GIS-Based Assessment of Flash Flood Susceptibility around Thuwal-Rabigh Region, Saudi Arabia

doi:10.21203/rs.3.rs-4134684/v1

Download PDF

Research Article

GIS-Based Assessment of Flash Flood Susceptibility around Thuwal-Rabigh Region, Saudi Arabia

https://doi.org/10.21203/rs.3.rs-4134684/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Floods have become more frequent and severe across the globe, resulting in considerable loss of human lives, physical infrastructure, and livelihood. It is also applicable to Saudi Arabia, a country recognized for its arid climate, which has witnessed multiple flood events in the recent past; for example, Jeddah, a coastal Saudi city along the Red Sea experienced floods in 2007 and 2009. Flood susceptibility mapping and its spatio-temporal analysis is a vital component of flood mitigation projects as it identifies the most vulnerable regions of the project area based on physical properties. The present study intends to delineate the flood susceptibility zones, in the Thuwal-Rabigh region, located to the west of Saudi Arabia, by using a geographical information system (GIS) based multi-criteria decision analysis (MCDA) method called the analytical hierarchy process (AHP). The AHP technique was applied to compute the relative weights of nine flood governing factors namely digital elevation model (DEM), topographic wetness index (TWI), slope, normalized difference vegetation index (NDVI), land use/ land cover (LULC), rainfall, distance to waterways, distance to roads and drainage density. The final flood susceptibility map of the study area was obtained and reclassified into five zones (i.e., very low risk, low risk, medium risk, high risk, and very high risk) by using the overlay tool in ArcGIS. The results show that 39% of the study area has a very high to high risk of flooding. The model's sensitivity analysis demonstrates that the maps are reliable and Rainfall, TWI, DEM and slope appears to have a higher influence in the flood risk mapping of the study area. It was also found that Thuwal, city lies in an area of a very high flood risk zone (15%) and needs appropriate measures to ensure sustainable urban development in the future. This study helps to avoid further urban expansion in flood-prone areas and will assist decision-makers in implementing sustainable flood risk management plans.

Flood risk

Sustainable development

Analytical hierarchy process (AHP)

Geographical information system (GIS)

Saudi Arabia

Flooding is recognised as one of the most common natural disasters across the globe (Rahmati, Haghizadeh, & Stefanidis, 2016; Rahmati, Zeinivand, & Besharat, 2016). In the recent past, flash floods are occurring more frequently (Ushiyama, Kwak, Ledvinka, Iwami, & Danhelka, 2017), which is due to a number of factors, such as climate change, urbanisation, population increase, the partial or complete encroachment of roads or other structures into watercourses, and improperly sized rainwater drainage channels (Green, Parker, & Tunstall, 2000). They are caused by large amounts of rainfall that fall over a short period of time, exceeding the capacity of natural and artificial drainage systems to manage the excess water. Flash floods are characterized by their sudden onset, rapid water flow, and short duration, which can cause significant damage to human lives, properties, and livelihood (MURATA, OZAWA, KOZAI, ADACHI, & NISHIMURA, 2015; Xia, Falconer, Lin, & Tan, 2011). It is very difficult to predict when the water level will reach its crest hence; there is little time to issue warnings for local rapid-response teams to manage the potential risks and losses (Abdelkareem, 2017; Carpenter, Sperfslage, Georgakakos, Sweeney, & Fread, 1999; Collier, 2007). Flood catastrophes, including flash floods, have risen to the top of the list because of climate change impacts, urbanisation, and environmental degradation (Charlton, Fealy, Moore, Sweeney, & Murphy, 2006; Khosravi, Pourghasemi, Chapi, & Bahri, 2016; Scheuer, Haase, & Volk, 2017). In order to lessen the harmful effects of floods, access to reliable and up-to-date information is very essential. One of these crucial types of information is a flood susceptibility map (FSM) (Papaioannou et al., 2018).

Due to climate change and anthropogenic factors, flash floods have become more frequent in dry and semi-arid areas of the world. Flash floods are a recurring natural disaster in the Kingdom of Saudi Arabia (KSA), and their consequences can be catastrophic for both human life and property (Youssef, Sefry, Pradhan, & Alfadail, 2016). Flash floods in Saudi Arabia are driven by a combination of natural and human factors. (Youssef, Sefry, et al., 2016). The unique geography, geology, and climate of Saudi Arabia create conditions that make it highly susceptible to flash floods, which are sudden and intense floods that occur after heavy rainfall (Elhag & Abdurahman, 2020). The most common cause of flash floods in Saudi Arabia is heavy rainfall, particularly in the western and southwestern regions of the country (Elkhrachy, 2015; Subyani, 2011). The high-intensity rainfall during the wet season, coupled with the country's arid climate, makes it difficult for the soil to absorb water and triggers flash floods leading to soil erosion. Other factors that contribute to flash floods in Saudi Arabia include deforestation, rapid urbanization, and human activities such as mining and construction (Dano, 2020; Haider, Ghumman, Al-Salamah, Ghazaw, & Abdel-Maguid, 2019; Pham et al., 2020). Flash floods in the past have caused severe damages to infrastructure and agriculture in various parts of the country. During the period of 20 November 2009 to 1 January 2011, Jeddah City experienced two significant flash floods. The average amount of devastation caused by flash floods in the Jeddah region was roughly $3 billion USD in these two flood events. The consequences were disastrous, with significant infrastructure and property destruction (over 10,000 homes and 17,000 vehicles) killing 113 individuals (Youssef, Sefry, et al., 2016).

For sustainable development and effective management of floods, many studies have evaluated and forecasted flood risk in the past using climate change scenarios, flood risk modelling, and geomorphic and physical catchment characteristics (Abdelkareem, 2017; Arrighi, Brugioni, Castelli, Franceschini, & Mazzanti, 2018; Waqas et al., 2021; Zhang et al., 2015). In the flood risk assessments, the most important variables can be divided into the following three categories: (1) Physical and natural factors, including elevation, slope, aspect, slope curvature, lithology, topographic position index, and rainfall. (2) The hydrological factors include drainage density, distance to a river, topographic wetness index (TWI), and stream power index (SPI). (3) The third category includes man-made developments, such as land use and road distance (Al-Juaidi, Nassar, & Al-Juaidi, 2018; Costache, 2019; Darabi et al., 2019; Zhao, Pang, Xu, Yue, & Tu, 2018). Flood hazard mapping is a crucial component of flood risk management and sustainable development. It is an essential part of successful land use planning in floodplain management and flood mitigation measures (Hong & Abdelkareem, 2022). Water resource managers and decision-makers can use it to geographically identify the most hazardous locations for the implementation of flood mitigation plans (Bathrellos, Karymbalis, Skilodimou, Gaki-Papanastassiou, & Baltas, 2016; Billi, Alemu, & Ciampalini, 2015). Geographic information system (GIS) and remote sensing (RS) techniques have been utilized as efficient tools for flood risk assessment (SalehI & AI-Hatrushi, 2009). Since floods are multifaceted events with different spatio-temporal characteristics, ArcGIS represents a highly valuable computing tool to produce FSM by using logical and mathematical relations (Articte, 1995; Carver, 1991). The ArcGIS software can handle enormous amounts of spatial data and combine various types of data to predict and identify new water sources (Yin et al., 2018). The processing of RS data in ArcGIS allows the gathering of new datasets and reveals high flood-risk regions by generating FSM(Abdelkareem, 2017; Khosravi et al., 2016; Vojtek & Vojteková, 2019). When recent topographic maps or ground surveys are unavailable or out-of-date, remote sensing photographs constitute a valuable source of extensive geographical data. Moreover, soil and geological maps required for flood estimation can be produced from RS imageries in ArcGIS environment. In addition, RS images taken after a flood can also be used to calibrate the built floodplains within a hydrological basin (Dawod, Mirza, & Al-Ghamdi, 2012).

Multi-criteria decision analysis (MCDA) is one of the important techniques used for analysing flood susceptibility areas (Hu, Cheng, Zhou, & Zhang, 2017; Souissi et al., 2020). MCDA can use a variety of weighting procedures to prioritise the relative importance of the selected flood-contributing variables (Mahmoud & Gan, 2018; Tang, Zhang, Yi, & Xiao, 2018). One of the widely used MCDA methods is based on a weighted linear approach to identify flood-prone areas. The weight linear combination method entails multiplying each flood-controlling variable by its weight, and the sum of all values gives the final FSM (Drobne & Lisec, 2009). The most preferred technique used for defining weights is the knowledge-driven analytical hierarchy process (AHP). It uses pairwise comparisons to assess the degree to which one selection outranks another (T. Saaty, 1980). The AHP technique integrated weighted overlay method in ArcGIS, has been used successfully to simulate various environmental challenges and flood risk models (Abdekareem, Al-Arifi, Abdalla, Mansour, & El-Baz, 2022; Danumah et al., 2016; Waqas et al., 2021). The MCDA technique has been employed in numerous forecasting studies, as it provides solutions for complex problems based on hierarchical ordering criteria (Arulbalaji, Padmalal, & Sreelash, 2019; Priya et al., 2022; Sahu, Wagh, Mukate, Kadam, & Patil, 2022).

The purpose of this research is to set up a GIS/RS -based model in order to identify flash flood-prone locations around the Thuwal-Rabigh area, Saudi Arabia and thereby aid in future flood risk management. We use weighted overlay analysis method in the ArcGIS tool and widely used AHP technique. In this approach, a hierarchical tree with numerous levels is developed, and selected criteria are divided into several sub-criteria. This model's input is spatial data of nine flood causing-factors that include digital elevation model (DEM), TWI, slope, rainfall, land use/land cover (LULC) map, normalized difference vegetation index (NDVI), distance to waterways, distance to roads, drainage density. The scientific insights generated from this effort can help planners, engineers, and relevant government entities to take appropriate decisions to prevent and mitigate future floods in the study area and ensure sustainable development.

2.1 Study Area

The boundary of the study area lies in the west of Saudi Arabia between 22.0° − 22.9° N latitudes and 39.0°–39.9° E longitudes and constitutes an area of 10993.6 km² encompassing Thuwal and Rabigh cities (Fig. 1). The study area is flanked on the west by the Red Sea and on the east by various mountain chains, the highest of which is roughly 1600 metres above mean sea level. This area contains numerous wadi (Valley) systems that are largely covered by wadi deposits (suspended sand and mud) and represent almost 70% of the total surface (Vincent, 2008). These wadis are natural systems formed by erosion of channels by cutting into the Precambrian-age rocks of various lithology’s (Grainger, 2007). The majority of waterways dissect the area, beginning with these mountains (general split), which are about 40 to 50 km inland, and progress towards the coastal plain of the red sea (Fig. 1). The region is marked by heavy rainfall, which often causes flash floods in the low-lying areas. Water flowing into the Red Sea through the wadi basins can endanger coastal cities, neighbourhoods, and other developmental projects. The study area encompasses a number of important strategic developments, such as Thuwal City, King Abdullah Economic City, and King Abdullah University of Science and Technology (KAUST).

Thuwal is a small city located, along the coast of the red sea, about 80 kilometres north of Jeddah and 100 kilometres south of Yanbu. Thuwal's proximity to Mecca also makes it an important area for providing logistical support during the Hajj season. Thuwal's King Abdulaziz International Airport is one of the busiest airports in the region, and it handles a significant amount of air traffic during the Hajj season. Thuwal contributes immensely to the country's economic, scientific, and technological development. Its strategic location and infrastructure make it a hub for trade. Therefore, flash flood susceptibility modelling in this area will be very helpful to identify possible risks to important infrastructure, which has been raised in recent years as a result of increased urbanisation and climate change.

2.2 Data Collection

Different techniques were found in the literature for the selection of flood governing factors to assess flood susceptibility. For Instance, the research carried out by Ogato et al. (Ogato, Bantider, Abebe, & Geneletti, 2020), analysed elevation, slope, land use, drainage density, soil type, and rainfall as flood risk factors. Whereas, Pham et al. (Pham et al., 2020) chose slope, aspect, elevation, distance from faults, rainfall, lithology, soil type, and land use as identifying criteria for flash floods mapping. Likewise, Kazakis et al. (Kazakis, Kougias, & Patsialis, 2015) picked rainfall intensity, geology, distance from the drainage network, slope, and elevation as parameters for identifying flood danger zones. On the other hand, Youssef et al. (Youssef, Pradhan, & Sefry, 2016) used geology, curvature, land use, soil drain, and distance from the stream as parameters for assessing flash flood vulnerability. After a thorough literature review, here, we have used nine flood-controlling factors in a similar scenario. The details of the selected variables, their data types and sources are presented in (Table 1).

Table 1

Data, sources, and map types
	Data item	GIS data Type	Source	Map developed
1	Digital Elevation Model (DEM)	Raster	USGS (2020)	Elevation Slope TWI Drainage density Waterways
2	Landsat 9 Imagery	Raster	USGS (2021)	Land use and land cover (LULC) Normalized digital vegetation index (NDVI)
3	Rainfall distribution	Raster	CRU TS (V. 4.05)	Rainfall
4	Waterways	Raster	DEM	Distance to waterways
5	Roads	Vector, Shapefile	MapCruzin	Distance to roads

In this study, both satellite radar and optical imageries were collected to characterise the flash flood-prone area in the Thuwal-Rabigh region, Saudi Arabia. The Shuttle Radar Topography Mission (SRTM), digital elevation model (DEM) data (30 m spatial resolution) are accessed and downloaded from the United States Geological Survey (USGS) website (https://earthexplorer.usgs.gov/). The SRTM-DEM data was used to map the topographic features and compute the catchment characteristics. Land-use/land-cover (LULC) classification was also prepared using optical data. Landsat 9 images, having the least amount of cloud cover, were downloaded from USGS to develop the LULC categories by utilizing data management tools in ArcGIS.

2.3 Methods

The methodology implemented in this study is summarized and presented in Fig. 2. The objective of the current study is to uncover flood-prone areas in the specified region using satellite imagery in ArcGIS environment coupled with AHP technique.

2.3.1 Flood-Causing Factors

Defining the relevant flood control factors and delineation of thematic layers is the first step in developing the FSM. For flash flood mapping in this study, nine specific layers were chosen including DEM, TWI, slope, rainfall, LULC, NDVI, distance to waterways, distance to roads, and drainage density (Fig. 2). As shown in Fig. 2, all the subject layers were developed from DEM by using spatial analyst tools in ArcGIS except rainfall, distance to roads, and LULC maps. LULC maps were created using Landsat 9 imagery. The FSM was produced by the superimposition of reclassified layers after testing the AHP model's pairwise matrix consistency. The flood-contributing factors were selected by reviewing the literature and physical characteristics of the study area. The details of flood-controlling factors are given below:

2.3.1.1. Digital Elevation Model (DEM)

DEM is the most basic type of digital topographic representation. It serves as a baseline map for TWI, slope, flow accumulation, stream network, and drainage density in flood susceptibility studies. It is used to compute the pathways and direction of the overflow, as well as the depth of the water level (Ogato et al., 2020). Floods are less likely to occur in regions with relatively high altitudes. Figure 3a depicts high-altitude areas in green and low-altitude areas in dark pink. The raster map of elevation was produced in the ArcGIS environment using DEM data and subdivided into five categories.

2.3.1.2. Slope

The amount of sharpness and the degree of inclination relative to the horizontal plane constitute the slope (Gigović, Pamučar, Bajić, & Drobnjak, 2017). It is a critical marker for identifying areas where flash flooding poses a high risk. The topographic slope, which is always expressed as a percentage or degree, has a significant effect on the flow velocity and flood energy. Consequently, regions with steep slopes are more prone to floods due to high flow velocity. While areas with level terrain and a minimal regional slope are relatively less susceptible to floods. The slope raster map was produced after assessing the slope factor from DEM, by using the built-in slope tool, in ArcGIS software. Subsequently, the slope raster was then reclassified into five sub-values.

2.3.1.3. Topographic Wetness Index (TWI)

The TWI can be used as an alternative to the traditional way of identifying locations that are capable of producing floods. TWI equation be defined mathematically as Eq. 1 (Pourali, Arrowsmith, Chrisman, Matkan, & Mitchell, 2016)

$\text{T}\text{W}\text{I}=\text{ln}\left(\frac{{\alpha }}{\text{t}\text{a}\text{n} {\beta }}\right)$ Eq. (1)

Here, “α” denotes the area with an upward slope per unit contour length through a point and tanβ denotes the slope angle there. This secondary topographical characteristic shows the wetness state of the catchment area. TWI is now frequently utilized in the modeling of biological processes, vegetation patterns, and studies of forests (Pourali et al., 2016). In this work, DEM, TWI, and flow accumulation maps were used to extract TWI using the map algebra tool in ArcGIS by selecting the flow accumulation layer as α and slope as β. The resultant TWI raster was divided into five sub-groups (Fig. 3b).

2.3.1.4. Normalized Difference Vegetation Index (NDVI)

NDVI is used to quantify vegetation greenness and estimate vegetation density. While areas with little or no vegetation are more vulnerable to floods and vice versa. Dense vegetation helps to avoid or reduce flooding (Shrestha et al., 2017). NDVI is calculated as a ratio between the red (R) and near-infrared (NIR) band values (Fig. 3f). The Landsat-9, Band 5 (NIR) and Band 4 (R), images were used to compute NDVI values by using Eq. 2 via map algebra in ArcGIS.

NDVI = (NIR - R) / (NIR + R). Eq. (2)

2.3.1.5. Land Use and Land Cover (LULC)

Land use is one of the main characteristics, used to determine places that are likely to experience flooding. Soil infiltration rate is influenced by different LULC and contributes differently to triggering food hazards (Kourgialas & Karatzas, 2011). For instance, forest, vegetation, and water-retaining crops favour water infiltration whereas, grazing, forest loss, growing urbanisation, farming, and impermeable surfaces lead to an increase in the frequency and severity of floods. From the official US Geological Survey website, the Landsat 9 images with the least amount of cloud cover were obtained. This raw dataset is imported into ArcGIS and bands are combined using the composite bands' tool. Finally, a supervised image classification procedure was then used to locate different land cover classes along with parallel verification through the google earth application. We produced five categories of land cover, including sea, built area, rocks, bare soil, and waterways (Fig. 3e).

2.3.1.6. Rainfall

Rainfall is considered as one of the crucial factors in flood susceptibility mapping. Among the nine factors listed, rainfall appears to have a direct and immediate impact on floods. To develop the mean rainfall map using ArcGIS, a high-resolution gridded (30’ × 30’) monthly rainfall dataset, for the years (2011–2020), was downloaded from the Climatic Research Unit gridded Time Series (CRU TS4.06) website through the following link https://crudata.uea.ac.uk (accessed date: 10 April 2023). The CRU is a widespread climate dataset that covers the whole world's landmasses, with the exception of Antarctica, on 0.5° latitude by 0.5° longitude grid. It is produced by extrapolating monthly climate anomalies from vast networks of weather station readings across the globe (Harris, Osborn, Jones, & Lister, 2020). The average annual rainfall map for the study area was created using the inverse distance weighting (IDW) method in the ArcGIS by using the spatial analyst tool. Finally, the rainfall map was reclassified into five sub-groups in order to visualise low, very low, moderate, high, and very high rainfall areas (Fig. 3d).

2.3.1.7. Distance from Waterways

Distance from waterways is another crucial flood-determining factor considering that rivers and the nearby lands are the primary conduits for floods. Lower areas near waterways are more vulnerable to flooding than regions that are comparatively at higher elevations and are located at a distance from the rivers. For this study, the distance from the waterways map was prepared from the stream-flow layer by manipulating DEM in ArcGIS. Subsequently, a rasterized map was created (Fig. 3g) by using a buffer tool and divided into five classes. It represents that floods are less likely to occur in locations that are distant from the water paths.

2.3.1.8. Distance from Roads

Distance from the road is also a controlling factor in flood susceptibility mapping. Roads and adjacent impermeable surfaces have a significant impact on flood levels by reducing the potential for infiltration and making them a source of surface runoff (Tehrany, Jones, & Shabani, 2019). In ArcGIS, the euclidean distance tool was used to calculate the distance from the road layer. This technique is very useful in a variety of scenarios when the data being used is an insufficient collection of distances and the output is a set of euclidean space points that realise the given distances. The final map is then reclassified into five classes which represent a low probability of floods in areas distant from the roads (Fig. 3h).

2.3.1.9. Drainage Density

Drainage density is influenced by permeability, vegetation, slope, etc. The drainage density factor is computed by dividing the total length of all waterways in the drainage catchment by the total area of the catchment (Harini et al., 2018). The stream layer was developed using DEM in order to generate a drainage density map by applying spatial analyst tools in ArcGIS. Subsequently, the drainage density was re-classified into five classes (Fig. 3i).

2.3.2. Analytic Hierarchy Process (AHP) and FSM

AHP, an MCDM technique, originally developed by Saaty (T. L. Saaty, 1980) was utilized and described as follows. In the first step, each flood factor was assigned a weight based on the literature review and experts’ opinions. Next to that, sub-classes were also given weights and ranked based on their contributions to the likelihood of a flood. A pair-wise comparison matrix was then used to compare the contribution of these flood conditioning factors. The chosen nine flood conditioning factors were arranged hierarchically (Table 3) in which each row's value describes the relationship between two flood factors. Each factor was assigned an arithmetic value between 1 and 9, depending on how important it was in relation to the other factor with which it was paired. If the arithmetic number is 1, then both elements are equally important. Whereas, 9 denotes that a row component is more important than the corresponding column factor. The consistency ratio (CR) was calculated by computing the principal eigenvalue (E_max) via the eigenvector technique. Finally, the consistency index (CI), was calculated through the following Eq. 3.

$$\text{C}\text{I}=\frac{\left({\text{E}}_{\text{m}\text{a}\text{x}}-\text{n}\right)}{\left(\text{n}-1\right)}$$

Where “n” is the number of conditioning factors and “E_max” is the mean value of the consistency vector. The AHP states that the CR shouldn't exceed 0.1 (Pham et al., 2020). The computed value of CR was 0.03, confirming the acceptable consistency of the pairwise matrix.

When the relative importance of each element was established, the ArcGIS overlay tool was used to produce a map of flood risk areas, which multiply each reclassification factor by its weight and add the results using Eq. 4.

$$\text{F}\text{S}=\sum _{\text{i}=1}^{\text{n}}{\text{W}}_{\text{i}}{\text{S}}_{\text{i}}$$

Whereas FS stands for flood susceptibility, W_i is the weight of factor i, and S_i represents classes of flood susceptibility for a given factor i.

3.1 Application of Analytic Hierarchy Process (AHP)

AHP technique was applied for the ranking and weighting of above-mentioned flood governing factors. These nine flood governing factors were then converted into reclassified raster maps, each showing five classes, in ArcGIS (Fig. 3). These rasterized maps when combined with the AHP method can provide a flood risk map in a particular region. The relative weight of each flood factor in a pair-wise comparison matrix is shown in Table 3. The weights of the flood governing factors were derived from the normalised matrix (Table 4), using a geometric weighting scheme to calculate the percent share of each factor (Table 5). The weighted sum values were calculated for each flood factor and divided by their weights (WSV/FFW) to obtain the principle eigenvector (E_max) value of the matrix. E_max is the average of the ratio of weighted sum value and flood factor weight. Finally, to calculate the CR of the matrix, the CI was computed by using Eq. 3 and divided with the standard random index (RI) value that was taken from Table 2.

Table 2

Random ranking index standard matrix (Ogato et al., 2020)
No.	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45	1.49	1.51	1.48	1.56	1.57	1.58

Table 3

Ranking of nine flood factors to build a pairwise comparison matrix
Flood Factor	TWI	Elevation	Slope	Rainfall	LULC	NDVI	Distance to water	Distance to roads	Drainage density
TWI	1	¼	⅓	⅐	⅕	3	¼	3	⅕
Elevation	4	1	2	½	2	7	3	7	3
Slope	3	½	1	¼	4	5	3	7	2
Rainfall	7	2	4	1	5	7	5	7	3
LULC	5	½	¼	⅕	1	6	3	3	⅓
NDVI	⅓	⅐	⅕	⅐	⅙	1	⅙	2	⅓
Distance to water	4	⅓	⅓	⅕	⅓	6	1	7	½
Distance to road	⅓	⅐	⅐	⅐	⅓	½	⅐	1	¼
Drainage density	5	⅓	½	⅓	3	3	2	4	1

Table 4

Normalized matrix for the selected flood-causing factors
Flood Factors	TWI	Elevation	Slope	Rainfall	LULC	NDVI	Distance to water	Distance to roads	Drainage density	Factor weight	Percent share
TWI	0.03	0.05	0.04	0.05	0.01	0.08	0.01	0.07	0.02	0.04	4.1
Elevation	0.13	0.19	0.24	0.17	0.13	0.18	0.17	0.17	0.28	0.19	18.5
Slope	0.10	0.10	0.12	0.09	0.25	0.13	0.17	0.17	0.19	0.15	14.6
Rainfall	0.24	0.38	0.47	0.34	0.31	0.18	0.28	0.17	0.28	0.30	29.6
LULC	0.17	0.10	0.03	0.07	0.06	0.16	0.17	0.07	0.03	0.10	9.5
NDVI	0.01	0.03	0.02	0.05	0.01	0.03	0.01	0.05	0.03	0.03	2.6
Distance to water	0.13	0.06	0.04	0.07	0.02	0.16	0.06	0.17	0.05	0.08	8.4
Distance to road	0.01	0.03	0.02	0.05	0.02	0.01	0.01	0.02	0.02	0.02	2.1
Drainage density	0.17	0.06	0.02	0.11	0.19	0.08	0.11	0.10	0.09	0.10	10.4

Table 5

Estimation of weighted sum values, factors weight, and percent share of the selected flood-governing factors
Flood Factors	TWI	Elevation	Slope	Rainfall	LULC	NDVI	Distance to water	Distance to roads	Drainage density	Weight sum value	Ratio WSV/FFW	Factor weight	Percent share
TWI	0.04	0.05	0.05	0.04	0.02	0.09	0.02	0.06	0.02	0.39	9.58	0.04	4.06
Elevation	0.16	0.19	0.30	0.15	0.20	0.21	0.24	0.14	0.30	1.89	10.21	0.19	18.58
Slope	0.12	0.10	0.15	0.08	0.40	0.15	0.24	0.14	0.20	1.57	10.78	0.15	14.60
Rainfall	0.28	0.38	0.60	0.30	0.50	0.21	0.40	0.14	0.30	3.11	10.49	0.30	29.64
LULC	0.20	0.10	0.04	0.06	0.10	0.18	0.24	0.06	0.03	1.01	10.57	0.10	9.51
NDVI	0.01	0.03	0.03	0.04	0.02	0.03	0.01	0.04	0.03	0.24	9.34	0.03	2.61
Distance to water	0.16	0.06	0.05	0.06	0.03	0.18	0.08	0.14	0.05	0.82	9.69	0.08	8.41
Distance to road	0.01	0.03	0.02	0.04	0.03	0.02	0.01	0.02	0.03	0.21	9.69	0.02	2.14
Drainage density	0.20	0.06	0.03	0.10	0.30	0.09	0.16	0.08	0.10	1.12	10.74	0.10	10.45

Table 6

Relative weights, classes, and corresponding ranks of the selected flood-governing factors
Flood factors	Relative weight	Reclassified raster	Ranks	Total weightage
TWI	4.06	-9.1 to -5.1	1	4.06
		-5 to -2.7	2	8.12
		-2.6 to 0.82	3	12.18
		0.83 to 4.8	4	16.24
		4.9 to 14	5	20.3
Elevation	18.58	-21 to 150	5	92.9
		160 to 380	4	74.32
		390 to 650	3	55.74
		660 to 980	2	37.16
		990 to 1600	1	18.58
Slope	14.60	0 to 9.2	5	73
		9.3 to 22	4	58.4
		23 to 38	3	43.8
		39 to 56	2	29.2
		57 to 260	1	14.6
Rainfall	29.64	71 to 81	1	29.64
		82 to 91	2	59.28
		92 to 99	3	88.92
		100 to 110	4	118.56
		110 to 120	5	148.2
LULC	9.51	Sea	1	9.51
		Built Area	3	28.53
		Rocks	4	38.04
		Bare Soil	2	19.02
		Water Ways	5	47.55
NDVI	2.61	-0.42 to -0.09	5	13
		-0.09 to 0.03	4	10.4
		0.03 to 0.05	3	7.8
		0.06 to 0.09	2	5.2
		0.09 to 0.54	1	2.6
Distance to water	8.41	0–40	5	42.05
		50–125	4	33.64
		125–235	3	25.23
		235–395	2	16.82
		395–700	1	8.41
Distance to road	2.14	0–40	5	10.65
		40–105	4	8.52
		105–170	3	6.39
		170–265	2	4.26
		265–500	1	2.13
Drainage density	10.45	0 to 38	1	10.44
		39 to 87	2	20.88
		88 to 120	3	31.32
		130 to 140	4	41.76
		150 to 210	5	52.2
Sum	100.00

Principle eigenvector value (E_max) = 10.12 (Average of WSV/FFW)

By using Eq. 2

Consistency Index (CI) = 10.12–9 / 8 = 0.14

Consistency ratio (CR) = CI/RI = 0.096

CR < 0.1 Hence, a pair-wise matrix is consistent

3.2 Flood Governing Factors

In the present investigation, nine flood governing factors in the raster format with the same pixel value were analysed in ArcGIS. After applying the AHP technique and obtaining a reasonable result, classes of each raster were reclassified into five sub-categories and ranked from 1 to 5 as shown in Fig. 3 (a-i).

Digital Elevation Model (DEM) stands as an indispensable determinant in the assessment of flood risk within any given geographical expanse. Lower DEM values suggest relatively lower surface altitudes and the maximum values represent relatively higher altitudes. Figure 3a is the reclassified DEM map of the study area, the north-eastern region shows the maximum elevation values ranging from 990 to 1600m (light yellow colour) while the western region of the study area represents the lowest values ranges between − 21 to 150 (blue colour). In the DEM map, blue colour constitutes the maximum portion, where most of the population and infrastructure exist, showing higher possibility of inundation in this area. TWI has a direct connection with flooding, indicating that areas with higher TWI values are more prone to flooding and vice versa. In Fig. 3b and Table 6, the minimum TWI value in the eastern region ranges between − 9.1 to -5.1 (green colour), and the maximum value is 4.9 to 14 (dark purple colour) in the western region of study area. Figure 3c depicts five classes of slopes representing lower slope values in green and higher slope values in orange. The eastern portion of the study location is a mountainous area thereby indicating a steeper slope (Fig. 1). Whereas, the western part mostly has a very low slope ranges between 0 ͦ to 9 ͦ (Table 6). Consequently, rainwater flows down the slope and pools in the lower region and generates flash flooding. In this case, slope classes with low values were assigned a higher rating and represent a higher tendency for flash floods. Rainfall is considered a key contributor to an area's flooding, which directly influences the occurrence of flash floods (Talha, Maanan, Atika, & Rhinane, 2019). Figure 3d shows five classes of rainfall representing minimum rainfall of 71 mm/year in the western region (brown colour) and maximum rainfall of 120 mm/year (orange colour) in the north-eastern region of the study area. The rainfall map shows heavy rainfall in the north-eastern region as compared to the other parts of the study area. This is a mountainous area with more rainfall events due to the topographically driven convective rains. According to Zaharia (Zaharia, Costache, Prăvălie, & Ioana-Toroimac, 2017), locations with slopes greater than 15° do not support the accumulation of water. Thus, intense rainfall over the rock-mountains promptly flows downward due to the steep slope and increases the likelihood of flash floods in the low-lying areas. The reclassified Land Use and Land Cover (LULC) map, as depicted in Fig. 3e, elucidates the predominant land features within the eastern precinct of the study area, primarily characterized by mountainous terrain (yellow segment). This geological configuration, marked by impervious rock formations, serves as the impetus for flood initiation, thereby engendering downstream repercussions in the western expanse (orange segment). It is reported that areas with bare soil and rock accelerate the peak discharge and considered highly vulnerable to floods (Al-Taani, Al-husban, & Ayan, 2023). Thus, higher ratings were assigned to the classes representing the bare soil and rocks in order to show high tendency for flash floods. The red portion in the western region of the NDVI map shows less vegetation and the dark purple portion in the east represent relatively more vegetation (Fig. 3f). The lowest value of NDVI ranges from − 0.42 to -0.09 and the highest value ranges from 0.09 to 0.54 (Table 6). It is evident from overall low NDVI values that most of the study region is an arid region. Therefore, relative contribution of NDVI in flash flood was kept only 2.5% (Table 6). The areas with low NDVI values are at more risk of flood because less vegetation fails to absorb rainfall, resulting in increased surface runoff. Conversely, high NDVI values represent thick vegetation thus less susceptible to high runoff. Reclassified five classes of distance from waterways can be seen in Fig. 3g. In general, the flooding is concentrated near the river and thus inundates the nearby areas. The dark brown and green colour in the map represent more distance from waterways hence, this area is less susceptible to flash flooding. Whereas, dark blue and yellow portion indicates proximity to the river and thus are at a higher risk of flash flooding. Similarly, in the case of distance from the roads map in Fig. 3h, the grey portion is at lower risk with rating 1 as compared to the orange portion with a rating value of 5 represents higher risk of the flash flood. Lastly, the minimum drainage density value is 0 and the maximum value is 210 (Table 6). The lowest drainage density is represented with dark blue and the maximum value with dark grey colour in the Fig. 3i. It can be observed that the western part of the study area has high drainage density and depicts the high susceptibility of flash floods with a given ranking value of 5, and vice versa for minimum drainage density values.

3.3 Flood Susceptibility Map of Study Area

According to the results, the CR value of the pairwise matrix was within the limit because and less than 0.10 (T. Saaty, 1980). This indicates that pairwise comparison exhibits a sensible consistency dimension, which allows us to proceed further. The weighted sum of all flood factors was computed and the final map of the flood-prone area was developed by using an overlay tool in the ArcGIS as shown in Fig. 4. Subsequently, FSM was categorized into five classes, ranging from very low to high flood potential by using the natural breaks (Jenks) grading procedure. These five classes are represented as very high risk (red colour), high risk (Pink colour), medium risk (green colour), low risk (yellow colour), and very low risk (blue colour).

In terms of the percentage of flood susceptibility classes by area, the very high risk (13%) and very low risk (9%) classes had the lowest area percentages. The highest area percentage (31%) was identified in the moderate flood susceptibility class. On the other hand, the high flood susceptibility class constitutes 26% of the land while the low flood susceptibility class accounts for 21%. Moreover, 39% of the study area exhibits very high to high flood susceptibility based on the specified flood governing factors and their associated weights. According to the FSM analysis in ArcGIS, 15% of the zones with extremely high flood risk are located close to major streams. The area with a very high flood risk level is distributed throughout and slightly higher in the western part of the study region, which possesses a low elevation and a low slope. Whereas, the eastern region exhibits relatively low flood risk, as it is characterized by non-porous rocks with more rainfall, high elevation, and steep slope that exceeds 50◦. This area favours the generation of flash floods, which becomes more visible during its downhill journey, in the low-lying areas, towards the Red Sea. It is noteworthy that Thuwal City is in the zone of very high flood risk area (Fig. 4b). Figure 4b is the magnified view of FSM of the study area, to observe the impact of the flood on Thuwal City. Gentle slope and low-elevation land beneath the Red Sea favours the high velocity downhill flow from high-elevation rocks situated in the eastern part of the study area.

This study provides baseline information in Thuwal and Rabigh regions, which must be taken into account for flood management practices. It is important to consider that the final flash flood susceptibility map does not consider sea flooding along the Red Sea banks. It is, therefore, recommended that the final flood susceptibility map must be combined with the Red Sea flood inundation map in order to obtain a full overview of the flood risk in the area to ensure sustainable future planning and management. After 2009 and 2011 floods in the Jeddah city, KSA government has taken significant steps to mitigate the flash floods by constructing drainage channels in different regions of Saudi Arabia. The Jeddah flood mitigation project was designed in order to control and divert storm water towards to the sea. The whole project consists of 8 Packages with 5 Dams and 3 Channels/Drainage and outfalls. The New Drainage system runs at a stretch of 36 kilometers from southeast to northwest of the City of Jeddah diving at a head of 27 meters from the highest point. It is recommended to plan such projects in the adjacent areas like Thuwal and Rabigh, possessing similar local conditions and likelihood of floods due to the impact of climate change.

3.4 Sensitivity Analysis

It’s recommended to perform a sensitivity analysis after each round of criteria analysis to ensure and verify the stability of the results due, to possible uncertainties in expert judgments (Dano et al., 2019). The evaluation of single parameter sensitivity aims to understand the impact of factors on the flood hazard index. There are two methods for conducting sensitivity analyses; (i) adjusting weights based on expert evaluations and (ii) assigning weights to all influencing factors. In this study we employed an approach to test the reliability of the flood susceptibility model. This involved conducting a parameter one at a time (OAT) sensitivity analysis and exploring equal weight sensitivity. As shown in Fig. 5(e) when equal weights were assigned to all flood causing factors the resulting FSM exhibited minute variations from the original map. The variations in the percentage of areas at risk are obvious and consistent with the original flood patterns, though they are still small-scale indicating the reliability of original map.

Table 7 compares the proportion of each class's flood risk area between the original FSM and the FSM produced by employing equal weights for each flood criterion. In the original assessment, the "Moderate" flood risk class has the highest susceptibility, whereas the "FSM by Equal Weights" places the "Moderate" and "Very High" classes at the top in terms of susceptibility. This indicates that the equal weights approach has re-prioritized the factors contributing to flood susceptibility, potentially assigning more significance to those factors that were not as heavily weighted in the original assessment.

Table 7

Relative weights, classes, and corresponding ranks of the selected flood-governing factors
S.no	Flood risk class	Original FSM map (%)	FSM by equal weights (%)
1	Very low	9	9
2	Low	21	19
3	Moderate	31	37
4	High	26	22
5	Very high	13	14

The second approach implements the OAT (One-At-a-Time) method, where all the criteria values are modified in a systematic manner to assess their influence on the FSMs. These modified weights are then inputted into GIS software to generate multiple FSMs. The OAT method, widely used and straightforward, is employed to individually adjust the weight of a single criterion and observe the resulting impact on the outputs (Dano et al., 2019). Hence, a total of 9 scenarios were executed in the ArcGIS to determine the flood susceptible zones in the study area. Figure 6 (a-d) portrays the Flood Susceptibility Maps (FSMs) of the study region, revealing notable discrepancies among them. Specifically, subfigures denoted as a, b, c, and d correspond respectively to Rainfall, Topographic Wetness Index (TWI), Digital Elevation Model (DEM), and Slope maps, each exhibiting substantial divergences when compared with the initial FSM. Meanwhile, the FSMs associated with TWI, Land Use and Land Cover (LULC), and Normalized Difference Index (NDI) showed comparatively less variance in relation to the original Flood Risk Map. Therefore, the flood controlling factors of Rainfall, TWI, DEM, and Slope appear to contribute more significantly in the flood susceptibility investigation of the study area. The Flood Susceptibility Maps (FSMs) derived through the utilization of sensitivity analysis by means of two distinct methodologies offer valuable insights into the verification of the precision of the initial FSM. Such an approach is highly advocated within studies characterized by restricted or unavailable field data, as it contributes to enhancing the credibility of the outcomes.

Flash flooding is a natural phenomenon that poses a significant threat to human lives and physical properties around the globe. The occurrence of flash flood events becomes more frequent and severe due to climate change impacts. Thus, flood susceptibility mapping is necessary for effective town planning and management in order to ensure sustainable urban growth. Flood risk occurrence can be better comprehended in the specified region with the use of spatial analysis in the GIS environment. This study utilizes a GIS-based AHP technique to generate flood susceptibility map by employing a number of geo-climatic variables derived from RS data. ArcGIS is an effective tool to identify and evaluate flood risk zones and the factors contributing to the intensification of flash flood risk. Nine thematic layers of flood-controlling factors were analysed in ArcGIS including DEM, Slope, TWI, NDVI, LULC, rainfall, distance to waterways, distance to roads, and drainage density. The final output map shows that the flood risk zones of the study area are located in the western region of the study area. The flash flood map was reclassified into five classes ranging from very low, low, moderate, high, and very high risk, in terms of flood susceptibility. Analysis revealed that 39% of the total study area is under very high-risk and high-risk zones and the western region is relatively at higher risk due to low elevation and slope. It is noted that most of the human settlement lies in the same region, a zone of very high flood risk, including Thuwal City. Even though the average rainfall in the western coastal regions of Saudi Arabia is minimal, high rainstorms in the nearby eastern mountains can cause flash floods, such as those that were experienced in 2009, 2010– 2011, and 2017 around Jeddah, Saudi Arabia. It should be noted that the allocated rainfall weighting of 29% in this study may not be realistic in present times, but it can meet that level in the future as a result of climate change. FSM obtained through sensitivity analysis by two methods gives useful insights to check the accuracy of original FSM and recommended in such kind of studies where field data is limited/not available. The findings of this investigation can help relevant decision-makers and urban planners to take appropriate flood management actions and cope with the climate-induced challenges to ensure sustainable development in the study area.

Funding: Not Applicable

Institutional Review Board Statement: Not applicable.

Data Availability Statement: Data used in this research are available in the manuscript.

Acknowledgments: The researchers would like to thank Sami G. Alghamdi for his guidance in the study

Conflicts of Interest: The authors declare no conflict of interest.

Ethical Statement: We ensure that all research is conducted in accordance with ethical principles of the journal of Natural Hazards. All research is conducted with an ethic of respect for cultures, communities, the individual/person, and independent knowledge.

Author Contributions Statement

Abdul Shakoor and Mohammad Arif designed and wrote the main article and prepared figures and tables
Abdul Razzaq Ghumman supervised the study
Ghufran Ahmad Pasha Performed a Critical Analysis
Amjad Masood provided important data

Abdekareem, M., Al-Arifi, N., Abdalla, F., Mansour, A., & El-Baz, F. (2022). Fusion of Remote Sensing Data Using GIS-Based AHP-Weighted Overlay Techniques for Groundwater Sustainability in Arid Regions. Sustainability, 14(13), 7871.
Abdelkareem, M. (2017). Targeting flash flood potential areas using remotely sensed data and GIS techniques. Natural Hazards, 85(1), 19-37.
Al-Juaidi, A. E., Nassar, A. M., & Al-Juaidi, O. E. (2018). Evaluation of flood susceptibility mapping using logistic regression and GIS conditioning factors. Arabian Journal of Geosciences, 11, 1-10.
Al-Taani, A., Al-husban, Y., & Ayan, A. (2023). Assessment of potential flash flood hazards. Concerning land use/land cover in Aqaba Governorate, Jordan, using a multi-criteria technique. The Egyptian Journal of Remote Sensing and Space Science, 26(1), 17-24.
Arrighi, C., Brugioni, M., Castelli, F., Franceschini, S., & Mazzanti, B. (2018). Flood risk assessment in art cities: The exemplary case of Florence (Italy). Journal of Flood Risk Management, 11, S616-S631.
Articte, P. N. (1995). Raster procedures for M ulti-criteria/multi-0biective decisions. Photogramm. Eng. Remote Sens, 61(5), 539-547.
Arulbalaji, P., Padmalal, D., & Sreelash, K. (2019). GIS and AHP techniques based delineation of groundwater potential zones: a case study from southern Western Ghats, India. Scientific reports, 9(1), 2082.
Bathrellos, G., Karymbalis, E., Skilodimou, H., Gaki-Papanastassiou, K., & Baltas, E. (2016). Urban flood hazard assessment in the basin of Athens Metropolitan city, Greece. Environmental Earth Sciences, 75, 1-14.
Billi, P., Alemu, Y. T., & Ciampalini, R. (2015). Increased frequency of flash floods in Dire Dawa, Ethiopia: Change in rainfall intensity or human impact? Natural Hazards, 76, 1373-1394.
Carpenter, T., Sperfslage, J., Georgakakos, K., Sweeney, T., & Fread, D. (1999). National threshold runoff estimation utilizing GIS in support of operational flash flood warning systems. Journal of Hydrology, 224(1-2), 21-44.
Carver, S. J. (1991). Integrating multi-criteria evaluation with geographical information systems. International Journal of Geographical Information System, 5(3), 321-339.
Charlton, R., Fealy, R., Moore, S., Sweeney, J., & Murphy, C. (2006). Assessing the impact of climate change on water supply and flood hazard in Ireland using statistical downscaling and hydrological modelling techniques. Climatic change, 74(4), 475-491.
Collier, C. (2007). Flash flood forecasting: What are the limits of predictability? Quarterly Journal of the Royal Meteorological Society: A journal of the atmospheric sciences, applied meteorology and physical oceanography, 133(622), 3-23.
Costache, R. (2019). Flash-flood Potential Index mapping using weights of evidence, decision Trees models and their novel hybrid integration. Stochastic Environmental Research and Risk Assessment, 33(7), 1375-1402.
Dano, U. L. (2020). Flash flood impact assessment in Jeddah City: An analytic hierarchy process approach. Hydrology, 7(1), 10.
Dano, U. L., Balogun, A.-L., Matori, A.-N., Wan Yusouf, K., Abubakar, I. R., Said Mohamed, M. A., . . . Pradhan, B. (2019). Flood susceptibility mapping using GIS-based analytic network process: A case study of Perlis, Malaysia. Water, 11(3), 615.
Danumah, J. H., Odai, S. N., Saley, B. M., Szarzynski, J., Thiel, M., Kwaku, A., . . . Akpa, L. Y. (2016). Flood risk assessment and mapping in Abidjan district using multi-criteria analysis (AHP) model and geoinformation techniques,(cote d’ivoire). Geoenvironmental Disasters, 3, 1-13.
Darabi, H., Choubin, B., Rahmati, O., Haghighi, A. T., Pradhan, B., & Kløve, B. (2019). Urban flood risk mapping using the GARP and QUEST models: A comparative study of machine learning techniques. Journal of Hydrology, 569, 142-154.
Dawod, G. M., Mirza, M. N., & Al-Ghamdi, K. A. (2012). GIS-based estimation of flood hazard impacts on road network in Makkah city, Saudi Arabia. Environmental Earth Sciences, 67, 2205-2215.
Drobne, S., & Lisec, A. (2009). Multi-attribute decision analysis in GIS: weighted linear combination and ordered weighted averaging. Informatica, 33(4).
Elhag, M., & Abdurahman, S. G. (2020). Advanced remote sensing techniques in flash flood delineation in Tabuk City, Saudi Arabia. Natural Hazards, 103(3), 3401-3413.
Elkhrachy, I. (2015). Flash flood hazard mapping using satellite images and GIS tools: a case study of Najran City, Kingdom of Saudi Arabia (KSA). The Egyptian Journal of Remote Sensing and Space Science, 18(2), 261-278.
Gigović, L., Pamučar, D., Bajić, Z., & Drobnjak, S. (2017). Application of GIS-interval rough AHP methodology for flood hazard mapping in urban areas. Water, 9(6), 360.
Grainger, D. J. (2007). The geologic evolution of Saudi Arabia: a voyage through space and time: Saudi geological survey.
Green, C. H., Parker, D. J., & Tunstall, S. M. (2000). Assessment of flood control and management options. WCD Thematic reviews. World Commission on Dams Secretariat, South Africa.
Haider, H., Ghumman, A. R., Al-Salamah, I. S., Ghazaw, Y., & Abdel-Maguid, R. H. (2019). Sustainability evaluation of rainwater harvesting-based flood risk management strategies: a multilevel decision-making framework for arid environments. Arabian Journal for Science and Engineering, 44, 8465-8488.
Harini, P., Sahadevan, D. K., Das, I., Manikyamba, C., Durgaprasad, M., & Nandan, M. (2018). Regional groundwater assessment of Krishna River basin using integrated GIS approach. Journal of the Indian Society of Remote Sensing, 46, 1365-1377.
Harris, I., Osborn, T. J., Jones, P., & Lister, D. (2020). Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset. Scientific data, 7(1), 109.
Hong, Y., & Abdelkareem, M. (2022). Integration of remote sensing and a GIS-based method for revealing prone areas to flood hazards and predicting optimum areas of groundwater resources. Arabian Journal of Geosciences, 15(1), 114.
Hu, S., Cheng, X., Zhou, D., & Zhang, H. (2017). GIS-based flood risk assessment in suburban areas: A case study of the Fangshan District, Beijing. Natural Hazards, 87, 1525-1543.
Kazakis, N., Kougias, I., & Patsialis, T. (2015). Assessment of flood hazard areas at a regional scale using an index-based approach and Analytical Hierarchy Process: Application in Rhodope–Evros region, Greece. Science of the Total Environment, 538, 555-563.
Khosravi, K., Pourghasemi, H. R., Chapi, K., & Bahri, M. (2016). Flash flood susceptibility analysis and its mapping using different bivariate models in Iran: a comparison between Shannon’s entropy, statistical index, and weighting factor models. Environmental monitoring and assessment, 188, 1-21.
Kourgialas, N. N., & Karatzas, G. P. (2011). Flood management and a GIS modelling method to assess flood-hazard areas—a case study. Hydrological Sciences Journal–Journal des Sciences Hydrologiques, 56(2), 212-225.
Mahmoud, S. H., & Gan, T. Y. (2018). Multi-criteria approach to develop flood susceptibility maps in arid regions of Middle East. Journal of Cleaner Production, 196, 216-229.
MURATA, M., OZAWA, H., KOZAI, T., ADACHI, N., & NISHIMURA, H. (2015). Post Aswan High Dam flash floods in Egypt: Causes, consequences and mitigation strategies. Bulletin of Center for Collaboration in Community Naruto University of Education, 29, 173-186.
Ogato, G. S., Bantider, A., Abebe, K., & Geneletti, D. (2020). Geographic information system (GIS)-Based multicriteria analysis of flooding hazard and risk in Ambo Town and its watershed, West shoa zone, oromia regional State, Ethiopia. Journal of Hydrology: Regional Studies, 27, 100659.
Papaioannou, G., Efstratiadis, A., Vasiliades, L., Loukas, A., Papalexiou, S. M., Koukouvinos, A., . . . Kossieris, P. (2018). An operational method for flood directive implementation in ungauged urban areas. Hydrology, 5(2), 24.
Pham, B. T., Avand, M., Janizadeh, S., Phong, T. V., Al-Ansari, N., Ho, L. S., . . . Bozchaloei, S. K. (2020). GIS based hybrid computational approaches for flash flood susceptibility assessment. Water, 12(3), 683.
Pourali, S., Arrowsmith, C., Chrisman, N., Matkan, A., & Mitchell, D. (2016). Topography wetness index application in flood-risk-based land use planning. Applied Spatial Analysis and Policy, 9, 39-54.
Priya, U., Iqbal, M. A., Salam, M. A., Nur-E-Alam, M., Uddin, M. F., Islam, A. R. M. T., . . . Rak, A. E. (2022). Sustainable groundwater potential zoning with integrating GIS, remote sensing, and AHP model: a case from North-Central Bangladesh. Sustainability, 14(9), 5640.
Rahmati, O., Haghizadeh, A., & Stefanidis, S. (2016). Assessing the accuracy of GIS-based analytical hierarchy process for watershed prioritization; Gorganrood River Basin, Iran. Water resources management, 30, 1131-1150.
Rahmati, O., Zeinivand, H., & Besharat, M. (2016). Flood hazard zoning in Yasooj region, Iran, using GIS and multi-criteria decision analysis. Geomatics, Natural Hazards and Risk, 7(3), 1000-1017.
Saaty, T. (1980). The analytical hierarchy process McGraw-Hill. New York.
Saaty, T. L. (1980). TheAnalyticHierarchyProcess. McGrawhill, Juc. New York.
Sahu, U., Wagh, V., Mukate, S., Kadam, A., & Patil, S. (2022). Applications of geospatial analysis and analytical hierarchy process to identify the groundwater recharge potential zones and suitable recharge structures in the Ajani-Jhiri watershed of north Maharashtra, India. Groundwater for Sustainable Development, 17, 100733.
SalehI, A., & AI-Hatrushi, S. (2009). Torrential flood hazards assessment, management, and mitigation, in Wadi Aday, Muscat area, Sultanate Of Oman, a GIS and RS approach. Egyptian Journal of Remote Sensing and Space Sciences, 12.
Scheuer, S., Haase, D., & Volk, M. (2017). Integrative assessment of climate change for fast-growing urban areas: Measurement and recommendations for future research. PLoS One, 12(12), e0189451.
Shrestha, R., Di, L., Eugene, G. Y., Kang, L., SHAO, Y.-z., & BAI, Y.-q. (2017). Regression model to estimate flood impact on corn yield using MODIS NDVI and USDA cropland data layer. Journal of Integrative Agriculture, 16(2), 398-407.
Souissi, D., Zouhri, L., Hammami, S., Msaddek, M. H., Zghibi, A., & Dlala, M. (2020). GIS-based MCDM–AHP modeling for flood susceptibility mapping of arid areas, southeastern Tunisia. Geocarto International, 35(9), 991-1017.
Subyani, A. M. (2011). Hydrologic behavior and flood probability for selected arid basins in Makkah area, western Saudi Arabia. Arabian Journal of Geosciences, 4(5), 817.
Talha, S., Maanan, M., Atika, H., & Rhinane, H. (2019). Prediction of flash flood susceptibility using fuzzy analytical hierarchy process (Fahp) algorithms and Gis: a study case of guelmim region In Southwestern of Morocco. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 42, 407-414.
Tang, Z., Zhang, H., Yi, S., & Xiao, Y. (2018). Assessment of flood susceptible areas using spatially explicit, probabilistic multi-criteria decision analysis. Journal of Hydrology, 558, 144-158.
Tehrany, M. S., Jones, S., & Shabani, F. (2019). Identifying the essential flood conditioning factors for flood prone area mapping using machine learning techniques. Catena, 175, 174-192.
Ushiyama, T., Kwak, Y., Ledvinka, O., Iwami, Y., & Danhelka, J. (2017). Interdisciplinary approach for assessment of continental river flood risk: a case study of the Czech Republic. Paper presented at the EGU General Assembly Conference Abstracts.
Vincent, P. (2008). Saudi Arabia: an environmental overview: CRC Press.
Vojtek, M., & Vojteková, J. (2019). Flood susceptibility mapping on a national scale in Slovakia using the analytical hierarchy process. Water, 11(2), 364.
Waqas, H., Lu, L., Tariq, A., Li, Q., Baqa, M. F., Xing, J., & Sajjad, A. (2021). Flash flood susceptibility assessment and zonation using an integrating analytic hierarchy process and frequency ratio model for the Chitral District, Khyber Pakhtunkhwa, Pakistan. Water, 13(12), 1650.
Xia, J., Falconer, R. A., Lin, B., & Tan, G. (2011). Numerical assessment of flood hazard risk to people and vehicles in flash floods. Environmental Modelling & Software, 26(8), 987-998.
Yin, H., Shi, Y., Niu, H., Xie, D., Wei, J., Lefticariu, L., & Xu, S. (2018). A GIS-based model of potential groundwater yield zonation for a sandstone aquifer in the Juye Coalfield, Shangdong, China. Journal of Hydrology, 557, 434-447.
Youssef, A. M., Pradhan, B., & Sefry, S. A. (2016). Flash flood susceptibility assessment in Jeddah city (Kingdom of Saudi Arabia) using bivariate and multivariate statistical models. Environmental Earth Sciences, 75(1), 12.
Youssef, A. M., Sefry, S. A., Pradhan, B., & Alfadail, E. A. (2016). Analysis on causes of flash flood in Jeddah city (Kingdom of Saudi Arabia) of 2009 and 2011 using multi-sensor remote sensing data and GIS. Geomatics, Natural Hazards and Risk, 7(3), 1018-1042.
Zaharia, L., Costache, R., Prăvălie, R., & Ioana-Toroimac, G. (2017). Mapping flood and flooding potential indices: a methodological approach to identifying areas susceptible to flood and flooding risk. Case study: the Prahova catchment (Romania). Frontiers of Earth Science, 11, 229-247.
Zhang, D.-w., Quan, J., Zhang, H.-b., Wang, F., Wang, H., & He, X.-y. (2015). Flash flood hazard mapping: A pilot case study in Xiapu River Basin, China. Water Science and Engineering, 8(3), 195-204.
Zhao, G., Pang, B., Xu, Z., Yue, J., & Tu, T. (2018). Mapping flood susceptibility in mountainous areas on a national scale in China. Science of the Total Environment, 615, 1133-1142.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

GIS-Based Assessment of Flash Flood Susceptibility around Thuwal-Rabigh Region, Saudi Arabia

Status:

Version 1

Abstract

Figures

1. Introduction

2. Material and Methods

2.1 Study Area

2.2 Data Collection

2.3 Methods

2.3.1 Flood-Causing Factors

2.3.1.1. Digital Elevation Model (DEM)

2.3.1.2. Slope

2.3.1.3. Topographic Wetness Index (TWI)

2.3.1.4. Normalized Difference Vegetation Index (NDVI)

2.3.1.5. Land Use and Land Cover (LULC)

2.3.1.6. Rainfall

2.3.1.7. Distance from Waterways

2.3.1.8. Distance from Roads

2.3.1.9. Drainage Density

2.3.2. Analytic Hierarchy Process (AHP) and FSM

3. Results and Discussion

3.1 Application of Analytic Hierarchy Process (AHP)

3.2 Flood Governing Factors

3.3 Flood Susceptibility Map of Study Area

3.4 Sensitivity Analysis

4. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1