |
0.598 |
0.758 |
|
|
30º – 45º |
20% |
0.198 |
0.209 |
0.593 |
0.791 |
|
|
45º – 60º |
4% |
0.299 |
0.198 |
0.503 |
0.802 |
|
|
60º – 80º |
0% |
0.000 |
0.201 |
0.799 |
0.799 |
|
|
Aspect (Appendix, Figure C3) |
North |
13% |
0.052 |
0.136 |
0.812 |
0.864 |
|
Northeast |
12% |
0.155 |
0.121 |
0.724 |
0.879 |
|
|
East |
10% |
0.159 |
0.121 |
0.719 |
0.879 |
|
|
Southeast |
13% |
0.244 |
0.107 |
0.648 |
0.893 |
|
|
South |
17% |
0.132 |
0.123 |
0.744 |
0.877 |
|
|
Southwest |
15% |
0.090 |
0.131 |
0.779 |
0.869 |
|
|
West |
10% |
0.088 |
0.129 |
0.783 |
0.871 |
|
|
Northwest |
11% |
0.080 |
0.131 |
0.790 |
0.869 |
|
|
Rainfall (Appendix, Figure C4) |
700 - 800 |
23% |
0.103 |
0.226 |
0.671 |
0.774 |
|
800 - 900 |
32% |
0.168 |
0.208 |
0.623 |
0.792 |
|
|
900 - 1000 |
15% |
0.283 |
0.181 |
0.536 |
0.819 |
|
|
1000 – 1100 |
13% |
0.229 |
0.192 |
0.579 |
0.808 |
|
|
1100 - 1300 |
17% |
0.216 |
0.193 |
0.591 |
0.807 |
|
|
NDVI (Appendix, Figure C5) |
-0.5 to -0.07 |
22% |
0.426 |
0.147 |
0.427 |
0.853 |
|
-0.07 – 0.16 |
29% |
0.226 |
0.198 |
0.576 |
0.802 |
|
|
0.16 – 0.25 |
25% |
0.142 |
0.224 |
0.634 |
0.776 |
|
|
0.25 – 0.37 |
17% |
0.094 |
0.223 |
0.683 |
0.777 |
|
|
0.37 – 0.77 |
7% |
0.112 |
0.208 |
0.680 |
0.792 |
|
|
Geology (Appendix, Figure C6) |
Gercus Formation |
3% |
0.023 |
0.053 |
0.923 |
0.947 |
|
Pilaspi Formation |
4% |
0.020 |
0.054 |
0.927 |
0.946 |
|
|
Sehkaniyan and Sarki Formations |
1% |
0.000 |
0.053 |
0.947 |
0.947 |
|
|
Chia Gara, Barsarin, Naokelekan and Sargelu
Formation |
3% |
0.028 |
0.053 |
0.919 |
0.947 |
|
|
Balambo, Garagu and Sarmord Formation |
5% |
0.033 |
0.053 |
0.914 |
0.947 |
|
|
Qamchuqa Formation |
19% |
0.052 |
0.050 |
0.898 |
0.950 |
|
|
Tanjero Formation |
1% |
0.052 |
0.052 |
0.895 |
0.948 |
|
|
Aqra-Bekhme Formation |
34% |
0.032 |
0.060 |
0.908 |
0.940 |
|
|
Shiranish Formation |
8% |
0.092 |
0.048 |
0.860 |
0.952 |
|
|
Fatha (Lower Fars)
Formation |
4% |
0.000 |
0.055 |
0.945 |
0.945 |
|
|
Mukdadiyah (Lower Bakhtiari)
Formation |
3% |
0.000 |
0.054 |
0.946 |
0.946 |
|
|
Injana (Upper Fars)
Formation |
4% |
0.019 |
0.054 |
0.927 |
0.946 |
|
|
Slope deposits |
3% |
0.127 |
0.049 |
0.824 |
0.951 |
|
|
Kolosh Formation |
1% |
0.000 |
0.053 |
0.947 |
0.947 |
|
|
Bal Hassan (Upper
Bakhtiari) Formation |
1% |
0.171 |
0.051 |
0.778 |
0.949 |
|
|
Alluvial fan
deposits |
4% |
0.078 |
0.051 |
0.871 |
0.949 |
|
|
Flood plain
deposits |
0% |
0.000 |
0.053 |
0.947 |
0.947 |
|
|
River terraces |
0% |
0.219 |
0.052 |
0.730 |
0.948 |
|
|
River |
1% |
0.053 |
0.052 |
0.895 |
0.948 |
|
|
Soil (Appendix, Figure C7) |
Chestnut |
8% |
0.448 |
0.192 |
0.360 |
0.808 |
|
Lithosols,
Rendzinas, Chromic cambisols |
62% |
0.297 |
0.135 |
0.569 |
0.865 |
|
|
Lithosols,
Rendzinas, Calcic Xerosols, Chromic cambisols |
25% |
0.070 |
0.255 |
0.675 |
0.745 |
|
|
Lithosols, Calcaric Regosols, Calcic Xerosols, Chernozems |
5% |
0.186 |
0.209 |
0.605 |
0.791 |
|
|
Rough |
1% |
0.000 |
0.209 |
0.791 |
0.791 |
|
|
Land use (Appendix, Figure C8) |
Water Body |
1% |
0.415 |
0.196 |
0.390 |
0.804 |
|
Forest |
25% |
0.057 |
0.240 |
0.703 |
0.760 |
|
|
Built-up |
3% |
0.158 |
0.199 |
0.642 |
0.801 |
|
|
Barren Land |
59% |
0.143 |
0.183 |
0.675 |
0.817 |
|
|
Cultivated Land |
12% |
0.228 |
0.182 |
0.590 |
0.818 |
|
|
Distance from Road (Appendix, Figure C9) |
0 –25 m |
29% |
0.333 |
0.124 |
0.543 |
0.876 |
|
25 –50 m |
24% |
0.195 |
0.171 |
0.634 |
0.829 |
|
|
50 –75 m |
19% |
0.153 |
0.178 |
0.669 |
0.822 |
|
|
75 –100 m |
14% |
0.079 |
0.184 |
0.737 |
0.816 |
|
|
100 –125 m |
10% |
0.146 |
0.173 |
0.681 |
0.827 |
|
|
>125 m |
4% |
0.093 |
0.171 |
0.736 |
0.829 |
|
|
Distance from
River (Appendix, Figure C10) |
0 –50 m |
33% |
0.495 |
0.099 |
0.407 |
0.901 |
|
50 –100 m |
29% |
0.184 |
0.190 |
0.626 |
0.810 |
|
|
100 –150 m |
20% |
0.196 |
0.180 |
0.624 |
0.820 |
|
|
150 –200 m |
11% |
0.126 |
0.180 |
0.694 |
0.820 |
|
|
200 –250 m |
5% |
0.000 |
0.178 |
0.822 |
0.822 |
|
|
>250 m |
2% |
0.000 |
0.173 |
0.827 |
0.827 |
|
|
Drainage density (Appendix, Figure C11) |
0 - 0.56 |
60% |
0.077 |
0.315 |
0.608 |
0.685 |
|
0.56 - 0.86 |
12% |
0.144 |
0.186 |
0.670 |
0.814 |
|
|
0.86 - 1.43 |
15% |
0.323 |
0.145 |
0.533 |
0.855 |
|
|
1.43 - 1.72 |
8% |
0.301 |
0.169 |
0.530 |
0.831 |
|
|
1.72 - 2.0 |
5% |
0.156 |
0.185 |
0.659 |
0.815 |
Table A2. The final
weight of the significant elements represents
the flood's potential importance in the AHP technique.
|
Factors |
Factor weights |
Class |
Class weights |
Pixels |
Weights (%) |
|
Slope |
0.052 |
0º – 15º |
0.020 |
2847957 |
38% |
|
15º – 30º |
0.020 |
2849154 |
38% |
||
|
30º – 45º |
0.010 |
1480220 |
20% |
||
|
45º – 60º |
0.002 |
327470 |
4% |
||
|
60º – 80º |
0.000 |
36494 |
0% |
||
|
Aspect |
0.027 |
North |
0.003 |
968570 |
13% |
|
Northeast |
0.003 |
870225 |
12% |
||
|
East |
0.003 |
738698 |
10% |
||
|
Southeast |
0.003 |
963574 |
13% |
||
|
South |
0.005 |
1270832 |
17% |
||
|
Southwest |
0.004 |
1122706 |
15% |
||
|
West |
0.003 |
761336 |
10% |
||
|
Northwest |
0.003 |
845354 |
11% |
||
|
Elevation |
0.035 |
402 - 800 |
0.014 |
2911766 |
39% |
|
800 - 1300 |
0.013 |
2787192 |
37% |
||
|
1300 - 1600 |
0.007 |
1410943 |
19% |
||
|
1600 - 2000 |
0.002 |
369680 |
5% |
||
|
>2000 |
0.000 |
61714 |
1% |
||
|
Rainfall |
0.098 |
700 - 800 |
0.022 |
1713774 |
23% |
|
800 - 900 |
0.031 |
2395279 |
32% |
||
|
900 - 1000 |
0.015 |
1156930 |
15% |
||
|
1000 - 1100 |
0.013 |
991732 |
13% |
||
|
1100 - 1300 |
0.017 |
1283549 |
17% |
||
|
Drainage density |
0.059 |
0 - 0.56 |
0.035 |
4523909 |
60% |
|
0.56 - 0.86 |
0.007 |
937667 |
12% |
||
|
0.86 - 1.43 |
0.009 |
1133371 |
15% |
||
|
1.43 - 1.72 |
0.005 |
575488 |
8% |
||
|
1.72 - 2.0 |
0.003 |
370829 |
5% |
||
|
Distance from
River |
0.12 |
0 –50 m |
0.039 |
2475505 |
33% |
|
50 –100 m |
0.034 |
2152474 |
29% |
||
|
100 –150 m |
0.023 |
1471414 |
20% |
||
|
150 –200 m |
0.014 |
859135 |
11% |
||
|
200 –250 m |
0.006 |
403107 |
5% |
||
|
> 250 m |
0.003 |
179629 |
2% |
||
|
Land use |
0.082 |
Water Body |
0.001 |
87193 |
1% |
|
Forest |
0.021 |
1916937 |
25% |
||
|
Built-up |
0.002 |
228532 |
3% |
||
|
Barren Land |
0.048 |
4435024 |
59% |
||
|
Cultivated Land |
0.009 |
873492 |
12% |
||
|
NDVI |
0.138 |
-0.5 to -0.07 |
0.030 |
2601235 |
22% |
|
-0.07 – 0.16 |
0.041 |
3470022 |
29% |
||
|
0.16 – 0.25 |
0.034 |
2925801 |
25% |
||
|
0.25 – 0.37 |
0.023 |
1964136 |
17% |
||
|
0.37 – 0.77 |
0.010 |
822089 |
7% |
||
|
Distance from Road |
0.038 |
0 –50 m |
0.011 |
2219874 |
29% |
|
50 –100 m |
0.009 |
1819891 |
24% |
||
|
100 –150 m |
0.007 |
1425567 |
19% |
||
|
150 –200 m |
0.005 |
1033994 |
14% |
||
|
200 –250 m |
0.004 |
748374 |
10% |
||
|
> 250 m |
0.001 |
293564 |
4% |
||
|
0.176 |
Gercusi iformation |
0.006 |
251405 |
3% |
|
|
Pilaspii iformation |
0.007 |
296001 |
4% |
||
|
Sehkaniyani andi Sarkii iformation |
0.002 |
92034 |
1% |
||
|
Chiai Gara,i Barsarin,i Naokelekani andi Sargelui iformation |
0.005 |
211047 |
3% |
||
|
Balambo,i Garagui andi Sarmordi iformation |
0.008 |
353511 |
5% |
||
|
Qamchuqai iformation |
0.034 |
1450876 |
19% |
||
|
Tanjeroi iformation |
0.003 |
111483 |
1% |
||
|
Aqra-Bekhmei iformation |
0.059 |
2548840 |
34% |
||
|
Shiranishi iformation |
0.013 |
569951 |
8% |
||
|
Fathai (Loweri Fars)i iformation |
0.007 |
282264 |
4% |
||
|
Mukdadiyahi (Loweri Bakhtiari)i iformation |
0.005 |
203665 |
3% |
||
|
Injanai (Upperi Fars)i iformation |
0.007 |
307547 |
4% |
||
|
Slopei ideposits |
0.005 |
229963 |
3% |
||
|
Koloshi iFormation |
0.002 |
97375 |
1% |
||
|
Bali Hassani (Upperi iBakhtiari)i iformation |
0.002 |
68402 |
1% |
||
|
Alluviali fani deposits |
0.007 |
297859 |
4% |
||
|
Floodi plaini deposits |
0.001 |
30479 |
0% |
||
|
Riveri iterraces |
0.001 |
26734 |
0% |
||
|
iRiver |
0.003 |
110493 |
1% |
||
|
Soil |
0.175 |
iChestnut |
0.014 |
585913 |
8% |
|
Lithosols,i Rendzinas,i Chromici icambisols |
0.108 |
4647205 |
62% |
||
|
Lithosols,i Rendzinas,i Calcici Xerosols,i Chromici cambisols |
0.044 |
1876501 |
25% |
||
|
Lithosols,i Calcarici Regosols,i Calcici Xerosols,i iChernozems |
0.008 |
353653 |
5% |
||
|
iRough |
0.002 |
77992 |
1% |
Figure C1. Elevation map of the study area. Figure C3. Aspect map of the study
area. Figure C2. Slope map of the study area. Figure C4. Rainfall
map of the study area.
Figure C9. Distance from the Road map of the study area. Figure C10. Distance from main river map of the study area. Figure C7. soil map of the study area. Figure C11. Drainage density of the study area. Figure C8. Land use
and land cover map of the study area.
This is an open access under a CC BY-NC-SA 4.0
license (https://creativecommons.org/licenses/by-nc-sa/4.0/)