Landslide Prediction in Mountainous Terrain Using Weighted Overlay Analysis Method: A Case Study of Al Figrah Road, Al-Madinah Al-Munawarah, Western Saudi Arabia

Abstract

This study applies the Weighted Overlay Analysis (WOA) method integrated with GIS to assess landslide susceptibility along Al Figrah Road in Al-Madinah Al-Munawarah, western Saudi Arabia. Seven key conditioning factors, elevation, slope, aspect, drainage density, lithology, soil type, and precipitation were integrated using high-resolution remote sensing data and expert-assigned weights. The output susceptibility map categorized the region into three zones: low (93.5 million m²), moderate (271.2 million m²), and high risk (33.1 million m²). Approximately 29% of the road corridor lies within the low-risk zone, 48% in the moderate zone, and 23% in the high-risk zone. Ten critical sites with potential landslide activity were detected along the road, correlating well with the high-risk zones on the map. Structural weaknesses in the area, such as faults, joints, foliation planes, and shear zones in both igneous and metamorphic rock units, were key contributors to slope instability. The findings offer practical guidance for infrastructure planning and geohazard mitigation in arid, mountainous environments and demonstrate the applicability of WOA in data-scarce regions.

1. Introduction

Landslides rank among the most devastating natural hazards globally, responsible for considerable fatalities and economic setbacks each year [1]. These geomorphological processes involve the downward displacement of rock, soil, or debris due to gravitational forces [2], and are frequently initiated by triggers such as seismic disturbances, prolonged rainfall, snowmelt, or anthropogenic activities like urban development, deforestation, and excavation [3]. Their sudden occurrence and unpredictability pose significant risks to both communities and infrastructure [4].

On the international stage, high-impact landslides such as the 2017 Xinmo landslide in China, the 2018 Hokkaido event in Japan, and the 2020 Kerala disaster in India have highlighted the urgent need for improved predictive and mitigation strategies [5,6,7]. Within Saudi Arabia, landslides have become an increasing concern, particularly in mountainous zones undergoing accelerated urbanization and infrastructure development. In 2019, Regions such as Asir, Al-Baha, and Jazan have experienced slope failures that disrupted transportation networks and damaged infrastructure [8,9]. Al-Madinah Al-Munawarah, in particular, has witnessed landslide events that threaten its strategic and religious significance, especially along roads like Al Figrah that traverse complex geological terrains [10,11,12]. Despite these recurring events, existing studies in Saudi Arabia have primarily focused on general hazard mapping or rely heavily on qualitative assessments. There is a notable lack of detailed, quantitative, GIS-based susceptibility models specifically tailored to arid, mountainous environments like Al-Madinah. Additionally, field-validated studies integrating multiple geospatial layers using a transparent methodological framework are rare.

The application of Geographic Information Systems (GIS) and remote sensing has greatly improved the ability to identify and monitor landslide-prone regions [13]. Diverse modeling approaches, ranging from statistical techniques to machine learning and heuristic methods, have been employed in landslide risk analysis [14,15]. Among these, the Weighted Overlay Analysis (WOA) method is particularly valued for its capacity to incorporate both qualitative insights and quantitative measurements, enabling decision-makers to evaluate complex spatial problems through structured analysis [16,17].

This study addresses the identified research gaps by applying a GIS-based WOA method to assess landslide susceptibility along the Al Figrah Road. The key innovations of this study include: (i) the integration of seven conditioning factors tailored to the arid mountainous terrain of western Saudi Arabia; (ii) the use of expert-driven weighting supported by field insights; and (iii) the validation of model outputs with ten known landslide occurrences, enhancing the credibility of the approach. The objectives of this study are to: identify and analyze key environmental factors influencing landslides in the study area; generate a landslide susceptibility map using the WOA technique; and validate the results through field survey and spatial analysis. The results aim to provide decision-makers with reliable tools for risk management, particularly in road development across similarly vulnerable terrains in Saudi Arabia and beyond.

2. Materials and Methods

2.1. Study Area

This study focuses on a segment along Al Figrah Road, a vital transportation link located in the western region of Saudi Arabia. The study area lies between the geographical coordinates 38°57′ E, 24°18′ N and 39°07′ E, 24°20′ N (Figure 1). The route traverses varied topography, with elevation levels ranging from 795 m at the valley base to approximately 1866 m at the summit of Al Figrah Mountain. These pronounced elevation shifts present logistical challenges during field investigations, particularly in narrow segments of the road where access is limited. Additionally, elevation variations influence local microclimates, such as temperature and precipitation levels, which in turn play a significant role in determining environmental stability and landslide risk [18,19,20,21].

Al Figrah Road functions as an important shortcut connecting inland regions to coastal cities along the Red Sea, such as Yanbu. Geologically, the Al Figrah Mountain belongs to the Red Sea Mountains, formed by tectonic uplift of Precambrian rocks due to rifting along the Red Sea margin. Over geological time, erosional processes have produced steep slopes along mountain flanks. The lithology of the area is diverse, with dominant formations including coarse-grained granites composed mainly of quartz, feldspar, and mica. In addition, localized basaltic rocks appear in volcanic zones. Metamorphic units, particularly schist and gneiss, are also widespread, characterized by strong foliation and banding. Sedimentary deposits such as sandstone and alluvial material are found predominantly in valleys, reflecting depositional processes shaped by water transport and weathering over millions of years [22].

The climate in this region mirrors that of much of Saudi Arabia, marked by significant diurnal and seasonal thermal variability. Winter months, particularly January, see average high temperatures around 24 °C, while in summer, particularly June, temperatures can reach average highs of 43 °C (Climate-Data.org). Rainfall is generally scarce and primarily occurs during the winter season, with historical precipitation ranging between 3 mm and 90 mm over the past 30 years [23]. Orographic effects caused by the region’s complex topography promote precipitation in elevated areas, which then drains through the network of seasonal streams. Most drainage systems extend eastward, while mountainous formations define the area’s northern, western, and southern borders.

2.2. Datasets

The integration of remote sensing and GIS techniques provides an efficient framework for mapping and analyzing landslide susceptibility [24,25]. This approach consolidates data from various sources, including geological, geomorphological, hydrological, and meteorological domains, to develop a spatial representation of risk. Key datasets used in this study include Sentinel-1 satellite imagery and a 12.5 m resolution Digital Elevation Model (DEM), which were utilized to derive essential terrain parameters such as slope gradient, drainage density, and aspect orientation [24]. Additional raster datasets, such as lithology, soil types, and precipitation, were also incorporated, with some obtained from the Ministry of Environment, Water, and Agriculture.

A comprehensive landslide risk model was developed by integrating several key conditioning factors: slope, elevation, aspect, precipitation, drainage density, geology, and soil characteristics. These variables were processed using the Weighted Overlay Analysis (WOA) technique, a robust spatial analysis tool that enables the evaluation of multiple contributing factors by assigning weights to each thematic layer based on its relative importance. Through this process, a composite susceptibility map is generated, illustrating spatial variations in landslide risk intensity [11]. WOA has been widely adopted in domains such as land-use management, environmental assessment, and natural hazard analysis due to its flexibility in handling diverse datasets.

We adopted the weight percentages directly from the Al-Hada Road landslide susceptibility study [11]. During field trips, the authors compared and ranked eight GIS layers based on their influence on landslides (35% drainage density, 30% slope, 10% elevation, and 5% each for precipitation, lithology, soil, aspect, and land use/land cover) and iteratively validated these weights against 11 documented landslide events to ensure regional robustness. Because our study area shares similar topographical, geological, and climatic conditions with Al-Hada Road, we applied these field-validated weights unchanged in our WOA scoring.

In this study, the weighted thematic layers were integrated within a GIS platform, producing a detailed landslide susceptibility map. Each factor was normalized and assigned a specific weight reflecting its contribution to landslide occurrence, allowing for the aggregation of multiple risk criteria. The resulting output classifies the study area into zones of high, moderate, and low landslide susceptibility, offering a valuable spatial tool for risk evaluation.

The final step involved validating the susceptibility map through targeted field surveys (Figure 2). These site visits focused on zones identified as high-risk to verify the alignment between model predictions and on-ground conditions. Field validation is a critical component of the modeling process, as it ensures the practical accuracy of the risk map and provides an opportunity to refine the model based on observed data [26].

3. Results and Discussion

Despite the implementation of various mitigation strategies along Al Figrah Road, such as improved drainage networks, slope reinforcement methods, and early warning mechanisms, landslide occurrences have continued to affect the region periodically since the road’s initial construction (Figure 3). These recurring events underscore the limitations of structural solutions alone and highlight the necessity for advanced spatial assessments to better understand and predict landslide-prone areas. Remote sensing data play a pivotal role in enhancing spatial analyses for landslide susceptibility [27]. By integrating high-resolution geospatial datasets, researchers can gain a more nuanced understanding of terrain dynamics and environmental conditions that contribute to slope instability. Accurate landslide risk assessment requires the use of multiple data layers, each reflecting a specific conditioning factor influencing the likelihood of slope failure.

3.1. Drainage Density

Drainage density serves as a critical indicator of the extent and concentration of stream networks within a given surface area. It reflects the hydrological behavior of the landscape and plays an essential role in guiding land-use planning, water resource management, and environmental protection strategies [28]. Mathematically, drainage density is expressed as the ratio of the total length of streams and rivers within a region to its surface area, commonly measured in kilometers per square kilometer (km/km²). This parameter is closely linked to terrain characteristics, with higher drainage density often associated with steep slopes, impermeable surfaces, and limited infiltration capacity, all of which can elevate the risk of landslides [29]. In the context of Al Figrah Road, the convergence of surface runoff and ephemeral streams during periods of heavy rainfall increases slope instability, thereby contributing to landslide occurrences (Figure 4). To facilitate spatial analysis, the study area was classified into five distinct drainage density zones: 0–0.90, 1.00–1.50, 1.60–2.10, 2.20–2.80, and 2.90–4.50 km/km² (Figure 4a). This stratification enables the identification of zones with higher hydrological stress and informs the overall susceptibility assessment for landslide risk in the region.

3.2. Topography

Elevation is a fundamental factor influencing landslide initiation and propagation, as it governs the gravitational potential energy within a landscape. Higher altitudes are typically associated with steeper slopes, which increase the likelihood of rapid debris movement and accumulation, ultimately elevating the risk of slope failure [30]. To capture and analyze these variations, a Digital Elevation Model (DEM) was utilized, offering a three-dimensional representation of the terrain surface and serving as a key tool in topographic analysis [31]. In the case of the Al Figrah Road region, the elevation ranges dramatically—from a maximum of 1866 m above mean sea level (AMSL) in the western mountainous zones to a minimum of 795 m AMSL in the eastern lowlands. For analytical clarity, the area’s elevation was divided into five categories: 795–980, 981–1153, 1154–1335, 1336–1524, and 1525–1866 m AMSL (Figure 4b). This classification aids in correlating altitude levels with susceptibility to landslide events.

3.3. Slope

Slope inclination, defined as the angle between the terrain surface and a horizontal plane, is a critical topographic variable in landslide susceptibility studies [32]. Examining slope gradients provides valuable insights into processes such as erosion, sediment transport, and the impact of anthropogenic activities on terrain stability [33]. Steeper slopes generally correlate with higher landslide risk due to their increased potential to accelerate mass movement and magnify its destructive impact [34,35]. In the Al Figrah Road study area, slope gradients exhibit substantial spatial variation, particularly in the mountainous western sector, where inclinations reach up to 73°. These high-slope regions have historically experienced frequent landslide activity. Furthermore, zones with limited vegetation cover are more vulnerable to slope failure, as the absence of root structures reduces natural soil cohesion. To facilitate spatial classification, slope gradients were categorized into five classes: 0–9°, 10–18°, 19–26°, 27–34°, and 35–73° (Figure 5a). This gradient-based zoning helps in pinpointing areas of elevated landslide susceptibility across the region.

3.4. Annual Precipitation

The climate of Al-Madinah Al-Munawwarah is predominantly arid, with most annual rainfall occurring between November and January. In the Al Figrah Road area, the southern region records the highest annual precipitation, reaching up to 52.9 mm. While the overall rainfall distribution across the study area remains relatively stable, even slight variations in annual precipitation can significantly affect slope stability. For analytical purposes, the precipitation data were segmented into five classes: 45.6–47.3 mm, 47.4–48.7 mm, 48.8–50.0 mm, 50.1–51.4 mm, and 51.5–52.9 mm per year (Figure 5b). Although the region receives modest rainfall, even limited amounts can infiltrate rock joints and fractures, elevating pore-water pressure and consequently reducing the shear strength of both soil and rock materials [36,37]. This weakening effect increases the susceptibility of slopes to failure, particularly in areas where other contributing factors, such as lithology or steep topography, are also present

3.5. Lithology and Soil

The geological composition and the degree of chemical and mechanical weathering significantly influence a slope’s susceptibility to landslides [38]. The lithological framework of Al-Madinah Al-Munawwarah encompasses a diverse range of igneous and metamorphic rocks, primarily granites, schists, and gneisses (Figure 6a), as outlined in the detailed geological mapping by Johnson and Kattan [39]. Recognizing the mineralogical and structural attributes of these rocks is essential for assessing landslide-prone zones [11]. Granitic rocks are especially prone to landslide activity due to their common occurrence of deep fracturing and intense weathering. These processes often lead to exfoliation and the formation of clay minerals, which reduce the mechanical integrity of the rock mass. Additionally, schists and gneisses exhibit pronounced foliation, a planar fabric that introduces inherent structural weaknesses. When these foliated planes are oriented parallel to the slope surface, they can serve as potential slip planes, thus significantly compromising slope stability [40]. Field observations along Al Figrah Road reveal widespread jointing and foliation within the bedrock (Figure 7 and Figure 8), which facilitate water infiltration and accelerate the weathering process. Such structural features create preferential pathways for moisture movement, further weakening the rock mass, particularly on steep slopes and denudational hills, thereby increasing the likelihood of landslide occurrence [41].

Soil characteristics significantly influence the occurrence of natural hazards such as floods and landslides [42]. Within the study area, the predominant soil types are Calciorthids and Torriorthents (Figure 6b). Calciorthids are calcium-rich soils that form distinct soil horizons with variable depth and exhibit textures ranging from sandy to loamy, particularly prevalent in the eastern segment of Al Figrah Road. On the other hand, Torriorthents typically develop on actively eroding slopes composed of resistant parent materials. These soils encompass a range of textures including loamy sand, fine sandy loam, sandy loam, loam, clay loam, and their respective gravelly variants [43,44].

A key factor influencing landslide susceptibility in the area is the presence of clay, primarily generated through the chemical weathering of granitic rocks. This clay contributes to the formation of potential slip surfaces, particularly in regions with steep slopes, increasing the likelihood of slope instability and collapse [45]. Due to its low permeability and high porosity, clayey soil retains significant amounts of water, which leads to a reduction in shear strength. This decreased shear strength ultimately compromises slope stability and elevates the risk of landslides [46,47].

3.6. Aspect

The aspect of a slope, defined as the compass direction it faces, is a critical topographic parameter in landslide susceptibility analysis. In mountainous terrain, slope aspect influences microclimatic conditions such as solar radiation, moisture retention, and vegetation cover, all of which affect the stability of slopes. More importantly, in areas with human interventions such as road cuts, understanding aspect is vital for identifying zones of increased vulnerability. When the structural planes of rock formations align with the direction of slope exposure, particularly along artificial cuts such as roads, the risk of slope failure is significantly heightened. This alignment enables movement along pre-existing planes of weakness, making landslides more likely to occur [48,49,50].

In this study, aspect analysis was employed to evaluate the dip direction of slopes along Al Figrah Road, with the goal of determining orientations more susceptible to landslides. The aspect data were derived from the digital elevation model and classified into five directional intervals: 0–67° representing northeast-facing slopes, 68–139° for northeast to southeast, 140–208° for southeast to southwest, 209–283° for southwest to northwest, and 284–360° for northwest-facing slopes (Figure 9). This classification aids in recognizing directional trends associated with landslide occurrences and enhances the effectiveness of risk mitigation strategies. The understanding of aspect, when combined with other terrain and geological parameters, allows for more comprehensive and reliable spatial predictions of landslide-prone zones.

3.7. Landslide Susceptible Zones and Validation

Field investigations revealed that the most prominent threats along Al Figrah Road stem from rapid downslope landslides, typically triggered by severe rainstorms. These intense precipitation events produce substantial surface runoff on the steep slopes of Al Figrah Mountain, mobilizing loose rock material and debris, which are subsequently transported downslope and onto the roadway. The problem is further compounded by anthropogenic activities such as road cuts, which significantly disturb the natural slope structure. These construction operations not only remove stabilizing materials but also alter the terrain geometry, weakening slope stability and increasing the likelihood of mass-wasting events such as landslides [51,52]. On-site observations further confirmed that, following periods of heavy rainfall, watercourses traverse the mountainous terrain and converge in concentrated drainage zones [53,54]. Notably, the western portion of the study area, characterized by steep slopes and denudational hills at higher elevations, experienced more frequent and intense landslides. This susceptibility is largely attributed to the geological composition of the area, including highly jointed igneous and metamorphic rocks with pronounced foliation, which act as planes of weakness and facilitate slope failure.

The Weighted Overlay Analysis (WOA) was employed to identify zones of varying landslide susceptibility within the study area.

Table 1. The thematic layers applied in the overlay analysis to determine landslide-prone areas, along with their corresponding influence percentages, classification categories, and ranking scales.

Parameter	Weight (%)	Classes	Rank
Slope (Degree)	32	0–9	1
		10–18	2
		19–26	3
		27–34	4
		35–73	5
Precipitation (mm)	5	45.6–47.3	1
		47.4–48.7	2
		48.8–50.0	3
		50.1–51.4	4
		51.5–52.9	5
Elevation (m)	10	795–980	1
		981–1153	2
		1154–1335	3
		1336–1524	4
		1525–1866	5
Lithology	5	Rhyolite, Andesite, and Clastic Sediments	4
		Granite	3
		Granite and Granodiorite	3
		Quaternary Deposits	1
		Rhyolite	3
Drainage density (km/km2)	38	0.0–0.9	1
		1.0–1.5	2
		1.6–2.1	3
		2.2–2.8	4
		2.9–4.5	5
Soil	5	Calciorthids	1
		Torriorthents	2
Aspect	5	0–67° (NE)	1
		68–139° (NE–SE)	2
		140–208° (SE–SW)	3
		209–283° (SW–NW)	4
		284–360° (NW)	5

presents the relative influences of each contributing factor, expressed as a weight percentage. Among the parameters analyzed, drainage density was found to be the most influential, with a weight of 38%, followed by slope at 32%. Elevation ranked third with a weight of 10%, while precipitation, lithology, soil, and aspect each contributed 5% to the overall susceptibility model. Figure 10 illustrates the spatial distribution of landslide risk based on the WOA output. The low-risk areas, shown in green, cover an estimated 93,552,333 m². Moderate-risk zones, highlighted in yellow, span approximately 271,180,722 m², while high-risk areas, marked in red, encompass 33,053,839 m². As expected, high-risk zones predominantly align with high-elevation regions characterized by steep gradients and fractured geological formations.

Excavating slopes for road construction and expansion is a frequent practice in hilly areas, and it has a notable influence on slope stability [55,56,57,58]. The current work employs remote sensing data and leverages the WOA analysis method in the GIS environment to detect past landslide occurrences and forecast likely future landslide locations. During the field trip, a total of 10 landslide spots were identified and utilized to verify the risk zones generated for the current study (Figure 11). This validation step is crucial for eliminating any erroneous results. All ten landslides occurred in the high-risk areas, confirming the findings of the WOA research. Regarding Al Figrah Road, out of the overall 26.2-km route, 7.5 km is classified as low risk, 12.6 km as moderate risk, and 6.1 km as high danger.

4. Conclusions

The current study utilized GIS-based Weighted Overlay Analysis (WOA) and remote sensing data to develop a landslide susceptibility map for the Al Figrah Mountain region in western Saudi Arabia. The analysis integrated multiple contributing factors, including drainage density, elevation, slope, precipitation, lithology, soil type, and aspect, to delineate the spatial distribution of landslide-prone zones within the study area. The results indicated that low-risk zones comprise 23.52% of the total area, moderate-risk zones account for 68.17%, and high-risk zones represent 8.31%. Field surveys along Al Figrah Road confirmed 10 historical landslides, predominantly occurring in areas characterized by high drainage concentration and steep slopes. Based on the risk assessment, the Al Figrah Road corridor was categorized into three segments: 28.63% of the route falls within low-risk zones, 48.09% within moderate-risk zones, and 23.28% within high-risk zones. This study serves as a valuable foundation for future landslide risk assessments and planning initiatives not only in Saudi Arabia but also in other regions with similar geological and topographical conditions. Continued research is necessary to evaluate the influence of climate change on landslide activity in the mountainous areas of the kingdom.