Introduction

In recent times, many urban areas, infrastructure and lifelines (bridges, railways, highways, power, and gas lines), and agricultural lands in all countries have been affected by flood hazards, which are considered the most common catastrophic natural hazards (Vojtek and Vojteková 2019; Mishra and Sinha 2020; Sarkar and Mondal 2020; Ahmad and Afzal 2020, 2022; Li et al. 2022). The rapid increase in population is forcing people to settle the low-lying areas that intersect or are close to the wadis/rivers. These areas are becoming more vulnerable to flooding due to current and future predicted climate changes and extreme meteorological events (Bubeck and Thieken 2018; Alexander et al. 2019; Ali et al. 2019; Huang et al. 2019; Xu et al., 2019; Khan et al. 2021). Floods are often the most dangerous natural hazards (more dangerous than landslides, earthquakes, and volcanoes), causing enormous loss of life and injury, massive economic disruption, and contamination of areas with disease (Ceola et al. 2014; Dandapat and Panda 2017). According to the statistical data from the United Nations Office for Disaster Risk Reduction (UNISDR), between 1995 and 2015, approximately 150,061 flood events occurred globally, and about 157,000 fatalities due to floods, accounting for about 11% of global disaster victims (Wahlstrom and Guha-Sapir 2015). In 2019, EM-DAT recorded 396 natural disasters with 11,755 fatalities, 95 million people affected, and $103 billion in economic losses around the world. The impacr was not evenly distributed, as Asia was the hardest hit. Floods were the deadliest disaster type, accounting for 43.5% of fatalities (CRED 2020). Several studies have reported that floods affect approximately 200 million people annually and cause economic losses of US$95 billion worldwide (Ceola et al. 2014; Mabuku et al. 2018).

Most flash floods occur with sever intensity, in a short period of time, and with high flow velocity occurring suddenly and with little or no time to react, posing a great danger to human and property (Sene 2013). Precipitation in arid areas is limited by cloud size (Laity 2008). In the last two decades, arid regions have faced many extreme rainfall events that caused immense devastation and loss of life, including Morocco 2008 2014; Algeria 2007, 2008, and 2013; and Saudi Arabia 2009, 2001, 2015, 2017, and 2018 (Kenyon 2007; Yamani et al. 2016; Echogdali et al. 2018; Abu-Abdullah et al. 2020).

Accordingly, flood vulnerability mapping is an extremely important step for predicting flood probability and mitigating and controlling future floods (Kourgialas and Karatzas 2011). Recently, various techniques and models have been applied to delineate flood-prone areas using remote sensing data and GIS (Mandal and Chakrabarty 2016; Siahkamari et al. 2018; Dano et al. 2019; Kanani-Sadata et al. 2019; Liu et al. 2019; Mahmood and Rahman 2019; Wang et al. 2019; Sahana et al. 2020). One approach that deals with the concept of flood inundation is depending mainly on the hydrological characteristics of a given watershed to estimate the peak discharge during a given return period. The high-resolution digital elevation model of urban regions is used to apply the inundation model and produce inundation maps with water depth and flood velocity using hydrological models such as the Soil Water Assessment Tool (SWAT) and Hydraulic Engineering Center-River Analysis System (HEC-RAS) (Getahun and Gebre 2015; Pal and Pani 2016; Prasad and Pani 2017; Youssef et al. 2021). Additionally, other studies are mainly based on flood-related factors that have a significant impact on flood hazard assessment (e.g., lithology, slope, aspect, curvature, elevation, distance from streams, drain type, slope length (LS), topographic wetness index (TWI), and land use/land cover pattern). Several authors have applied various techniques and approaches to assess the flood susceptibility of a region, such as (1) heuristic (multicriteria analysis), (2) statistical techniques, and (3) machine learning techniques. Flood susceptibility maps (FSMs) could play an extremely important role in establishing early warning systems, contingency plans, reducing and preventing future inundations, and implementing flood management policies and regulations (Mandal and Chakrabarty 2016; Tehrany et al. 2019). Each approach has its own advantages and disadvantages. For example, heuristic models are highly subjective and largely depend on human perception, judgment, and experience to determine the weighting of each flood-related factor (Bathrellos et al. 2017; Dandapat and Panda 2017; Souissi et al. 2019; Vojtek and Vojteková 2019; Nachappa et al. 2020). On the other hand, bivariate and multivariate models have recently been used to overcome human judgment and enhance the accuracy of flood vulnerability by using various computational methods (e.g., weights-of-evidence, frequency ratio (FR), information value (IV), Shannon entropy (SE), statistical index (SI), weighting factor, logistic regression (LR), fuzzy logic (FL), and neuro-fuzzy logic) (Kourgialas and Karatzas 2011; Kia et al. 2012; Feng et al. 2015; Park et al. 2019; Paul et al. 2019; Sahana and Patel 2019; Ali et al. 2020; Sahana et al. 2020). Recently, the most machine learning techniques (MLTs) were developed and used in different hazard susceptibility assessment (identifing and predicting flood-prone areas) among them artificial neural networks (ANNs), general linear models (GLMs), adaptive neuro-fuzzy interface systems (ANFIS), decision trees (DT), random forest (RF), support vector regression (SVR), boosted regression tree (BRT), GLMs, and classification and regression trees (CART) (Feng et al. 2015; Albers et al. 2016; Gizaw and Gan 2016; Muñoz et al. 2018; Zhao et al. 2018; Park et al. 2019; Dodangeh et al. 2020; Nhu et al. 2020; Tabbussum and Dar 2021; Sellami et al. 2022).

Egypt is located in an arid climate characterized by sparse rainfall, except for the northern and eastern parts, which are characterized by intense rainfall (El-Ghani et al. 2017). Many areas in Egypt have been increasingly affected by flooding in recent decades, resulting in economic collapses, deaths and injuries, and infrastructure disruptions. These areas that experienced dangerous flooding include Drunka village in November 1994, Wadi Al-Arish in January 2010, Alexandria and Beheira governorate in 2015, the northwestern region of Egypt in 2016, and New Cairo district in April 2018 and October 2019 (Elkhrachy et al. 2021). To the best of our knowledge, few documents have been found that specifically relate to these flood events in the study area and its vicinity. For example, in October 2016, severe flooding occurred in Ras Gharib aea, damaging thousands of homes, destroying much of the infrastructure, and claiming the lives of 22 people (Youssef and Hegab 2019; Hermas et al. 2021). Accordingly, flood disasters have greatly increased, mainly due to increasing urbanization (residential areas and buildings) and infrastructure (highways, railways, and roads) in flood-prone areas (Moawad 2013; Moawad et al. 2016; Youssef and Hegab 2019; El-Haddad et al. 2021). Therefore, it is crucial to manage floods and minimize or avoid their risk, which requires flood forecasting programs and inundation modeling. Bubeck et al. (2012) pointed out that flood forecasting minimizes flood-related hazards (fatalities and associated economic losses). The core concept of the flood control strategy is the ability to delineate flood-prone areas (Sarhadi et al. 2012).

The objective of the current study is to compare the performances of seven advanced MLTs to determine the most optimal flood susceptibility model using remote sensing and GIS approaches. These models include BRT, functional data analysis (FDA), GLM, multivariate adaptive regression spline (MARS), multivariate discriminant analysis (MDA), RF, and support vector machine (SVM). They were applied based on several characteristics. Flood modeling analysis using MLTs is new to Egypt; they are suitable for small and medium applications, they have an objective statistical basis, they can quantitatively analyze the contribution of factors to flood development, and they are mainly based on RS data rather than detailed field work. The results of our study can make important scientific contributions, and the optimal flood susceptibility map can be suitable for disaster management analysis to identify and outline flood-prone areas so that decision makers and land use planners can select favorable locations for future urban development.

Study area

The study area extends from Safaga to Ras Gharib, with an area of approximately 10,537 km2. It is situated between latitudes 26°40′00″ and 28°20′00″ N and longitudes 32°50′00″ and 34°0′00″ E (Fig. 1). The watershed is characterized by various physiographic features, including mountains, hills, main wadis, and streams. The elevation ranges from 0 m (Red Sea coast in the east) to 2068 m (mountainous areas in the west) above the mean sea level. The slope angle varied between 0 and 72° (with an average of 8.2° and a standard deviation of 10.4°). Approximately 16.1% of the total area has a slope greater than 30°, 16.4% of the area has a slope between 15 and 30°, 61.7% of the area has a slope between 5 and 15°, and 5.8% of the area has a slope of less than 5°. The study area is characterized by various rock units including bedrock complex in the west (40.8% of the study area), sedimentary rocks, and alluvial soils (wadi deposits), which occupy approximately 40.8%, 15.1%, and 44.1% of the study area, respectively.

Fig. 1
figure 1

a Location map of the study area along the Red Sea and the Gulf of Suez coast; b close-up of the study area showing cities, road networks, holiday areas, flood detection points (training and validation points), and comparison sites (1, 2, 3, and 4)

Full size image

Egypt is located in an arid and semiarid climate with low rainfall in the south and west and intense rainfall in the north and east. Desert areas account for more than 90% of the land area (El-Ghani et al. 2017). Flash floods are the worst meteorologically induced disasters in Egypt. They are dangerous, occur suddenly and unpredictably (with high intensity and in short duration), and cause economic collapses (Ezz 2017). In the study area, precipitation is typically infrequent and occurs in the form of heavy thunderstorms from November to April. Unfortunately, there are no records of precipitation, as there are few precipitation stations in the area. The area under study has been largely developed due to floods in 2014 and 2016, where the study area experienced heavy and short-duration rainfall that caused devastating destruction for Hurghada-Safaga and Qena-Safaga highways, the Hurghada airport, and inundated Hurghada urban area (Fig. 2) In addition, rockfall along highways are affected the area due to heavy rainfall (Fig. 2). Consequently, the future planning and development in the study area will be affected by flood hazards. The area is characterized by numerous main streams that cut through the area, making it a flood-prone area (e.g., Wadi Abu Naakhra, Bali, Aish, Milaha, Abu Had, and Gharib). The eastern part of the study area, the low-lying areas, are particularly vulnerable to flooding from the western part. The area, which is constantly subject to flood damage, undergoes cascading changes over time. This presents a constraint for the spatial flood assessment. If the site information is erroneous, it can cause significant problems in the spatial analysis. However, drainage structures and water supply systems may have an impact on flood vulnerability assessments. The change in land use in the eastern portion of the study area from desert to residential and infrastructure and the lack of action plans or inadequate engineering solutions to prevent flood events.

Fig. 2
figure 2

Various photographs taken and collected during different flood events in the study area (by authors and from social media). Impact of the flood in 2014: a, b The flooding of Hurghada-Safaga Highway; c, d the impact of the flood on Hurghada airport. Impact of flood in 2016: e flooding of Qena-Safaga highway; f rockfall along highways due to heavy rainfall; g, h the inundation of the Hurghada City

Full size image

Data and methodology

This study demonstrates the use of different datasets that can be used in flood susceptibility mapping. Several critical steps of the methodology were followed in this study to ensure the reliability of the yield models. These critical steps are shown in Fig. 3 and are explained in the following sections.

Fig. 3
figure 3

A diagram showing data used and modeling steps applied to provide an accurate FSMs

Full size image

Data used

Table 1 describes the various datasets that were collected and extracted for this study. Field surveys were conducted to collect various features and evidence related to the consequences of flood events that affected the study area. Questionnaires with residents of the area (local people and Badwins) and historical documents (from the Civil Defense Agency and the Department for Transport) were collected and used to understand previous flood events. Photographs were taken and maintained documenting various flood events that affected different parts of the study area. Remote sensing data were acquired for the study area, including Landsat 8 and OLI sensor (Operational Land Imager) (acquired in 2019, 30-m spatial resolution) from the Earth Explorer website (https://earthexplorer.usgs.gov). The image mosaic (30-m resolution) was created by overlaying the bands (1–7) and then fused with the panchromatic band (15-m resolution) to generate the final image mosaic (15-m resolution). Additional high-resolution images were obtained using an astro digital 2.5-m resolution and Professional Google Earth. Remote sensing imagery was used to create land use/land cover, flood inventory, lithology, and hydrolithology unit layers. In addition, a 30-m resolution DEM was obtained from ALOS World 3D-30m. DEM was used to generate various datasets (for example, elevation, slope aspect, lithology, LULC, LS, TWI, slope angle, plan and profile curvatures, stream power index (SPI), and hydrolithology units). Finally, a 1:100,000-scale geologic map was prepared and digitized to delineate different lithological units and hydrolithological units. The data of this study with different resolutions due to different sources, as previously described, were converted into themes with a grid size of 30-m resolution and stored in a digital database with a uniform projection (UTM zone 36 and WGS84 datum).

Table 1 Data utilized and applied in the current work
Full size table

Flood inventory map

Based on historical data and previous flood events, flooded areas were extracted to construct an inventory map. The inventory map is an extremely crucial element in flood susceptibility modeling (Sarkar and Mondal 2020). Several authors have pointed out that areas that have been exposed to past flood events under the same conditions are most likely to be vulnerable to current flood events (Fotovatikhah et al. 2018). To prepare susceptibility maps, it is necessary to determine the relationship between the inventory map (existing problems) and various factors that are relevant to susceptibility (Petley 2008). Different types of data (e.g., historical records, field visits, and satellite imagery interpretation) were used to generate an inundated inventory layer (Fig. 2b). Previously flooded areas (in the form of points) were extracted by comparing the study area before and after the flood events (2014 and 2016) using visual inspection of (1) high-resolution imagery (Google Earth and astrodigital imagery) and (2) medium-resolution imagery (Landsat 8 OLI). Flooded site data were examined and identified during field investigations following the 2014 and 2016 flood events (Fig. 2). Additional inundation data in the form of coordinated locations were collected from the Civil Defense Agency and past news over the past three decades. To isolate the exact flooded areas using medium- and high-resolution remote sensing images, Landsat 8 (2014) imagery with a spatial resolution of 15 m and Astro digital (2016) imagery with a spatial resolution of 2.5 m were used in two time periods. Cloud-free images were acquired before and after the flood events in 2014 and 2016. Visual inspection of the true color images (bands 1, 2, and 3 in RGB) using ArcGIS 10.8 software was used to extract the flooded areas (Fig. 4). The inundated areas identified using satellite imagery were verified using field investigations and civil defense data. Finally, a point feature layer (420 flooded locations) of the inundated locations was created to produce the flood inventory layer (Fig. 1b). In flood susceptibility maps (FSM), spatial prediction consists of a binary classification of the data into two groups: flood and non-flood. In this study, a total of 420 flooded locations and an equal number of non-flooded locations (areas with a slope angle greater than 10°) were identified. Based on Tehrany et al. (2019), the non-flooded locations in the study area can enhance the accuracy of the results. The data points were randomly partitioned using R statistical software to divide the data into training and validation datasets (Naimi and Araújo 2016). The training datasets (70% of floods (295 locations) and non-floods (295 locations)) were used to build the flood susceptibility models, and the remaining (30% of floods (125 locations) and non-floods (125 locations)) were used as the validation dataset for model evaluation. The training and validation datasets were converted to a raster format. Flood and non-flood locations were coded as 1 and 0, respectively (Wang et al. 2019).

Fig. 4
figure 4

Close-up medium-resolution Landsat 8 images at sites 1 and 2 (as shown in Fig. 1b) before (a, c) and after (b, d) the 2014 flash flood event; high-resolution astro-digital images at sites 3 and 4 (as shown in Fig. 1b) before (e, g) and after (f, h) the 2016 flash flood event

Full size image

FRFs

The determination of key flood-related factors (FRFs) is essential for flood susceptibility modeling (Sanyal and Lu 2004), and they vary according to catchment characteristics (Waqas et al. 2021). Rainfall is considered the most influential factor in the occurrence of floods. Lawal et al. (2012) pointed out that there are several other flood-related factors that contribute significantly to flood hazards. Runoff along the catchment depends on the characteristics of the catchment (e.g., catchment area, topography, and LULC types) (Hölting and Coldewey 2019). In the current study, eleven flood-related factors (FRFs) were selected as thematic layers based on the sound information from different types of literature (the most commonly used factors in flood vulnerability assessment literature), data availability related to the current study area, and field investigation (Al-Juaidi et al. 2018; Kanani-Sadata et al. 2019; Liu et al. 2019; Paul et al. 2019; Wang et al. 2019; Vojtek and Vojteková 2019). These FRFs include altitude, slope aspect, lithology, land use/land cover (LULC), slope length (LS), topographic wetness index (TWI), slope angle, profile curvature, plan curvature, stream power index (SPI), and hydrolithology units (Fig. 5). They were generated and stored in spatial database themes with a grid cell size of 30 × 30 m in an ArcGIS environment for data processing. A digital elevation model (DEM) of the study area with a spatial resolution of 30-m was obtained from ALOS World 3D-30m), from which eight layers were generated. Of these, five factors, slope aspect, slope angle, altitude, plan curvature, and profile curvature were extracted using ArcGIS 10.8 software. The other three themes, including TWI, LS, and SPI, were generated using the SAGA software. Other factors such as lithology, land use/land cover (LULC), and hydrolithology unit maps were extracted using remote sensing images (Landsat 8-OLI and Google Earth), geological maps, and field surveys. Different types of FRFs were used in the present study, such as nominal (lithology, slope aspect, land use/land cover, and hydrolithology unit layers) and ordinal (altitude, TWI, slope angle, LS, profile curvature, plan curvature, and SPI).

Fig. 5
figure 5figure 5

FRF maps applied in the current study: a altitude, b slope aspect, c lithology, d landuse/landcover (LULC), e slope length (LS), f topographic witness index (TWI), g slope angle, h profile curvature, i plan curvature, j stream power index (SPI), and k hydrolithology units

Full size image

Altitude

According to several authors, altitude is influenced by various factors (e.g., lithologic unit, wind action, precipitation, and erosion) (Waqas et al. 2021). The occurrence of flooding is likely to be influenced by elevation, which is considered an influential factor in flooding. Low-elevation regions (flat areas) are more susceptible to flooding than higher-elevation areas because water flows from high altitudes to lower-elevation areas (Kia et al. 2012; Cao et al. 2016). The altitude layer was extracted from the DEM using ArcGIS and ranged from 0 to 2173 m (Fig. 5a).

Slope aspect

The slope aspect is the direction of the maximum inclination of the Earth’s surface. It affects the direction of runoff, which maintains the soil moisture (Chu et al. 2020). The slope aspect may indirectly affect flooding, as the inclined shaded regions are characterized by relatively high soil moisture, indicating high runoff (Islam et al. 2021). The slope-aspect theme was created from the DEM map of the ArcGIS platform. The slope aspect map was divided into nine categories (Fig. 5b).

Lithology

Because of the varying permeability of rocks and sediments in a watershed, lithological units play a crucial role in hydrological processes (variations in the quantity and rate of water flow and sediment production) (Ward and Robinson 2000). The drainage density depends on the type of material used. Çelik et al. (2012) and Srivastava et al. (2014) indicated that a low drainage density is associated with highly resistant rock or highly permeable subsoil material. Stefanidis and Stathis (2013) concluded that flood hazard zones are influenced by geological units, especially torrential formations characterized by erodibility and permeability. In the current study, lithological units were generated from lithological maps (1:100,000-scale). Four main geological units were identified: (1) wadi deposits (alluvum), (2) sandstone, (3) limestone, (4) evaporates, (5) shales, and (6) basement rocks (Fig. 5c).

LULC

Land use/land cover type (LULC) plays a critical role in runoff velocity, interception, infiltration, and evaporative transport (Yalcin et al. 2011). Various LULC features can affect the infiltration and surface flow generation in a catchment (Rahmati et al. 2015). Tehrany et al. (2019) indicated that forested areas can infiltrate more water into the subsurface than other LU/LC types. Many studies have shown that LULC types have a significant impact on distinguishing flood-vulnerable areas (Karlsson et al. 2017; Komolafe et al. 2018). The LULC layer was generated based on 2018 Landsat 8 satellite imagery (OLI) and classified into five categories using supervised classification in ENVI 5.4 software: wetlands, bare rock, bare soil, built-up area, and sandy soil with trees (Fig. 5d).

LS

The slope length (LS) is one of the influential factors determining soil erosion, where soil erosion accelerates with increasing slope length owing to the effects of higher accumulation of surface runoff (Bera 2017). LS shows the combined impacts of gradient length and steepness and affects particle transport “soil loss” and upland (mountainous) hydrological processes (Park et al. 2019). In this study, LS was calculated from the DEM layer according to the slope gradient and specific basin area using SAGA software based on the universal soil loss equation (USLE) (Eq. 1) (Moore and Burch, 1986):

$$\mathrm{LS}={\left(\frac{A_{\mathrm{s}}}{22.13}\right)}^{0.4}{\left(\frac{\mathrm{Sin}\beta }{0.0896}\right)}^{1.3}$$
(1)

A s (m2) is the specific area of the catchment, and β is the slope angle in degrees. In this study, the slope length (LS) ranged from 0 to 59.1 (Fig. 5e).

TWI

The TWI reflects the variation in the quantity of water gathered in a basin (wetness values) and is the relationship between the specific basin area and the gradient (Beven 2011; Gokceoglu et al. 2005). TWI can be strongly correlated with locations within a catchment that have a high potential for flooding (Chen and Yu 2011; Manfreda et al. 2011; Abdel Hamid et al. 2020). Tehrany et al. (2019) pointed out that flat areas can absorb more water than steep terrain. Accordingly, areas near drainage networks and flat lands (flood-prone areas) have higher TWI values than those in areas with slopes (Meles et al. 2020; Zhang et al. 2020). The TWI index value was extracted based on Eq. (2) (Beven and Kirkby 1979):

$$\mathrm{TWI}= lin\ \frac{\mathrm{A}}{\tan B}$$
(2)

where A is the cumulative basin area (m2), and β is the slope angle (in degrees) at a point. In this work, the TWI was created using SAGA-GIS software ranging from 1.5 to 22.8 (Fig. 5f).

Slope angle

The slope angle is a crucial physiographic element for flood behavior and occurrence (Meraj et al. 2015). High-gradient areas have less time for perculation, which leads to the acceleration of runoff velocity, resulting in the accumulation of immense runoff in the lower lying areas (around the river or in flat areas) and are more vulnerable to flooding (Stevaux et al. 2020). The slope-angle layer was generated from a DEM map using ArcGIS. The slope angle ranged from 0 to 72° (Fig. 5g).

Plan and profile curvatures

The curvature represents the slope shape and the terrain morphology. It is one of the key terrain elements used in several geomorphometric works (Rau et al. 2019; Torcivia and López 2020). Curvature is a major flood-controlling factor in flood vulnerability mapping (Ahmadlou et al. 2019). Cao et al. (2016) reported that the curvature has a significant impact on surface flow and infiltration. Shahabi et al. (2020) stated that areas with zero curvature values are more prone to flooding than areas with positive or negative curvatures. The curvature can be represented by the plan and profile curvatures. The plan curvature is directly correlated with the convergence and dispersion of surface runoff (Nasiri Aghdam et al. 2016). At the same time, Xiao et al. (2019) indicated that profile curvature impacts material deposition on the slope (by controlling the deposition increasing or decreasing of these materials). In the following study, plan and profile curvatures were generated from the DEM layer using ArcGIS software. The values of the plane and profile curvatures ranged from −0.0249 to 0.0233 and −0.0193 to 0.0208, respectively (Fig. 4h, i).

SPI

The SPI is a crucial hydrological factor that plays a vital role in assessing the spatial variation of flood-vulnerable areas (Deepak et al. 2020). SPI directly correlates with the erosive power of the catchment, soil water content status in a basin, and discharge relative to a specific area within the watershed (the power of flood water to flow downward) (Cao et al. 2016). High SPI values indicate high flood power, and lower values indicate that the terrain in the watershed has the potential for impound flow (Turoglu and Dolke 2011). The SPI of the catchment was calculated using Eq. (3) (Wu et al. 2020).

$$\mathrm{SPI}={A}_{\mathrm{s}}\ \tan \beta$$
(3)

A s is the specific basin area, and β is the local slope angle (in degrees). The SPI values in the current study range from 0 to 3.24 (Fig. 5j).

Hydrolithology units

Some soil types have a decisive influence on rainfall runoff mechanisms. Hydrolithology impacts the rate of water infiltration and ultimately the accumulation potential at the soil surface (Norbiato et al. 2008). The higher the infiltration rate, the less likely is the occurrence of flooding (Phillips et al. 2019; Xie et al. 2019). In this study, a hydrolithology unit map was created by integrating Landsat 8 satellite imagery (OLI), Google Earth imagery, geological data, and field investigations. According to the national soil classes and soil taxonomy, the hydrolithology unit map of the study area was classified into three categories: well-drained (covered 10.1% of the total area), semi-drained (covered 41.9% of the total area), and impervious (covered 48.0% of the total area) (Fig. 5k).

The FRF effectiveness and contribution

Multicollinearity analysis is a technique used to determine the effectiveness of independent variables in a model (Dormann et al. 2012). It is a statistical method in which independent parameters in a model are highly correlated using multiple regression techniques, and the parameters with high collinearity are deleted (Saha 2017). The multicollinearity technique uses two indicators, namely variance inflation factors (VIF) and tolerance (TOL) (Eqs. 4 and 5):

$$\mathrm{TOL}=1-{R}_J^2$$
(4)
$$\mathrm{VIF}=\left[\frac{1}{T}\right]$$
(5)

\({R}_J^2\) represents the regression coefficient of explanatory factor J for the rest of the descriptive factors. Previous studies have shown that a TOL < 0.10 and a VIF > 5 account for problems of multicollinearity (Menard 2001).

Evaluating the importance of independent factors is crucial for flood susceptibility analysis. It can be applied to determine the contribution of various flood-related factors and accurately determine their role in modeling production. Several methods have been applied to evaluate relationships between related factors and events. These techniques have been used and received much attention, including random forest (RF) and partial least squares (PLS) (Wang et al. 2016; Huang et al. 2018). PLS was used in this study. PLS is a strong multivariate regression technique that enables a broad spectrum of processes to be performed (Martens and Martens 2000). It has many advantages, such as it allows a quick understanding of the essential sequence of variations in the data; it is suitable for the analysis of noisy data, collinear, and even incomplete parameters, and it helps to detect errors in the input data (Wold et al. 2001). PLS was used for multivariate calibration of a dependent parameter against many independent parameters. Accordingly, it was suitable for the selected critical factors in the analysis. Details of PLS functions and applications have been explained in various studies (e.g., Lowry and Gaskin 2014). In the present study, the contribution and importance of all FRFs to flooding occurrence were evaluated using partial least squares (PLS).

Theoretical background of methods used

Problems related to natural hazards, such as floods, landslides, and ground subsidence, have been identified and solved using various machine learning techniques (MLTs) (Park et al. 2014; Shi et al. 2016; Ghorbanzadeh et al. 2019; Kavzoglu et al. 2019; Sevgen et al. 2019; Eini et al. 2020). Despite the continued advantages of MLTs as a powerful method, human expertise still plays an essential role in hazard assessment (Marjanović et al. 2011). In the current study, seven MLTs were utilized to evaluate their effectiveness in flood susceptibility mapping. These include SVM, RF, MARS, BRT, FDA, GLM, and MDA, which are discussed in detail in the following sections.

SVM model

The key elements of the SVM model are the utilization of classification and regression, which relate to the learning control approach (Vapnik 2013). SVM is a supervised learning method that deals with binary classification models (Amiri et al. 2019). The results provided minimal clustering errors and determined the optimal response (Vapnik 2013). This provides a key advantage in effectively identifying and analyzing factors (Micheletti et al. 2014). SVM has been used to create flood-prone areas (Yang and Cervone 2019). Many authors have provided detailed studies on SVM techniques (e.g., Yao et al. 2008).

RF model

Random forest (RF) is an ensemble learning approach based on regression trees, where many classification trees are aggregated to quantify a classification (Calle and Urrea 2010; Micheletti et al. 2014; Thanh Noi and Kappas 2018; Hawryło et al. 2018). The RF model is a robust ML model owing to several advantages, including a large number of trees in the analysis, insensitivity to noise, unbiased estimation of generalization error, acceptance of most types of data, and determination of significant variables (Breiman 2001; Rodrigues and De la Riva 2014; Kim et al. 2018). RF can overcome outliers in predictors, automatically deal with omitted data, and increase diversity among classification trees (Breiman and Cutler 2015). The model RF was run in R software version R 3.5.3, using the package “randomForest” (Breiman and Cutler 2015).

MDA model

MDA, a supervised classification algorithm, is a linear discriminant analysis (LDA) in which a cluster is suggested as part of the closest group (Fraley and Raftery 2002). The normal distribution of variables is utilized to calculate the distance to the nearest collection, assuming that the variability and correlation between variables is uniform (Lombardo et al. 2006). MDA applies multiple normal distributions in every class. MDA can be derived from linear combinations using Eq. (6) (Hair et al. 1998).

$$Y=W1X1+W2X2+ WnXn$$
(6)

Y represents the discriminant value, Wi (i = 1, 2, 3, …, n) are discriminant weights, and Xi (I = 1, 2, 3, …, n) are independent variables. The MDA analysis was run in R software version R 3.5.3, using the package “mda” (Hastie et al. 2017).

MARS model

MARS is a powerful regression algorithm owing to its flexibility in predicting events (Adnan et al. 2019). MARS considers both linear and non-linear relationships between independent and dependent factors and reflects these functions as coefficients used to calculate the effects of these factors separately (Busto Serrano et al. 2020). MARS has been used in various uses to evaluate relationships between different disciplines (e.g., geophysics, climatology, ecology, and geomorphology) (Deichmann et al. 2002; Hjort and Luoto 2013; Abdulelah Al-Sudani et al. 2019). It also allows the determination of the relative importance of the independent variables in the predictions (Adnan et al. 2019). It is also used to split the datasets into multiple splines on an equivalent interval basis; each spline can be subdivided into subclasses by generating knots (Friedman 1991). The predictor MARS can be determined using Eq. 7, according to Hastie et al. (2001):

$$f(x)={\beta}_0\sum_{j=1}^p\sum_{b=1}^B\left[{\beta}_{jb}\left(+\right)\operatorname{Max}\left(0,{x}_j-{H}_{bj}\right)+{\beta}_{jb}\left(-\right)\operatorname{Max}\left(0,{H}_{bj}-{x}_j\right)\right]$$
(7)

where x, f(x), P, and B are the input, output, predictor variable, and basis function, respectively. Max (0, x − H) and Max (0, H − x) are BF and do not need to exist if their coefficients are 0. The H values are referred to as the nodes.

There are three steps in applying the MARS algorithm: (1) applying a stepwise forward algorithm to select spline basis functions, (2) deleting BFs until the “best” set is found by applying a stepwise backward algorithm, and (3) providing the final MARS approximation with some degree of continuity by performing a smoothing method. Generalized cross-validation (GCV) criteria were applied to delete the BF in order of least contribution using Eq. 8 (Craven and Wahba 1979).

$$\mathrm{GCV}=\frac{\frac{1}{N}\sum_{i=1}^N{\left[{y}_i-f\left({x}_i\right)\right]}^2}{{\left[1-\frac{C(B)}{N}\right]}^2}$$
(8)

N stands for the number of data points, and C(B) stands for a complexity penalty escalated with the number of BFs in the technique and is determined by Eq. 9:

$$C(B)=\left(B+1\right)+ dB$$
(9)

Here, d represents the penalty for every BF incorporated in the technique. This can also be considered a smoothing variable. The MARS technique was run in the software R 3.5.3, using the package “MARS” (Deichmann et al. 2002).

GLM model

The generalized linear model (GLM) is a linear regression model that can quantify and incorporate specific and temporal variables (Ozdemir and Altural 2013). The use of GLM can increase the accuracy and quality of the results because it uses multiple regression to develop a clear relationship between the dependent and independent variables (Scott et al. 1991). Moreover, it can predict numerous events as it can identify the best regression model (Federici et al. 2007; Payne 2015). Several authors have applied GLM to different spatial models (e.g., Dumbser et al. 2020). The relationship between the response variable and explanatory variables can be constructed using the GLM link function (Ahmedou et al. 2016; Kéry and Royle 2016; Soch et al. 2017). The predictions and variances of the response factors were estimated using Eqs. (10) and (11):

$${\mu}_i=E\left[{Y}_i\right]={g}^{-1}\left(\sum_j{X}_{ij}{\beta}_j+{\varepsilon}_i\right)$$
(10)
$$\operatorname{var}\left[{Y}_i\right]=\frac{\phi V\left({\mu}_i\right)}{\omega_i}$$
(11)

Y i denotes the vector of response parameters, Xij is the matrix of explanatory parameters, βj is the vector of floating variables, εi is the interference terms, g(x) is the corresponding link function, V(x) is the variance function, ϕ is the dispersion parameter of V(x), and ωi is the weight of the ith observed value.

In this work, it is assumed that Y is the response parameter representing the flooded area in a grid, and Xi is the ith flood-related parameter. Thus, the occurrence probability of a flooding event Y is represented by Eq. (12): By logistic transformation, the link function g(yi) is represented by Eq. (13).

$$P=\frac{\exp \left({c}_0+{c}_1{X}_2+{c}_2{X}_2+\dots +{c}_i{X}_i\right)}{1+\exp \left({c}_0+{c}_1{X}_2+{c}_2{X}_2+\dots +{c}_i{X}_i\right)}$$
(12)
$$g\left({y}_i\right)={c}_0+\sum {c}_i{x}_i+{\varepsilon}_i$$
(13)

where P is the probability of occurrence of event Y; c0, c1, …, ci are logistic regression coefficients; and εi is the residual error.

In the present study, R software was used to construct the GLM model. A simple Gaussian family is determined as a link function for normally distributed response data. The independent factors were included in the model separately, using a smoothing spline with only two degrees of freedom in a polynomial of degree 2 to avoid overfitting (Aertsen et al. 2009)

FDA model

Ramsay and Dalzell (1991) proposed the FDA model as a statistical method to analyze the effect factors. The crucial concept of FDA is to treat an observed object with functional properties as an integral, regardless of the order of the observed values (Battista et al. 2016; Wagner-Muns et al. 2018). It can discriminate unsupervised work, where each class is divided into subcategories with a unique value (Chamroukhi et al. 2012; Zou et al. 2019). The FDA is a non-parametric method that is widely used in problem classification (Lu 2007; Seifi Majdar and Ghassemian 2017). Ray et al. (2019) summarized that the FDA model is a combination of regression models that perform an unseen operation for each category in the modeling analysis when applying complex class models. The basic tasks in applying the FDA model include (1) implementing a functional data representation by selecting training and testing datasets, (2) using functional principal component analysis (FPCA) to extract functional data features, (3) using machine learning methods to classify data features, and (4) testing datasets to verify the validity of the classification method. In this study, the FDA model was used to generate a flood vulnerability map using the species distribution modeling (SDM) package in R software (Naimi and Araújo 2016).

BRT model

Friedman (2001) proposed BRT, which uses an integration of statistical and machine-learning techniques. The advantages of the BRT model are as follows: (1) the ability to improve the model performance by fitting and combining several models, (2) no data transformation or outlier removal is required, (3) sophisticated non-linear relationships can be fitted, and (4) interaction influences between variables are automatically accounted for (Elith et al. 2008; Park and Kim 2019). The combined strength of the regression tree and boosting algorithms can improve model accuracy and minimize variance (Aertsen et al. 2010). Model accuracy is improved by boosting, a powerful learning method that iteratively fits new trees to the residual errors of the existing tree composition (Döpke et al. 2017). The BRT model was run in R software version R 3.5.3, using the package “brt” (Ridgeway et al. 2013).

Modelling prediction and performance

Evaluating the predictive and performance accuracy of the susceptibility models used was critical. The cross-validation approach using receiver operating characteristic (ROC) and area under the curve (AUC) has been applied quantitatively and graphically by various authors (Akgun et al. 2012; Ozdemir and Altural 2013;Youssef and Hegab 2019). The cross-validation approach offers many advantages, including quantitative evaluation of model prediction, determination of a better prediction approach, ability to compare the predictive capabilities of different models, ability to distinguish the less and most vulnerable areas, identification of the influencing factors and their contribution to prediction, evaluation of the effectiveness of the input parameters to the models, and improvement of the quality of model prediction. The ROC method is a statistical indicator of model performance based on the rates of true and false positives (sensitivity and 1-specificity) (Chung and Fabbri 2003; Mathew et al. 2009). The acceptable susceptibility model must have an AUC value between 0.5 and 1. The effectiveness, accuracy, and reliability of the model were improved by a higher AUC value (equal to or close to 1.0). An AUC value of less than 0.5 is considered a random model (Marzban 2004). Sajedi-Hosseini et al. (2018) stated that the overall performance of the model can be identified by categorizing the AUC values as follows: incompetent model (AUC from 0.5 to 0.6), model with poor performance (AUC from 0.6 to 0.7), model with moderate performance (AUC between 0.7 and 0.8), and model with high fitness and performance (AUC 0.8).

Results and discussions

Multicollinearity test and variable importance

Multicollinearity analysis of the 11 flood-related factors utilized in this study is shown in Table 2. The tolerance ToL and VIF values indicated that flood-related variables selected in the current work were more than 0.1 (ToL = 0.468) and less than 5 (VIF = 2.135), respectively. Consequently, there is no multicollinearity among the selected FRFs so that they can contribute significantly to the model construction in this study.

Table 2 Multicollinearity results of flood-related factors
Full size table

Furthermore, a partial least squares (PLS) method was applied to evaluate the significance of influential flood-related factors. Figure 6 presents the PLS results, which show that slope angle, altitude, LS, and TWI are the most important factors, followed by LULC, SPI, slope aspect, hydrolithology units, lithology, and plan curvature, which are moderately important flood-related factors. Thus, profile curvature was less critical.

Fig. 6
figure 6

The importance of flood-related factors using partial least squares (PLS)

Full size image

FSMs

Due to exponential population growth, future sustainable development requires an accurate understanding of the spread of natural hazards in each area. Predictions made through the application of modeling and simulation techniques are critical to natural resources and sustainable development studies. This provides agencies, decision makers, planners, and engineers with practical, reliable, and accurate information about an area’s level of vulnerability to natural hazards. Seven MLT models (SVM, RF, MDA, MARS, GLM, FDA, and BRT) were successfully used for comparative analysis. Using the training dataset, these MLT models were used to generate the flood susceptibility models using ArcGIS 10.8 software for the study area (Fig. 7a–g). The natural break classifier (Jenks) was used in this study (Nicu 2018). These FSMs were then categorized into four groups: low, moderate, high, and very high susceptibility zones. The percentage of relative areas in each group was calculated for each model (Fig. 8). The results showed that the areas of low, moderate, high, and very high classes correspond to the flood susceptibility map (FSM): 38.0%, 17.3%, 10.8%, and 33.9% of the total area for SVM; 36.2%, 19.5%, 27.4%, and 16.9% of the total area for RF; 28.3%, 22.1%, 26.8%, and 22.8% of the total area for MDA; 40.9%, 21.2%, 22.2%, and 15.7% of the total area for MARS, 35.1%, 17.3%, 26.6%, and 21.1% of the total area for GLM; 27.3%, 22.3%, 26.9%, and 23.6% of the total area for FDA; and 35.1%, 4.7%, 10.4%, and 49.8% of the total area for BRT. The actual floodplains were extracted from the high-resolution imagery (Astro Digital and Google Earth imagery with spatial resolution of 2.5 m and 1 m, respectively) subsequent to the 2016 flood event to verify the accuracy and performance of the models used. The comparison indicates that there is good agreement, and the flood-prone areas in these seven models mainly occupy the main wadis and low-lying areas. However, low-susceptibility zones are located in highland regions.

Fig. 7
figure 7figure 7

Flood susceptibility models using MLTs models: a SVM, b RF, c MDA, d MARS, e GLM, f FDA, and g BRT

Full size image
Fig. 8
figure 8

FSM classes’ areas % for various MLTs in the current study

Full size image

FSM validation

The predictive performance of the MLTs producing flood vulnerability maps was evaluated using the ROC–AUC method. This method plots the sensitivity (percentage of currently inundated pixels correctly predicted by the model) against the 1-specificity (percentage of predicted inundated pixels in the entire area) (Fig. 7). The extracted FSMs were validated using the prediction rate method. The validation datasets (30% of the total flood locations), which were not previously used in building the models, were run to test how well the model predicted flooding (see Fig. 9). According to the AUC classification (Sajedi-Hosseini et al. 2018), as shown in Table 3, the AUC results of the current study indicated that the differences in model performance among MLTs were relatively moderate to high. Models with high performance were RF (AUC = 81.3%), GLM (AUC = 80.2%), and MARS (AUC = 80.1%). This was followed by the models with moderate performance, including BRT (AUC = 77.7%), MDA (AUC = 76.8%), FDA (AUC = 76.3%), and SVM (AUC = 73.3%) (Fig. 9).

Fig. 9
figure 9

Prediction rate curves for the FSMs generated in this study for different MLT models

Full size image
Table 3 AUC results of MLT models in the current study
Full size table

Discussion

Severe flood events, which are becoming more frequent in different areas as part of the effects of climate change, require significant efforts through the analysis of flood hazards, vulnerabilities, and risks. Therefore, flood management is the most important requirement for averting and reducing flood hazards. One such technique is flood vulnerability modeling, which is crucial for protecting people and developing viable and effective mitigation and management strategies worldwide (Sahana et al. 2020; Wang et al. 2020). The MLTs used in the present work, such as SVM, RF, MARS, BRT, FDA, GLM, and MDA, to generate flood-susceptibility maps provide remarkable results. One of these models is BRT, which is used in this study and provides a prediction rate of 77.7%, which is considered a reclosable value, and is used by other authors, suggesting that BRT is one of the best approaches for accurately determining flood-vulnerable areas (Darabi et al. 2020). Selecting 11 flood-related factors that were evaluated to determine the most flood-prone areas according to seven MLTs, the resulting map shows the statistical relationships between flood inventory data (real flooded areas) and flood-related factors. These eleven thematic layers were extracted and generated from various sources, including remote sensing data (high to medium resolution from 1 to 30 m), digital elevation models (30 m resolution), geological and topographic maps, and various field surveys. The inventory map (actual flooded areas) was created through various methods such as field visits (survey of flooded areas with GPS), historical documents (civil defense and local population), and high-resolution satellite imagery (1-m, 2.5-m, and 15-m resolutions). Potentially, flood-prone areas were mapped using seven MLTs in the current study. The results showed a significant correlation between model results. The validation of the resulting models was based on validation datasets not used in the training phase and provided relevant results ranging from 73.3 to 81.3% (values above 70%). The Red Sea region (especially the section between Safaga-Ras Gharib), which includes many urban areas (Ras-Gharib, Hurghada, and Safaga), various resorts, and attractive tourist sites, which are crossed by many lifelines (roads, power lines, railways, and highways), is crossed by many wadis that affect the area. For example, one of these events struck the area on October 18, 2016, in the proximity of Ras-Gharib City. This resulted in extensive flooding problems, 22 fatalities, many injuries, destruction of approximately 5000 houses, damage to many vehicles, erosion effects, and destruction of lifelines (main roads and streets) (Youssef and Hegab 2019).

The current work is new and innovative and, as expected, provides relevant mapping results, especially for preliminary planning management. The comparison of flood susceptibility maps produced by different machine learning models in the current study with flood hazard maps produced using LULC and DEM as the main parameters showed some differences, possibly because of the different types of data used in each model (Asare-Kyei et al. 2015; Mousavi et al. 2019). Flood hazards generated using LULC and DEM data are based on the extraction of runoff coefficients from land use/land cover (detection from remote sensing images), slope gradient layer and hydrological soil types, and precipitation intensity. Peak flows can be integrated with elevation data to generate a flood hazard map with a high performance rate of 87.83% (Mousavi et al. 2019).

In the current study, MLTs were used to identify flood-susceptible areas. The model prediction shows a minimum value of prediction performance of 73.3% for the SVM method and a maximum value of 81.3% for the RF approach. Even so, the prediction values are not high enough, and the machine learning techniques to produce appropriate flood-susceptibility maps are still high. In the current study, among the seven individual models applied in modeling flood susceptibility, the RF model showed better performance than the SVM, MARS, BRT, FDA, GLM, and MDA models. The main reason for this result may be the model’s ability to handle large databases and integrate large input variables without changing the variable. Also, RF uses high variance between trees accepting each tree for class membership. Then, it determines the class based on the highest number of votes. Similarly, RF can detect and predict non-linear interactions and relationships between effective factors (Catani et al. 2013). In previous studies on modeling natural hazards such as landslides (Chen et al. 2017), earth fissures (Choubin et al. 2019), gully erosion (Avand et al. 2019), air quality (Choubin et al. 2020), and floods (Chen et al. 2020), the RF model was also found to perform well. Our results for the prediction accuracy of RF (81.3%) and BRT (77.7%) are in agreement with the findings of Lee et al. (2017). They indicated that the RF approach provides a higher performance than the boosted tree technique. In the current work, the prediction performance based on RF was 79.18%, whereas it was 77.26% for BRT. Additionally, Mosavi et al. (2020) applied GLM, FDA, MARS, and RF for flood susceptibility, indicating that the prediction values using AUC are AUC of 93%, 92%, 89%, and 96%, respectively. They indicate that RF provides high performance value than other models. Also, in Costache et al. (2021), using various MLTs for flood susceptibility analysis, their results showed that RF provides the highest AUC value (97.3%). Satarzadeh et al. (2021) show that RF (AUC = 91.1%) has higher performance than SVM (AUC = 89.9%). Also, the strong capacity of FR to predict flood risk has also been demonstrated by Avand et al. (2020), Nachappa et al. (2020), and Norallahi and Kaboli (2021). Also, Rahmati et al. (2019) pointed out that when using SVM, BRT, and GAM for modelling multihazard delling, BRT has the highest validation and shows high accuracy for flood hazards (AUC = 94.2%). Various new and robust machine learning models have been applied by many researchers to achieve highly precise and accurate flood vulnerability mapping (Al-Abadi 2018; Paul et al. 2019; Wang et al. 2019; Ali et al. 2020; Nachappa et al. 2020; Wang et al. 2020; Islam et al. 2021). However, the prediction accuracy of these models is still difficult because of the different characteristics of the studied areas. The main advantages of MLTs are enormous, including their relative ease of use, which for reducing workload and time, and a variety of applications, it can handle different types of data. Their predictive accuracy usually outperforms some conventional methods, and they help us find ways to modernize technology. However, there are some limitations and side effects, such as the possibility of a high error rate due to the large amount of data, the inconsistency of data due to the large amount of data for training and testing, and selection of an algorithm is still a manual task, so the process is very time consuming. This helps us find various innovative ways to reduce these problems. Finally, these MLTs can automatically detect the relationships between flood-related factors and overcome uncertainties when reliable inventory maps (actual flooded areas) are acquired.

This work, using various MLTs, helped to understand the extent of flood hazard in the study area, which is crucial because many important facilities, critical highways, and tourist sites are located in this section of the area along the Red Sea and the Gulf of Suez. In addition, the MLTs provided deep insights into the importance of flood risk management. As millions of people visit these areas each year, they represent future income for the country, in addition to the residents of these areas. Therefore, a safe environment for this area is essential. This study has shown that many coastal areas are in worrisome areas. For this reason, we recommend that agencies, managers, and developers in the area study areas of high flood vulnerability and carefully consider the results of future development models to select the areas with the lowest risk. Areas with high or very high flood vulnerability should be studied in detail to determine flood protection measures.

Conclusions

In the context of climate change, floods are considered the most damaging natural disasters, causing socio-economic disruptions, loss of life, and property damage. Accordingly, low-lying areas surrounded by mountainous regions are at risk of flooding. These areas are prone to flooding. Effective and reliable techniques are required to delineate flood-prone areas. In this study, the spatial distribution of flood-prone areas in the coastal area of the Red Sea (between Safaga-Ras Gharib) in Egypt was investigated using seven machine learning models (SVM, RF, MARS, BRT, FDA, GLM, and MDA). The identification of flood-prone areas is crucial for most planners, the private sector, and decision makers. The analysis was mainly based on the identification of 420 flood points (295 points were used as training and 125 points for validation). Eleven flood-related factors were used, including elevation, slope, lithology, LULC, LS, TWI, slope, profile curvature, plan curvature, SPI, and soil runoff. Multicollinearity diagnostic tests (VIF and TOL) were used to test the suitability of lood-related factors. The partial least squares (PLS) approach was applied to identify the significance of flood-related factors. The ROC curve was constructed to check the flood-vulnerability models based on the validation datasets. Accordingly, the ability of seven MLTs to map most flood-prone areas is presented in this paper. The evaluation of the reliability and predictive performance of the FSMs produced by the SVM, RF, MARS, BRT, FDA, GLM, and MDA models showed that RF performed the best, followed by GLM and MARS, which produced more than 80%, indicating significantly better results. The BRT, MDA, and FDA algorithms provided moderately significant results. Finally, the SVM provides less significant results than the other models. The average results of all MLTs showed that 34.4% of the study area had low flood susceptibility, 17.8% had moderate flood susceptibility, 21.6% had high flood susceptibility, and 26.3% had very high flood susceptibility (extremely flood prone). Our results show that MLTs provide prediction values greater than 0.7% (70%), indicating that the models are adequate for flood susceptibility mapping in the area under consideration. Furthermore, the results indicate that these techniques have moderate to high performance in analyzing flood susceptibility, with such small differences. Therefore, these results can be viewed with greater confidence and applied in future studies to investigate the flood hazard distribution and provide helpful knowledge for decision makers to be proactive in flood management, hazard mitigation measures, and land use regulations. Recently, developers, planners, local governments, and other agencies have acquired flood susceptibility modeling as an important step in identifying flood-prone areas that need to be studied in more detail to prevent future flooding.