Previous Article in Journal
Effects of Climate Change and Changes in Land Use and Cover on Water Yield in an Equatorial Andean Basin
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Hydrology
Volume 11
Issue 10
10.3390/hydrology11100158
hydrology-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Urban Flood Vulnerability Assessment in Freetown, Sierra Leone: AHP Approach

by
Abdulai Osman Koroma
1,2,
Mohamed Saber
3,* and
Cherifa Abdelbaki
1,4,5
1
Pan African University, Institute of Water and Energy Sciences Including Climate Change, University of Tlemcen, P.O. Box 119, Tlemcen 13000, Algeria
2
Department of Geology, Fourah Bay College, University of Sierra Leone, Freetown, Sierra Leone
3
Disaster Prevention Research Institute (DPRI), Kyoto University, Goka-sho, Uji City 611-0011, Kyoto, Japan
4
Department of Hydraulics, Faculty of Technology, University of Tlemcen, P.O. Box 230, Tlemcen 13000, Algeria
5
Eau et Ouvrages dans Leur Environnement Laboratory, University of Tlemcen, P.O. Box 119, Tlemcen 13000, Algeria
*
Author to whom correspondence should be addressed.
Hydrology 2024, 11(10), 158; https://doi.org/10.3390/hydrology11100158
Submission received: 1 August 2024 / Revised: 10 September 2024 / Accepted: 18 September 2024 / Published: 25 September 2024
Browse Figures
Review Reports Versions Notes

Abstract

:
This study presents a comprehensive flood vulnerability assessment for Freetown, Sierra Leone, spanning the period from 2001 to 2022. The objective of this research was to assess the temporal and spatial changes in the flood vulnerability using Geographic Information System (GIS) tools and AHP-based Multi-Criteria Decision-Making (MCDM) analysis. This study identified the flood-vulnerable zones (FVZs) by integrating critical factors such as the rainfall, NDVI, elevation, slope, drainage density, TWI, distance to road, distance to river, and LULC. The analysis reveals that approximately 60% of the study area is classified as having medium to high vulnerability, with a significant 20% increase in the flood risk observed over the past two decades. In 2001, very-high-vulnerability zones covered about 68.84 km2 (10% of the total area), with high-vulnerability areas encompassing 137.68 km2 (20%). By 2020, very-high-vulnerability zones remained constant at 68.84 km2 (10%), while high-vulnerability areas decreased to 103.26 km2 (15%), and medium-vulnerability zones expanded from 206.51 km2 (30%) in 2001 to 240.93 km2 (35%). The AHP model-derived weights reflect the varied significance of the flood-inducing factors, with rainfall (0.27) being the most critical and elevation (0.04) being the least. A consistency ratio (CR) of 0.068 (< 0.1) confirms the reliability of these weights. The spatial–temporal analysis highlights the east and southeast regions of Freetown as consistently vulnerable over the years, while infrastructure improvements in other areas have contributed to a general decrease in very-high-vulnerability zones. This research highlights the urgent need for resilient urban planning and targeted interventions to mitigate future flood impacts, offering clear insights into the natural and human-induced drivers of the flood risk for effective hazard mitigation and sustainable urban development.

1. Introduction

Flood vulnerability assessment has become increasingly crucial, particularly within urban areas, due to the escalating impacts of climate change and rapid urbanization [1,2]. Flood vulnerability indices are essential tools for evaluating urban environments’ susceptibility to flooding, incorporating social, economic, environmental, and infrastructural dimensions [3,4]. Social vulnerability might include population density and the distribution of vulnerable groups, while economic vulnerability assesses potential financial losses. Environmental and infrastructural vulnerabilities consider land use, drainage capacities, and natural barriers [5,6].
In recent years, advancements in flood vulnerability assessments have stressed the need for robust methodologies that account for non-stationary approaches and real-time data sources to address climate change and increasing extreme weather events [7]. These non-stationary approaches enable researchers to account for changing environmental conditions, such as increased rainfall frequencies due to global warming, and have proven particularly useful in capturing the dynamic nature of flood risks. While our current analysis does not incorporate non-stationary models, we acknowledge this as a critical area for future research [8,9,10]. Freetown, Sierra Leone’s bustling capital, faces similar challenges, grappling with environmental pressure such as floods induced by rainfall variability and land use alterations over the past two decades [11].
Comparable to regions like Bhutan, Freetown encounters pronounced flood hazards worsened by climatic unpredictability, highlighting the need for a nuanced understanding of its flood vulnerability dynamics [12,13]. The vulnerability of Freetown to floods is compounded by factors such as unpredictable rainfall patterns and shifts in land use, posing significant challenges to community resilience and the infrastructure integrity [14]. Despite being acknowledged as a pressing issue, there remains a lack of comprehensive assessments that holistically evaluate flood vulnerability over extended periods, hindering effective risk management and mitigation efforts [15].
Over the years, several methodologies have been employed to assess flood risks, with the Geographic Information System (GIS) and Multi-Criteria Decision-Making (MCDM) frameworks emerging as widely used tools [16]. Among these, the Analytical Hierarchy Process (AHP) has proven to be highly effective at quantifying the relative importance of various flood-inducing factors, enabling a structured and consistent evaluation of flood vulnerability [17].
Recent advancements in flood vulnerability assessment have emphasized the need for context-specific approaches, particularly in regions where urbanization and land use changes rapidly increase the flood risks [18]. In many studies, AHP has been integrated with GIS to develop spatial vulnerability maps, allowing for the identification of high-risk areas and the prioritization of mitigation efforts. One study [19] systematically reviewed multidimensional flood vulnerability indices and highlighted the relevance of AHP in managing flood risks in diverse environments. Similarly, in the context of urban flooding, multi-criteria approaches combining socio-economic, environmental, and infrastructural factors have been adopted to provide comprehensive assessments [20].
AHP, as a decision-making tool, has evolved over the years and is now increasingly being applied in conjunction with modern data sources like remote sensing and machine learning techniques to improve the accuracy of flood risk predictions [21].
This body of work highlights the need for context-specific approaches to flood vulnerability assessment, which is particularly relevant for rapidly growing urban areas like Freetown. The integration of climate change projections into flood vulnerability assessments is particularly critical, as future flood risks are likely to be worsened by shifts in precipitation patterns and extreme weather events [22]
The significance of this research lies in its multifaceted approach to evaluating flood vulnerability, encompassing both climatic and human-induced factors. By leveraging advanced analytical techniques and longitudinal data analysis, this study aims to provide valuable insights into Freetown’s flood vulnerability landscape, empowering stakeholders with the knowledge necessary to enact sustainable and resilient urban development practices [23,24,25].
Flood vulnerability assessment has been the focus of numerous studies worldwide, each contributing valuable insights into the methodology and application of vulnerability indices [19,26]. Studies across Southeast Asia [18] have demonstrated the impact of rapid urbanization on the flood risk, while research in Europe has highlighted the importance of integrating climate change projections into flood risk assessments [27]. In Latin America, the focus has often been on the socio-economic impacts of flooding on low-income communities [28]. In Africa, the impacts of extreme events are seen to occur simultaneously due to the social and economic sectors and are accelerated by dry-land physical and environmental drivers like land use, topography, and proximity to rivers [29].
The novelty of this research lies in its focused analysis of Freetown’s flood vulnerability, using a combination of GIS tools and vulnerability indices fitted to the local context. Unlike previous studies, which often consider management factors, this study isolated climatic and land use variables, providing a clearer understanding of their direct impacts on the flood risk. This study aimed to assess the temporal and spatial changes in the flood vulnerability using Geographic Information System (GIS) tools and AHP-based Multi-Criteria Decision-Making (MCDM) analysis. This research endeavors to contribute to the overarching goal of fostering resilience and sustainability in the face of increasing flood risks within urban environments, epitomizing a crucial step towards safeguarding the well-being of Freetown’s populace and its urban infrastructure [30].

2. Materials and Methods

2.1. Study Area

The study area, Freetown, is the capital city of Sierra Leone, located on the western coast of Africa. As shown in Figure 1, it is situated on a natural harbor, known as the Sierra Leone Estuary, which opens into the Atlantic Ocean [30]. The city is built on hilly terrain, with the coastline stretching along the south and west [31]. Freetown is positioned at approximately 8.5° N latitude and 13.2° W longitude, covering an area of about 688 square kilometers [32].
Freetown’s climate is marked by two primary seasons: the rainy season from May to October and the dry season from November to April [33]. During the rainy season, the city experiences heavy rainfall, particularly between July and September, with monthly precipitation often exceeding 400 mm (15.7 inches). Conversely, the dry season is characterized by much lower precipitation, with monthly totals frequently dropping below 50 mm (2 inches). The relative humidity in Freetown averages between 75% and 85%, with higher levels during the rainy season [34]. The prevailing northeast trade winds contribute to the seasonal weather patterns, bringing cooler and drier conditions during the dry season and more humid, stormy weather during the rainy period. The city has an annual average temperature of 27 °C (81 °F), with a temperature range from 22 °C to 32 °C for the night and day, respectively [35].

2.2. Criterion Selection

The flood vulnerability index (FVI) was articulated using an indicator-based analytical hierarchy method, a widely employed approach for flood system vulnerability assessment globally [12,36]. The datasets used for this study can be seen in Table 1 below. This method integrates indicators such as the rainfall, NDVI, elevation, slope, drainage density, TWI, distance to road, distance to river, and land use land cover (LULC). According to [29], the AHP process involves several steps inspired by previous studies. Firstly, the main goal was identified, which involved weighting or assigning scores to the FV indicators based on their hazard and impact levels. Next, criteria were formulated, categorized into social, physical/economic, and environmental domains [12,37]. Subsequently, sub-criteria were prioritized using the Analytic Hierarchy Process (AHP) pairwise comparison method, including factors such as the rainfall, NDVI, elevation, slope, drainage density, TWI, distance to road, distance to river, and land use land cover (LULC) [13,17,38]. Finally, the weighting of the criteria was aggregated to produce a vulnerability scale ranging from very high to very low, as shown in Table S3.

2.3. Data Collection

The World Geodetic System 1984 Universal Transverse Mercator (WGS 84 UTM) with the coordinate reference system of zone 28 was used as the geographic projection system in all the developed maps.
The first step in this methodology for the development of flood vulnerability maps was creating the Digital Elevation Model (DEM), rainfall map, and land use land cover (LULC) map. Subsequently, the determination of the vulnerability employed several parameters over the study area, including the rainfall, Normalized Difference Vegetation Index (NDVI), elevation, drainage density, Topographic Wetness Index (TWI), distance to road, distance to river, slope, and land use land cover (LULC).
The source of the 30 m DEM that was extracted and downloaded was the United States Geographical Survey (USGS) Earth Explorer. It was later transferred to Arc Map and extracted to the study area of Freetown. The 30 m DEM of the study area was also used to extract the elevation and slope map of the study area [39].
This USGS-DEM was the basis for developing the spatial map of the drainage density through the following steps: filling the DEM, determining the flow direction, flow accumulation, and conditions, and calculating the total stream length, including the total basin area. The map of the slopes was derived by default from the USGS-DEM during the analysis process.
The distance-from-river data were downloaded from HydroSHEDS under HydroRiver and were later imported to ArcGIS 10.8 and extracted to the study area, whilst the distance-to-road data were downloaded from BBBike with a shapefile format and imported to ArcGIS for processing.
The supervised classification of the LULC downloaded from the USGS through the NASA LPDAAC Collections using MODIS Land Cover V6.1 was divided into 18 classes: evergreen needleleaf forests; evergreen broadleaf forests; deciduous needleleaf forests; deciduous broadleaf forests; mixed forests; closed shrublands; open shrublands; woody savannas; savannas; grasslands; permanent wetlands; croplands; urban and built-up lands; cropland/natural vegetation mosaics; permanent snow and ice; barren; water bodies; and unclassified [40]. The LULC map was based on supervised classification, which was extracted to the study area using ArcGIS 10.8.
The precipitation data used were downloaded from the satellite precipitation product developed by the Center for Hydrometeorology and Remote Sensing (CHRS) at the University of California, Irvine (UCI). PDIR-Now was implemented on the UCI CHRS global real-time satellite precipitation monitoring system—iRain (https://irain.eng.uci.edu (accessed on 1 August 2024))—and the entire study area was mapped by using daily rainfall records from the CFSR data. The main advantage of PDIR-Now, compared to other near-real-time precipitation datasets, is its reliance on high-frequency sampled IR imagery; consequently, the latency of PDIR-Now from the time of rainfall occurrence is very short (15–60 min). It has a coverage of 60° S to 60° N and a resolution of 0.04° × 0.04°. The annual rainfall data for 2001, 2010, and 2020 were downloaded using NetCDF and the map was developed using ArcGIS 10.8 and was later resampled and extracted to the study area.
The Normalized Difference Vegetation Index (NDVI) data were downloaded from USGS Earth Explorer using the MODIS Vegetation Indies V6.1 data type and were later further processed on ArcGIS 10.8 using remote sensing techniques to calculate the NDVI from the satellite using the following equation:
NDVI = B 2 − B 1 B 2 + B 1 ,
The NDVI data were generated to the study area and NDVI ranges were assigned to the various classes, such as water, built-up area, barren land, shrub and grassland, sparse vegetation, and dense vegetation, to assess the vegetation health and cover, which are indicative of the soil moisture and potential water absorption.
The Topographic Wetness Index (TWI) was generated using the 30 m DEM on ArcGIS 10.8 using the following method.
Firstly, the DEM data covering the study area were imported into ArcGIS, followed by pre-processing steps to fill in any sinks in the terrain. Following this, the flow direction and flow accumulation were computed to determine the direction and amount of water flow across the landscape. The slope was later calculated to characterize the steepness of the terrain. Using these calculated parameters, the TWI was then computed using a mathematical formula involving the natural logarithm of the ratio between the accumulated flow accumulation and slope, which is as follows:
TWI = ln   A tan S ,
where A is the accumulated flow accumulation, and S is the slope in radians. The resulting TWI raster was visualized and analyzed to identify areas of high and low wetness, providing insights into the hydrological processes and landscape characteristics, such as the soil moisture and drainage patterns. The entire flowchart illustrating the integration of various datasets and factors, including HydroSHEDS, MODIS Land Cover, SRTM, and others, into the Analytical Hierarchical Process (AHP) to assess and identify flood-vulnerable zones is shown in Figure 2 below.

2.4. AHP Model Application

The Analytic Hierarchy Process (AHP) and Multi-Criteria Decision-Making (MCDM) methods provide a structured approach for evaluating complex problems, such as flood vulnerability mapping. These methodologies facilitate the breakdown of the decision-making process into more manageable parts, allowing for the systematic evaluation of various criteria and alternatives. Specifically, in the context of flood vulnerability mapping, the AHP and MCDM enable stakeholders to prioritize risk factors, assess the vulnerabilities of different areas, and make informed decisions about flood management and mitigation strategies [41].

2.4.1. Weighting of Flood Vulnerability Indicators

The weighting of the flood vulnerability indicators was conducted using the Analytic Hierarchy Process (AHP) [12]. Specifically, pairwise comparisons were made between each indicator based on expert judgment [21]. These comparisons were then used to calculate the weights through the AHP matrix method. To enhance the repeatability, a detailed step-by-step guide on how the AHP was applied, including how the pairwise comparison matrix was constructed and how the weights were derived, is shown in Table S3.

2.4.2. Pairwise Comparison Matrix

The pairwise comparison matrix was generated by ascribing a numerical value between 1 and 9 to each factor, reflecting its relative significance. This matrix was built through comparisons based on previous literature from similar studies. These comparisons facilitated an assessment of how each criterion or factor relates to the others, ultimately assigning weights to the different factors influencing flood events based on their priority. Consequently, this matrix aided in establishing the relative weights of the criteria, which are shown in Table S2.

2.4.3. Normalized Pairwise Comparison Matrix

The normalized pairwise comparison matrix was then obtained by dividing each element in the pairwise comparison matrix by the sum of its respective columns [42]. This step was crucial for calculating the consistency ratio, which measures the consistency to ensure that reliable weights are assigned to the factors, as seen in Table S4. The consistency ratio was calculated using the belief comparison matrices and the evidence reasoning combination rules. A check of whether the comparison is consistent was performed using the equation below:
Consistency   Index   ( CI ) = ( n − 1 λ m a x − n )
where n represents the number of factors being compared in the matrix, and λmax is the highest eigenvalue of the pairwise comparison matrix. The maximum eigenvalue was computed using the following procedure:
  • Multiply each value in the column by the criterion weight;
  • Compute the weighted sum value by adding the values in the rows;
  • Calculate the ratio of each weighted sum value to the respective criterion weight;
  • Average the ratio of the weighted sum value to the criterion weight.

2.4.4. Consistency Ratio

The consistency ratio (CR) was calculated to measure the consistency of the expert judgments. If the consistency ratio (CR) falls below 0.10, the pairwise comparison matrix exhibits satisfactory consistency. Conversely, if the CR exceeds 0.1, it signifies that the pairwise comparison matrix lacks adequate consistency, necessitating the repetition of the process until the CR drops below 0.1. This helps in assessing the reliability and consistency of the decision-making process [17,43,44].
Finally, the consistency of the pairwise comparison matrix was calculated using consistency ratios, as represented in the following equation:
C R = C I R I
where RI is the Random Inconsistency Index dependent on the sample size according to the number of factors used in the pairwise matrix, and CI is the Consistency Index.

3. Results

3.1. Flood Vulnerability Criteria

3.1.1. Rainfall

According to [45], flood vulnerability is closely related to the rainfall intensity and distribution in urban areas like Douala and Freetown. The use of annual rainfall data in this study was driven by the need to capture long-term climatic patterns and their impact on the flood vulnerability in Freetown. While extreme rainfall events are critical for understanding short-term flash floods, the annual rainfall offered a broader view of the cumulative precipitation trends over the two decades examined. This approach aligned with the study’s objective of assessing persistent flood risks linked to seasonal precipitation, urbanization, and other environmental factors that influence flooding throughout the year. According to the rainfall map, which shows data from 2001 to 2020, the northwest region of Freetown experiences the highest rainfall. Data were sourced from Persian Rainfall using the PDIR-Now dataset and imported into ArcGIS 10.8 for processing and extraction. The annual rainfall map of Freetown (Figure 3a–c) highlights areas with the highest rainfall in red and areas with the lowest rainfall in green. The map was reclassified into two categories: High and Low. Areas with higher rainfall are more susceptible to flooding, while those with lower rainfall are less vulnerable. High-intensity rainfall can overwhelm drainage systems, leading to flash floods, particularly in Freetown. Additionally, reference [46] indicates that rainfall is a major factor contributing to the flood risk in West Africa, influenced by both human and environmental factors. Future projections based on regional climate models also predict increased flood risks and vulnerability in many areas [45,47].

3.1.2. Slope

Areas with gentle slopes are more prone to flooding compared to high-elevation areas due to the accumulation of water from higher regions. The extent of inundation is influenced by both the length and steepness of the slope. For example, areas with shorter and gentler slopes tend to experience more flooding than those with longer and steeper slopes [48]. The slope map (Figure 4) was derived from the DEM obtained from the SRTM digital elevation data provided by USGS. In Freetown, the slope ranges from 0° to 60.4°. The eastern part of the map, particularly the Western Area Rural region, exhibits the lowest slope values (0° to 4.97°) and is therefore more susceptible to flooding. This low-slope area is also found in patches in the northeast and northwest. Conversely, the middle portion of the Western Area Urban region has steeper slopes ranging from 18.48° to 60.4°, indicating lower flood susceptibility.
The slope map is classified into five categories: (0°–4.97°), (4.98°–11.37°), (11.38°–18.47°), (18.48°–26.29°), and (26.30°–60.4°). The map uses color coding to represent the different slope ranges, which measure the steepness or incline of the terrain. Steeper slopes can lead to faster runoff, which increases flood hazards in lower-lying areas [49,50].

3.1.3. Land Use Land Cover

Over the past two decades, from 2001 to 2020, the land cover and land use patterns in the observed area experienced notable changes, as shown in Figure 5a–c.
Evergreen needleleaf forests decreased from 0.6 km2 in 2001 to 0.2 km2 in 2010 and had almost disappeared by 2020. This loss of forest cover may increase the flood vulnerability due to the reduced water absorption capacity.
Evergreen broadleaf forests increased from 96.07 km2 in 2001 to 122.6 km2 in 2010, but then significantly decreased to 72.9 km2 by 2020. The sharp decline from 2010 to 2020 suggests deforestation, which could reduce flood mitigation through decreased evapotranspiration and soil stabilization.
Mixed forests decreased from 0.48 km2 in 2001 and had disappeared by 2010. The complete loss of mixed forests could heighten the flood risk due to the diminished biodiversity and water absorption.
Woody savannas increased from 117.12 km2 in 2001 to 121.5 km2 in 2010 and further expanded to 131.6 km2 by 2020. The growth in woody savannas may help mitigate the flood risk by enhancing water uptake, provided they are well maintained.
Savannas grew from 237.16 km2 in 2001 to 262.6 km2 in 2010 and increased to 266.3 km2 in 2020. While savannas generally absorb water effectively, their impact on flooding is less significant unless combined with other land use changes.
Grasslands expanded from 2.16 km2 in 2001 to 10.9 km2 in 2010 and increased significantly to 30.6 km2 by 2020. Increased grasslands can improve flood resilience if managed properly to maintain healthy soil and vegetation cover.
Permanent wetlands decreased from 120.48 km2 in 2001 to 73.2 km2 in 2010 and further dropped by 67.9 km2 by 2020. The reduction in wetlands is concerning, as they play a crucial role in flood control by absorbing excess rainfall.
Croplands grew from 7.73 km2 in 2001 to 10.4 km2 in 2010 and significantly increased to 11.8 km2 by 2020. While croplands provide food, they can contribute to the flood risk if they replace absorbent surfaces like forests or wetlands.
Urban built lands expanded from 74.88 km2 in 2001 to 80.7 km2 in 2010 and increased significantly to 97.1 km2 by 2020. The rise in urban areas likely results in more impervious surfaces, increasing runoff and flood vulnerability. This trend aligns with global urbanization patterns.
Cropland mosaics decreased from 32.65 km2 in 2001 to 7.9 km2 in 2010 but increased by 10.9 km2 by 2020. Changes in this category could be related to shifting agricultural practices, potentially affecting the water runoff and flood risk.
Barren land decreased from 0.88 km2 in 2001 to 0.3 km2 in 2010, and then increased by 1.2 km2 by 2020. This variation may not significantly impact the flood vulnerability unless these areas are converted to other land uses.
Water bodies increased slightly from 5.86 km2 in 2001 to 5.9 km2 in 2010 and remained stable at 5.9 km2 in 2020. Stability in water bodies is important for flood control.
A summary of the land use land cover (LULC) changes in Freetown from 2001 to 2020 is shown in Table S1 and the LULC detection change is further illustrated on a bar chart in Figure S1.

3.1.4. Drainage Density

The drainage density is a key factor in determining the flood vulnerability in Freetown. The drainage density map (Figure 6a) was generated from the DEM obtained from the SRTM digital elevation data provided by the USGS and was processed using ArcGIS 10.8 with the Spatial Analyst Tool. The region was classified into five categories: very low (0–0.432 km/km2), low (0.433–0.935 km/km2), moderate (0.936–1.58 km/km2), high (1.59–2.52 km/km2), and very high (2.53–4.59 km/km2). The area with the highest drainage density is located in the southeastern part of Freetown, which is more susceptible to flooding, with additional high-density patches in the northeast and northwest. Generally, effective drainage systems can mitigate flood risks, so areas with higher drainage densities might be less prone to flooding. However, areas with very high drainage densities are at a greater risk of inundation compared to those with lower drainage densities [50,51].

3.1.5. Distance to Road

Areas close to roads are often among the first to experience flooding due to their interaction with existing drainage systems. Conversely, areas further from roads face different flooding dynamics. The distance-from-road map of Freetown (Figure 6b), sourced from BBBike’s website, was analyzed using ArcGIS 10.8 and the Spatial Analyst Tool. This map categorizes the area into five vulnerability classes based on the proximity to roads: very high (0–0.002193 m), high (0.0021931–0.0069445 m), moderate (0.0069446–0.012793 m), low (0.012794–0.019493 m), and very low (0.019494–0.031068 m). Areas closest to roads (0–0.002193 m), marked in red, are distributed across the map with high flood vulnerability, while the central Freetown regions (0.019494–0.031068 m), marked in green, exhibiting lower flood risks. Drainage systems associated with roads can both mitigate and exacerbate flooding conditions [52]. Road networks are critical in disaster recovery, particularly for flood hazards [1]. Poor drainage or roads obstructing water flows can increase the flood risks for nearby areas. Proximity to roads may improve access for emergency services and evacuation but it also heightens the susceptibility to quick flooding from runoff. Conversely, areas farther from roads may experience delayed emergency responses but are less affected by direct runoff. Evaluating flood vulnerability involves balancing the benefits of immediate transportation access with the risks of increased runoff from impervious surfaces.

3.1.6. Topographic Wetness Index (TWI)

The Topographic Wetness Index (TWI) helps identify areas prone to water accumulation based on their topographic characteristics. The TWI map (Figure 7a) was generated using the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) and was processed with the Spatial Analyst Tool in ArcGIS 10.8. Higher TWI values, ranging from 13.98 to 22.3, are found in the eastern and southeastern parts of Freetown. These areas have greater potential for soil moisture saturation and are more vulnerable to flooding, especially during heavy rainfall events. Studies indicate that higher TWI values are often associated with increased flood hazards [53]. Conversely, regions with the lowest TWI values, between 2.985 and 6.014, are located from the northwest to the southwest of Freetown. These areas are less prone to retaining surface water and are considered less vulnerable to flooding [51]. The TWI has been categorized into five classes: 2.985–6.014, 6.015–7.908, 7.909–10.56, 10.57–13.97, and 13.98–22.3.

3.1.7. Distance to River

Proximity to rivers plays a crucial role in flood risk, as riverbanks can overflow during heavy rainfall. The “Distance from River” map of Freetown (Figure 7b), sourced from HydroSHEDS, was analyzed in ArcGIS 10.8 using the Spatial Analyst Tool. This analysis categorized the area into five vulnerability classes based on the proximity to rivers: very high (0–0.00776 m), high (0.007761–0.04536 m), moderate (0.04537–0.1056 m), low (0.1057–0.1271 m), and very low (0.1271–0.1771 m). Areas closest to rivers, marked in red (0–0.00776 m), are spread across the map, indicating high susceptibility to flooding. In contrast, areas farther from the river, colored in green (0.1057–0.1271 m), are found in isolated patches and show a reduced flood risk. Regions adjacent to rivers are particularly vulnerable to flooding, especially if the rivers overflow their banks [54,55]. The risk is further elevated for areas at lower elevations or next to rivers with high flow rates [56]. Although areas farthest from rivers are less susceptible to riverine flooding, they may still experience flooding from surface runoff or overwhelmed drainage systems.

3.1.8. Normalized Difference Vegetation Index (NDVI)

The Normalized Difference Vegetation Index (NDVI) evaluates the vegetation health and density by measuring the reflection of visible and near-infrared light. NDVI values range from −1 to +1, with higher values indicating denser and healthier vegetation [46]. The NDVI map of Freetown (Figure 8a), sourced from MODIS Vegetation Index data via the USGS, was processed using the Spatial Analyst Tool in ArcGIS 10.8. The map classifies the region into six categories: water (−0.28 to −0.015), built-up area (0.015 to 0.14), barren land (0.14 to 0.18), shrub and grassland (0.18 to 0.27), sparse vegetation, and dense vegetation. Areas identified as built-up land, marked in red with NDVI values between 0.0501 and 0.14, are particularly susceptible to flooding. This vulnerability is due to the prevalence of impervious surfaces, which impede water absorption and increase runoff, thereby raising flood risks [50,57]. Conversely, regions with dense vegetation, shown in dark green with NDVI values from 0.3601 to 0.74, are less prone to flooding. The dense vegetation in these areas effectively absorbs water, and the root systems help stabilize the soil, reducing runoff and lowering the flood risk [58].

3.1.9. Elevation

Elevation significantly impacts flood vulnerability, with lower-lying areas typically facing a higher risk of flooding, especially in coastal regions [59,60]. Conversely, higher-elevation areas are generally less susceptible to flooding but may face other risks, such as landslides if the terrain is steep and unstable [61,62]. The elevation map of Freetown (Figure 8b) was created using Shuttle Radar Topography Mission (SRTM) data from the USGS and was processed in ArcGIS 10.8. The map classifies the terrain into five elevation categories: very low (−13 to 80.29 m), low (80.3 to 216.6 m), moderate (216.66 to 363.76 m), high (363.77 to 525.24 m), and very high (525.25 to 902 m). The lowest elevation band, from −13 to 80.29 m, covers the areas most vulnerable to flooding, particularly in the eastern and southeastern parts of Freetown, including coastal zones prone to storm surges. In contrast, areas with elevations between 525.25 and 902 m are generally less at risk of flooding and are located mainly in central Freetown. However, high-elevation areas can contribute to the flood risk in lower regions by generating significant runoff during heavy rainfall.

3.2. Flood Vulnerability Maps of Freetown

The flood vulnerability maps of Freetown (Figure 9a–c) reveal the following trends over the two decades from 2001 to 2020.
In 2001, very-high-vulnerability areas (marked in red) in the Western Area Rural region covered approximately 68.84 km2, which represented about 10% of the total area. High-vulnerability areas (orange) covered around 137.68 km2, accounting for 20% of the area. Medium-vulnerability zones (yellow) extended over 206.51 km2 (30%), while low-vulnerability (light green) and very-low-vulnerability (green) areas covered 172.10 km2 (25%) and 103.26 km2 (15%), respectively.
By 2010, the very-high-vulnerability areas in the Western Area Rural region had decreased to 103.26 km2, representing 15% of the total area. High-vulnerability zones remained at 137.68 km2 (20%), but medium-vulnerability areas expanded to 240.93 km2 (35%), becoming the most dominant category. The Western Area Urban region also showed significant high vulnerability, with increased medium-vulnerability zones. High- to very-high-vulnerability areas were scattered in patches across the northwest and southeast regions.
In 2020, the trend of decreasing very-high-vulnerability areas continued, shrinking to 68.84 km2 (10%). High-vulnerability zones also slightly decreased to 103.26 km2 (15%). Medium-vulnerability areas remained prevalent, covering 240.93 km2 (35%). Low- and very-low-vulnerability zones covered 172.10 km2 (25%) and 103.26 km2 (15%), respectively. High- to very-high-vulnerability zones were primarily concentrated in the eastern and southeastern parts of Freetown.
Overall, the progression of the flood vulnerability in Freetown from 2001 to 2020 indicates a general decrease in areas classified as very-high-vulnerability. The very-high-vulnerability zones remained stable at 68.84 km2 (10%) throughout the period. High-vulnerability areas saw a slight reduction from 137.68 km2 (20%) in 2001 to 103.26 km2 (15%) in 2020. Conversely, medium-vulnerability zones expanded from 206.51 km2 (30%) in 2001 to 240.93 km2 (35%) in 2020, reflecting a shift towards more moderate-vulnerability classifications. These changes suggest that improvements in land use, infrastructure development, and other factors may have contributed to reducing the overall flood vulnerability in the region over the two decades.

4. Discussion

In the dynamic landscape of Freetown, where natural forces and human activities intersect, a vivid narrative of flood vulnerability and resilience emerges. This research, encompassing the years 2001 to 2020, meticulously examines how the intricate interplay of rainfall, topography, land use, and infrastructure weaves together to shape the flood risk profile of the city. The rainfall maps highlight areas of high flood vulnerability, particularly in Freetown’s northwest, where intense rainfall overwhelms the drainage systems, a pattern also observed in urban studies of Douala. This issue is consistent with trends across West Africa [63], where high-intensity storms can lead to flash floods. However, using annual rainfall data may overlook the impact of short-duration, high-intensity storms, affecting the precision of flood risk identification.
The gentle slopes in Freetown, with their gradual inclines, tend to retain water, creating vulnerable pockets, especially in the Western Area Rural region. This correlation between slope and flooding is supported by global geomorphological research [64].
Over the past two decades, the reduction in evergreen forests and the expansion of urban areas have heightened Freetown’s flood risks. The increase in impervious surfaces from 74.88 km2 to 97.1 km2 has intensified runoff, a common trend in rapidly urbanizing areas [55]. The loss of natural vegetation, which is crucial for water absorption, further contributes to flood susceptibility.
The drainage density map of Freetown reveals a dual narrative: areas with high drainage densities in the southeast channel flood swiftly, aligning with studies on hydrological patterns [65]. Yet, these areas also face the paradox of higher inundation risks if the systems falter.
Proximity to roads adds another layer of complexity. Areas near roads, depicted in red, are more susceptible to flooding due to runoff from impervious surfaces, an infrastructural imprint that reflects the challenge of balancing accessibility with the flood risk [66].
The Topographic Wetness Index (TWI) highlights regions with high soil moisture saturation, particularly in the east and southeast, indicating higher flood potential. This index, commonly used in environmental assessments, underscores the flood risks associated with these areas [67].
Proximity to rivers is another crucial factor; areas closest to watercourses are shown in red on the maps, demonstrating their high susceptibility to flooding, as supported by hydrological studies [68].
The NDVI maps illustrate the protective role of dense vegetation in mitigating flood risks. Healthier green zones, with their superior water absorption and soil stabilization, contrast with areas of reduced vegetation that face higher flood risks. Elevation data further complement this picture, showing that lower-lying coastal areas are at a higher risk of flooding, consistent with global studies on topography and flood risks [65].
The flood vulnerability maps from 2001 to 2020 depict a gradual shift from high to medium vulnerability, particularly in the Western Area Rural region. This shift suggests improvements in land management and infrastructure over time, reflecting broader regional adaptations to changing climatic conditions [69].

5. Conclusions

This study conducted a thorough assessment of the flood vulnerability in Freetown, Sierra Leone, covering the years 2001 to 2020. By utilizing the Analytical Hierarchy Process (AHP) alongside Multi-Criteria Decision Making (MCDM), the research effectively incorporated nine critical criteria: the rainfall (0.27); slope (0.18); land use land cover (LULC) (0.14); drainage density (0.09); distance to road (0.08); Topographic Wetness Index (TWI) (0.08); distance to river (0.06); Normalized Difference Vegetation Index (NDVI) (0.06); and elevation (0.04). This approach facilitated a detailed analysis and evaluation of their individual and collective impacts on the flood vulnerability.
The flood vulnerability maps indicate a shift from high to medium vulnerability over the two decades, particularly in the Western Area Rural region, suggesting that improvements in land management and infrastructure may have contributed to this trend. However, several limitations and potential sources of error must be acknowledged to provide a comprehensive understanding of the findings. Firstly, the present study did not account for the vulnerability of specific elements at risk, such as infrastructure, populations, or economic activities, which are critical factors in assessing the overall flood risk. The analysis was based solely on environmental and land use variables, excluding management factors like flood control measures and urban planning strategies. This exclusion may limit the ability to fully understand the drivers of flood vulnerability and the impact of mitigation efforts.
Additionally, the study employed annual rainfall data as a key factor in the flood vulnerability assessment. This choice was made due to the lack of available stream flow or river flow data, which are typically crucial for determining flood risks. While annual rainfall provides a broad overview of climatic influences, the use of maximum-24-hour rainfall data and specific flood events would likely improve the accuracy and confidence of the results. The absence of recorded flood data and river flow statistics also means that the results have not been validated against actual flood occurrences. As a result, while the findings offer valuable insights, they should be interpreted with caution.
Regarding the return period or recurrence interval of the flooding vulnerability, this study did not explicitly address these metrics due to the focus on the general vulnerability rather than specific flood events. Future research should incorporate these aspects to enhance the predictive power and applicability of flood hazard maps.
Despite these limitations, the integration of various environmental factors through the Analytical Hierarchy Process (AHP) and Multi-Criteria Decision Making (MCDM) has produced a valuable tool for identifying flood-prone areas. The results, while requiring cautious interpretation, provide a useful foundation for urban planning, disaster management, and future studies aimed at improving the flood resilience in Freetown.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/hydrology11100158/s1, Figure S1: LULC change detection of Freetown from 2001, 2010, and 2020; Table S1: LULC change for Freetown from 2001, 2010, and 2022 using MODIS classification; Table S2: Pairwise comparison of 9 × 9 decision matrix; Table S3: Weights of sub-classes using AHP comparison matrix; Table S4: Normalized pairwise comparison; Table S5: Pairwise comparison matrix scale and random index for AHP.

Author Contributions

Conceptualization, A.O.K. and M.S.; methodology, A.O.K., M.S. and C.A.; software M.S.; validation, M.S. and C.A.; formal analysis, A.O.K.; investigation, A.O.K. and M.S.; resources, M.S. and C.A.; data curation, A.O.K., M.S. and C.A.; writing—original draft preparation, A.O.K.; writing—review and editing, M.S. and C.A.; visualization, M.S. and C.A.; supervision, M.S. and C.A; project administration, C.A.; funding acquisition, M.S. and C.A. All authors have read and agreed to the published version of the manuscript.

Funding

This research received funding from the African Union Member States. The research grant was integral to the African Union Scholarship, generously supported by the German Government, administered through the Pan African University Institute for Water and Energy Sciences including Climate Change (PAUWES) (Grant Number: PAUWES/2022/MCC21).

Data Availability Statement

Rainfall data are available at PDIR-Now https://chrsdata.eng.uci.edu/ (accessed on 2 January 2024). For land cover data, the data are available at MODIS Land Cover V6.1 USGS Earth Explorer https://earthexplorer.usgs.gov/ (accessed on 3 January 2024). Distance-to-road data are available at BBBike https://extract.bbbike.org/ (accessed on 4 January 2024). Data for distance to river are available at https://www.hydrosheds.org/ (accessed on 1 August 2024). Data for NDVI MODIS Vegetation Indies V6.1 are available at USGS Earth Explorer https://earthexplorer.usgs.gov/ (accessed on 3 January 2024), and data for DEM, drainage density, and slope are available at USGS Earth Explorer https://earthexplorer.usgs.gov/ (accessed on 3 January 2024).

Acknowledgments

The authors extend their gratitude to the Pan-African University and the African Union Commission for their grant, which facilitated this research. The authors appreciate the insightful and constructive feedback from the anonymous reviewers, which significantly contributed to refining the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Park, K.; Lee, M.-H. The Development and Application of the Urban Flood Risk Assessment Model for Reflecting upon Urban Planning Elements. Water 2019, 11, 920. [Google Scholar] [CrossRef]
  2. Berndtsson, R.; Becker, P.; Persson, A.; Aspegren, H.; Haghighatafshar, S.; Jönsson, K.; Larsson, R.; Mobini, S.; Mottaghi, M.; Nilsson, J.; et al. Drivers of changing urban flood risk: A framework for action. J. Environ. Manag. 2019, 240, 47–56. [Google Scholar] [CrossRef] [PubMed]
  3. Salazar-Briones, C.; Ruiz-Gibert, J.M.; Lomelí-Banda, M.A.; Mungaray-Moctezuma, A. An Integrated Urban Flood Vulnerability Index for Sustainable Planning in Arid Zones of Developing Countries. Water 2020, 12, 608. [Google Scholar] [CrossRef]
  4. Abdrabo, K.I.; Kantoush, S.A.; Esmaiel, A.; Saber, M.; Sumi, T.; Almamari, M.; Elboshy, B.; Ghoniem, S. An integrated indicator-based approach for constructing an urban flood vulnerability index as an urban decision-making tool using the PCA and AHP techniques: A case study of Alexandria, Egypt. Urban Clim. 2023, 48, 101426. [Google Scholar] [CrossRef]
  5. Ogie, R.; Holderness, T.; Dunn, S.; Turpin, E. Assessing the vulnerability of hydrological infrastructure to flood damage in coastal cities of developing nations. Comput. Environ. Urban Syst. 2018, 68, 97–109. [Google Scholar] [CrossRef]
  6. López-Valencia, A.P. Vulnerability assessment in urban areas exposed to flood risk: Methodology to explore green infrastructure benefits in a simulation scenario involving the Cañaveralejo River in Cali, Colombia. Nat. Hazards 2019, 99, 217–245. [Google Scholar] [CrossRef]
  7. Cea, L.; Costabile, P. Flood Risk in Urban Areas: Modelling, Management and Adaptation to Climate Change. A Review. Hydrology 2022, 9, 50. [Google Scholar] [CrossRef]
  8. Westra, S.; Varley, I.; Jordan, P.; Nathan, R.; Ladson, A.; Sharma, A.; Hill, P. Addressing Climatic Non-Stationarity in the Assessment of Flood Risk. Australas. J. Water Resour. 2010, 14, 1–16. [Google Scholar] [CrossRef]
  9. Trenberth, K.E. Challenges for Observing and Modeling the Global Water Cycle. In Remote Sensing of the Terrestrial Water Cycle, 1st ed.; Geophysical Monograph Series; Lakshmi, V., Alsdorf, D., Anderson, M., Biancamaria, S., Cosh, M., Entin, J., Huffman, G., Kustas, W., van Oevelen, P., Painter, T., et al., Eds.; Wiley: Hoboken, NJ, USA, 2014; pp. 511–519. [Google Scholar] [CrossRef]
  10. Shahirnia, M.; Vidrio-Sahagún, C.T.; He, J.; Valeo, C.; van Duin, B.; Beaudry, M.; Neumann, N.F. Land use and rainfall influences on bacterial levels and sources in stormwater ponds. Environ. Sci. Pollut. Res. 2023, 30, 112236–112251. [Google Scholar] [CrossRef]
  11. Clark-Ginsberg, A.; Blake, J.S.; Patel, K.V. Hybrid governance and disaster management in Freetown, Sierra Leone, Monrovia, Liberia, and Dar es Salaam, Tanzania. Disasters 2022, 46, 450–472. [Google Scholar] [CrossRef]
  12. Tempa, K. District flood vulnerability assessment using analytic hierarchy process (AHP) with historical flood events in Bhutan. PLoS ONE 2022, 17, e0270467. [Google Scholar] [CrossRef] [PubMed]
  13. Syldon, P.; Shrestha, B.B.; Miyamoto, M.; Tamakawa, K.; Nakamura, S. Assessing the impact of climate change on flood inundation and agriculture in the Himalayan Mountainous Region of Bhutan. J. Hydrol. Reg. Stud. 2024, 52, 101687. [Google Scholar] [CrossRef]
  14. Cotugno, A.; Smith, V.; Baker, T.; Srinivasan, R. A Framework for Calculating Peak Discharge and Flood Inundation in Ungauged Urban Watersheds Using Remotely Sensed Precipitation Data: A Case Study in Freetown, Sierra Leone. Remote Sens. 2021, 13, 3806. [Google Scholar] [CrossRef]
  15. Cobbinah, P.B.; Addaney, M. Sustainable Urban Futures in Africa, 1st ed.; Routledge: New York, NY, USA, 2021; Available online: https://www.taylorfrancis.com/books/9781003181484 (accessed on 29 August 2024).
  16. Kanani-Sadat, Y.; Arabsheibani, R.; Karimipour, F.; Nasseri, M. A new approach to flood susceptibility assessment in data-scarce and ungauged regions based on GIS-based hybrid multi criteria decision-making method. J. Hydrol. 2019, 572, 17–31. [Google Scholar] [CrossRef]
  17. Saaty, T.L. Decision making with the analytic hierarchy process. Int. J. Serv. Sci. 2008, 1, 83. [Google Scholar] [CrossRef]
  18. Handayani, W.; Chigbu, U.E.; Rudiarto, I.; Putri, I.H.S. Urbanization and Increasing Flood Risk in the Northern Coast of Central Java—Indonesia: An Assessment towards Better Land Use Policy and Flood Management. Land 2020, 9, 343. [Google Scholar] [CrossRef]
  19. Dhawale, R.; Schuster-Wallace, C.J.; Pietroniro, A. Assessing the multidimensional nature of flood and drought vulnerability index: A systematic review of literature. Int. J. Disaster Risk Reduct. 2024, 112, 104764. [Google Scholar] [CrossRef]
  20. Lins-De-Barros, F.M. Integrated coastal vulnerability assessment: A methodology for coastal cities management integrating socioeconomic, physical and environmental dimensions—Case study of Região dos Lagos, Rio de Janeiro, Brazil. Ocean Coast. Manag. 2017, 149, 1–11. [Google Scholar] [CrossRef]
  21. Mourato, S.; Fernandez, P.; Pereira, L.G.; Moreira, M. Assessing Vulnerability in Flood Prone Areas Using Analytic Hierarchy Process—Group Decision Making and Geographic Information System: A Case Study in Portugal. Appl. Sci. 2023, 13, 4915. [Google Scholar] [CrossRef]
  22. Vidrio-Sahagún, C.T.; Ruschkowski, J.; He, J.; Pietroniro, A. A practice-oriented framework for stationary and nonstationary flood frequency analysis. Environ. Model. Softw. 2024, 173, 105940. [Google Scholar] [CrossRef]
  23. Hoque, M.A.-A.; Tasfia, S.; Ahmed, N.; Pradhan, B. Assessing Spatial Flood Vulnerability at Kalapara Upazila in Bangladesh Using an Analytic Hierarchy Process. Sensors 2019, 19, 1302. [Google Scholar] [CrossRef] [PubMed]
  24. Nelson, A.; Lindbergh, S.; Stephenson, L.; Halpern, J.; Arroyo, F.A.; Espinet, X.; González, M.C. Coupling Natural Hazard Estimates with Road Network Analysis to Assess Vulnerability and Risk: Case Study of Freetown (Sierra Leone). Transp. Res. Rec. J. Transp. Res. Board 2019, 2673, 11–24. [Google Scholar] [CrossRef]
  25. Das, S. Flood susceptibility mapping of the Western Ghat coastal belt using multi-source geospatial data and analytical hierarchy process (AHP). Remote Sens. Appl. Soc. Environ. 2020, 20, 100379. [Google Scholar] [CrossRef]
  26. Balica, S.F.; Wright, N.G.; Van Der Meulen, F. A flood vulnerability index for coastal cities and its use in assessing climate change impacts. Nat. Hazards 2012, 64, 73–105. [Google Scholar] [CrossRef]
  27. Alfieri, L.; Dottori, F.; Betts, R.; Salamon, P.; Feyen, L. Multi-Model Projections of River Flood Risk in Europe under Global Warming. Climate 2018, 6, 6. [Google Scholar] [CrossRef]
  28. Bauer, T.; Ingram, V.; De Jong, W.; Arts, B. The socio-economic impact of extreme precipitation and flooding on forest livelihoods: Evidence from the Bolivian Amazon. Int. For. Rev. 2018, 20, 314–331. [Google Scholar] [CrossRef]
  29. Karmaoui, A.; Balica, S. A new flood vulnerability index adapted for the pre-Saharan region. Int. J. River Basin Manag. 2021, 19, 93–107. [Google Scholar] [CrossRef]
  30. DeCorse, C.R. Sierra Leone in the Atlantic World: Concepts, contours, and exchange. Atl. Stud. 2015, 12, 296–316. [Google Scholar] [CrossRef]
  31. Redshaw, P.; Boon, D.; Campbell, G.; Willis, M.; Mattai, J.; Free, M.; Jordan, C.; Kemp, S.J.; Morley, A.; Thomas, M. The 2017 Regent Landslide, Freetown Peninsula, Sierra Leone. Q. J. Eng. Geol. Hydrogeol. 2019, 52, 435–444. [Google Scholar] [CrossRef]
  32. Cui, Y.; Cheng, D.; Choi, C.E.; Jin, W.; Lei, Y.; Kargel, J.S. The cost of rapid and haphazard urbanization: Lessons learned from the Freetown landslide disaster. Landslides 2019, 16, 1167–1176. [Google Scholar] [CrossRef]
  33. Taylor, E.T.; Kamara, I.S.; Bockarie, A. Rainfall Pattern in Freetown, Sierra Leone: From a Retrospective Viewpoint. In Proceedings of the International Conference ADAPTtoCLIMATE, Nicosia, Cyprus, 27–28 March 2014. [Google Scholar]
  34. Ibrahim, A.S.; Memon, F.A.; Butler, D. Seasonal Variation of Rainy and Dry Season Per Capita Water Consumption in Freetown City Sierra Leone. Water 2021, 13, 499. [Google Scholar] [CrossRef]
  35. Mustafa, E.K.; El-Hamid, H.T.A.; Tarawally, M. Spatial and temporal monitoring of drought based on land surface temperature, Freetown City, Sierra Leone, West Africa. Arab. J. Geosci. 2021, 14, 1013. [Google Scholar] [CrossRef]
  36. Baky, M.A.A.; Islam, M.; Paul, S. Flood Hazard, Vulnerability and Risk Assessment for Different Land Use Classes Using a Flow Model. Earth Syst. Environ. 2020, 4, 225–244. [Google Scholar] [CrossRef]
  37. Roy, D.C.; Blaschke, T. Spatial vulnerability assessment of floods in the coastal regions of Bangladesh. Geomat. Nat. Hazards Risk. 2015, 6, 21–44. [Google Scholar] [CrossRef]
  38. Ariff, H.; Salit, M.S.; Ismail, N.; Nukman, Y. Use of Analytical Hierarchy Process (AHP) for Selecting the Best Design Concept. J. Teknol. 2012, 49, 1–18. Available online: https://journals.utm.my/index.php/jurnalteknologi/article/view/188 (accessed on 29 August 2024). [CrossRef]
  39. Holmes, K.; Chadwick, O.; Kyriakidis, P. Error in a USGS 30-meter digital elevation model and its impact on terrain modeling. J. Hydrol. 2000, 233, 154–173. [Google Scholar] [CrossRef]
  40. Friedl M SMD. MODIS/Terra+Aqua Land Cover Type Yearly L3 Global 500 m 2024. Available online: https://lpdaac.usgs.gov/products/mcd12q1v061/ (accessed on 1 August 2024).
  41. de Brito, M.M.; Evers, M. Multi-criteria decision-making for flood risk management: A survey of the current state of the art. Nat. Hazards Earth Syst. Sci. 2016, 16, 1019–1033. [Google Scholar] [CrossRef]
  42. Dijkstra, T.K. On the extraction of weights from pairwise comparison matrices. Cent. Eur. J. Oper. Res. 2013, 21, 103–123. [Google Scholar] [CrossRef]
  43. Zhang, L.; Nan, Z.; Yu, W.; Ge, Y. Modeling Land-Use and Land-Cover Change and Hydrological Responses under Consistent Climate Change Scenarios in the Heihe River Basin, China. Water Resour. Manag. 2015, 29, 4701–4717. [Google Scholar] [CrossRef]
  44. Membele, G.M.; Naidu, M.; Mutanga, O. Examining flood vulnerability mapping approaches in developing countries: A scoping review. Int. J. Disaster Risk Reduct. 2022, 69, 102766. [Google Scholar] [CrossRef]
  45. Findi, E.N.; Wantim, M.N.; Ayonghe, S.N. Assessing Rainfall and Temperature Trend: Implication on flood patterns in vulnerable Communities of Limbe and Douala, Cameroon. Int. J. Environ. Sci. 2022, 7, 1–13. [Google Scholar]
  46. Tazen, F.; Diarra, A.; Kabore, R.F.W.; Ibrahim, B.; Bologo/Traoré, M.; Traoré, K.; Karambiri, H. Trends in flood events and their relationship to extreme rainfall in an urban area of Sahelian West Africa: The case study of Ouagadougou, Burkina Faso. J. Flood Risk Manag. 2019, 12, e12507. [Google Scholar] [CrossRef]
  47. Kundzewicz, Z. Flood risk and vulnerability in the changing climate. Ann. Wars Univ. Life Sci.—SGGW Land Reclam. 2008, 39, 21–31. Available online: http://archive.sciendo.com/SGGW/sggw.2008.39.issue--1/v10060-008-0002-9/v10060-008-0002-9.pdf (accessed on 29 August 2024). [CrossRef]
  48. Mishra, K.; Sinha, R. Flood risk assessment in the Kosi megafan using multi-criteria decision analysis: A hydro-geomorphic approach. Geomorphology 2020, 350, 106861. [Google Scholar] [CrossRef]
  49. Danumah, J.H.; Odai, S.N.; Saley, B.M.; Szarzynski, J.; Thiel, M.; Kwaku, A.; Kouame, F.K.; Akpa, L.Y. Flood risk assessment and mapping in Abidjan district using multi-criteria analysis (AHP) model and geoinformation techniques, (cote d’ivoire). Geoenvironmental Disasters 2016, 3, 10. [Google Scholar] [CrossRef]
  50. Nsangou, D.; Kpoumié, A.; Mfonka, Z.; Ngouh, A.N.; Fossi, D.H.; Jourdan, C.; Mbele, H.Z.; Mouncherou, O.F.; Vandervaere, J.-P.; Ngoupayou, J.R.N. Urban flood susceptibility modelling using AHP and GIS approach: Case of the Mfoundi watershed at Yaoundé in the South-Cameroon plateau. Sci. Afr. 2022, 15, e01043. [Google Scholar] [CrossRef]
  51. Radwan, F.; Alazba, A.A.; Mossad, A. Flood risk assessment and mapping using AHP in arid and semiarid regions. Acta Geophys. 2019, 67, 215–229. [Google Scholar] [CrossRef]
  52. Jamshed, A.; Birkmann, J.; Rana, I.A.; Feldmeyer, D. The effect of spatial proximity to cities on rural vulnerability against flooding: An indicator based approach. Ecol. Indic. 2020, 118, 106704. [Google Scholar] [CrossRef]
  53. Khoirunisa, N.; Ku, C.-Y.; Liu, C.-Y. A GIS-Based Artificial Neural Network Model for Flood Susceptibility Assessment. Int. J. Environ. Res. Public Health 2021, 18, 1072. [Google Scholar] [CrossRef]
  54. Ntajal, J.; Lamptey, B.L.; Mahamadou, I.B.; Nyarko, B.K. Flood disaster risk mapping in the Lower Mono River Basin in Togo, West Africa. Int. J. Disaster Risk Reduct. 2017, 23, 93–103. [Google Scholar] [CrossRef]
  55. Bhuyan, M.J.; Deka, N.; Saikia, A. Micro-spatial flood risk assessment in Nagaon district, Assam (India) using GIS-based multi-criteria decision analysis (MCDA) and analytical hierarchy process (AHP). Risk Anal. 2024, 44, 817–832. [Google Scholar] [CrossRef] [PubMed]
  56. Rudorff, C.M.; Melack, J.M.; Bates, P.D. Flooding dynamics on the lower Amazon floodplain: 2. Seasonal and interannual hydrological variability: Lower Amazon Floodplain Hydrological Variability. Water Resour. Res. 2014, 50, 635–649. [Google Scholar] [CrossRef]
  57. Wang, G.; Hu, Z.; Liu, Y.; Zhang, G.; Liu, J.; Lyu, Y.; Gu, Y.; Huang, X.; Zhang, Q.; Tong, Z.; et al. Impact of Expansion Pattern of Built-Up Land in Floodplains on Flood Vulnerability: A Case Study in the North China Plain Area. Remote Sens. 2020, 12, 3172. [Google Scholar] [CrossRef]
  58. Taherizadeh, M.; Niknam, A.; Nguyen-Huy, T.; Mezősi, G.; Sarli, R. Flash flood-risk areas zoning using integration of decision-making trial and evaluation laboratory, GIS-based analytic network process and satellite-derived information. Nat. Hazards 2023, 118, 2309–2335. [Google Scholar] [CrossRef]
  59. Swain, K.C.; Singha, C.; Nayak, L. Flood Susceptibility Mapping through the GIS-AHP Technique Using the Cloud. ISPRS Int. J. Geo-Inf. 2020, 9, 720. [Google Scholar] [CrossRef]
  60. Das, S. Geographic information system and AHP-based flood hazard zonation of Vaitarna basin, Maharashtra, India. Arab. J. Geosci. 2018, 11, 576. [Google Scholar] [CrossRef]
  61. Hammami, S.; Zouhri, L.; Souissi, D.; Souei, A.; Zghibi, A.; Marzougui, A.; Dlala, M. Application of the GIS based multi-criteria decision analysis and analytical hierarchy process (AHP) in the flood susceptibility mapping (Tunisia). Arab. J. Geosci. 2019, 12, 653. [Google Scholar] [CrossRef]
  62. Aladejana, O.; Salami, A.T.; Adetoro, O.O. Potential flood hazard zone mapping based on geomorphologic considerations and fuzzy analytical hierarchy model in a data scarce West African basin. Geocarto Int. 2021, 36, 2160–2185. [Google Scholar] [CrossRef]
  63. Ahmadalipour, A.; Moradkhani, H. A data-driven analysis of flash flood hazard, fatalities, and damages over the CONUS during 1996–2017. J. Hydrol. 2019, 578, 124106. [Google Scholar] [CrossRef]
  64. Li, L.; Yang, J.; Wu, J. Future Flood Risk Assessment under the Effects of Land Use and Climate Change in the Tiaoxi Basin. Sensors 2020, 20, 6079. [Google Scholar] [CrossRef]
  65. Chaubey, P.K.; Kundu, A.; Mall, R.K. A geo-spatial inter-relationship with drainage morphometry, landscapes and NDVI in the context of climate change: A case study over the Varuna river basin (India). Spat. Inf. Res. 2019, 27, 627–641. [Google Scholar] [CrossRef]
  66. Liu, H.; Fang, C.; Miao, Y.; Ma, H.; Zhang, Q.; Zhou, Q. Spatio-temporal evolution of population and urbanization in the countries along the Belt and Road 1950–2050. J. Geogr. Sci. 2018, 28, 919–936. [Google Scholar] [CrossRef]
  67. Beven, K.J.; Kirkby, M.J. A physically based, variable contributing area model of basin hydrology / Un modèle à base physique de zone d’appel variable de l’hydrologie du bassin versant. Hydrol. Sci. Bull. 1979, 24, 43–69. [Google Scholar] [CrossRef]
  68. Lehner, B.; Verdin, K.; Jarvis, A. New Global Hydrography Derived from Spaceborne Elevation Data. Eos Trans. Am. Geophys. Union 2008, 89, 93–94. [Google Scholar] [CrossRef]
  69. Jones, B.; O’neill, B.C. Spatially explicit global population scenarios consistent with the Shared Socioeconomic Pathways. Environ. Res. Lett. 2016, 11, 084003. [Google Scholar] [CrossRef]
Hydrology 11 00158 g001
Figure 1. (a) Map of Freetown; (b) map of Africa; (c) map of Africa.
Figure 1. (a) Map of Freetown; (b) map of Africa; (c) map of Africa.
Hydrology 11 00158 g001
Hydrology 11 00158 g002
Figure 2. Flowchart showing dataset integration of the study area.
Figure 2. Flowchart showing dataset integration of the study area.
Hydrology 11 00158 g002
Hydrology 11 00158 g003
Figure 3. (a) Map of annual rainfall for Freetown 2001; (b) map of annual rainfall for Freetown 2010; (c) annual rainfall for Freetown 2020 (source: Esri ArcGIS 10.8).
Figure 3. (a) Map of annual rainfall for Freetown 2001; (b) map of annual rainfall for Freetown 2010; (c) annual rainfall for Freetown 2020 (source: Esri ArcGIS 10.8).
Hydrology 11 00158 g003
Hydrology 11 00158 g004
Figure 4. Slope map of Freetown (source: Esri ArcGIS).
Figure 4. Slope map of Freetown (source: Esri ArcGIS).
Hydrology 11 00158 g004
Hydrology 11 00158 g005
Figure 5. (a) LULC map of Freetown 2001; (b) LULC map of Freetown 2010; (c) LULC map of Freetown 2020 (source: Esri ArcGIS).
Figure 5. (a) LULC map of Freetown 2001; (b) LULC map of Freetown 2010; (c) LULC map of Freetown 2020 (source: Esri ArcGIS).
Hydrology 11 00158 g005
Hydrology 11 00158 g006
Figure 6. (a) Drainage density map of Freetown; (b) distance-from-road map of Freetown (source: Esri ArcGIS).
Figure 6. (a) Drainage density map of Freetown; (b) distance-from-road map of Freetown (source: Esri ArcGIS).
Hydrology 11 00158 g006
Hydrology 11 00158 g007
Figure 7. (a) TWI map of Freetown; (b) distance-from-river map of Freetown (source: Esri ArcGIS).
Figure 7. (a) TWI map of Freetown; (b) distance-from-river map of Freetown (source: Esri ArcGIS).
Hydrology 11 00158 g007
Hydrology 11 00158 g008
Figure 8. (a) NDVI map of Freetown; (b) elevation map of Freetown (source: Esri ArcGIS).
Figure 8. (a) NDVI map of Freetown; (b) elevation map of Freetown (source: Esri ArcGIS).
Hydrology 11 00158 g008
Hydrology 11 00158 g009
Figure 9. (a) Flood vulnerability map 2001; (b) flood vulnerability map 2010; (c) flood vulnerability map 2020 (source: Esri ArcGIS).
Figure 9. (a) Flood vulnerability map 2001; (b) flood vulnerability map 2010; (c) flood vulnerability map 2020 (source: Esri ArcGIS).
Hydrology 11 00158 g009
Table 1. Datasets used in the study.
Table 1. Datasets used in the study.
S/NoDatasetsDescriptionData SourcesResolution
1RainfallPDIRPDIR-Now https://chrsdata.eng.uci.edu/ (accessed on 2 January 2024)1 km
2SlopeDerived from DEMDEM30 m
3LULCMODIS Land Cover V6.1USGS Earth Explorer https://earthexplorer.usgs.gov/ (accessed on 3 January 2024)500 m
4Drainage Density Extracted from DEMDEM30 m
5Distance to RoadDerived from BBBike and DEMBBBike https://extract.bbbike.org/ (accessed on 4 January 2024)30 m
6TWIExtracted from DEMDEM30 m
7Distance to RiverHydroSHEDSHydroSHEDS https://www.hydrosheds.org/ (accessed on 3 January 2024)3 arc seconds
8NDVIMODIS Vegetation Indices V6.1USGS Earth Explorer https://earthexplorer.usgs.gov/ (accessed on 3 January 2024)250 m
9ElevationSRTM USGS Earth Explorer https://earthexplorer.usgs.gov/ (accessed on 3 January 2024)30 m
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Koroma, A.O.; Saber, M.; Abdelbaki, C. Urban Flood Vulnerability Assessment in Freetown, Sierra Leone: AHP Approach. Hydrology 2024, 11, 158. https://doi.org/10.3390/hydrology11100158

AMA Style

Koroma AO, Saber M, Abdelbaki C. Urban Flood Vulnerability Assessment in Freetown, Sierra Leone: AHP Approach. Hydrology. 2024; 11(10):158. https://doi.org/10.3390/hydrology11100158

Chicago/Turabian Style

Koroma, Abdulai Osman, Mohamed Saber, and Cherifa Abdelbaki. 2024. "Urban Flood Vulnerability Assessment in Freetown, Sierra Leone: AHP Approach" Hydrology 11, no. 10: 158. https://doi.org/10.3390/hydrology11100158

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop