Geospatial assessment of urban flood resilience in rapidly growing Abbottabad City, Pakistan

Abstract

Urban flooding is one of the most recurrent and destructive natural hazards affecting rapidly urbanizing cities in developing countries. Pakistan faces a severe and growing urban flood challenge, yet spatially explicit, multi-dimensional resilience assessments remain scarce for fast-growing secondary cities. This study assessed urban flood resilience in Abbottabad City, Khyber Pakhtunkhwa, Pakistan, by quantifying flood hazard, exposure, susceptibility, and coping capacity through an integrated geospatial framework. The Urban Flood Resilience Model was applied, combining remote sensing data, GIS-based weighted overlay analysis, and the Analytical Hierarchy Process (AHP) to derive indicator weights and generate thematic component maps. Primary data on flood damages, economic capacity, institutional capacity, and educational level were collected through field surveys and questionnaires across flood-prone union councils. The final Urban Flood Resilience Index (UFRI) map showed marked regional variability. The northern and central parts of Abbottabad faced relatively high flood hazard, exposure, and susceptibility. Yet these areas showed high to very high resilience, because stronger infrastructure, institutional support, and coping ability offset the risk. By contrast, the eastern and south-eastern zones showed low to very low resilience. Weak coping capacity there worsened the effects of flood hazard and vulnerability. All AHP consistency ratios were within the acceptable threshold (Consistency Ratio < 0.1), confirming internal consistency of the pairwise comparisons. This assessment quantified both physical and socioeconomic dimensions of flood resilience, and the resulting Urban Flood Resilience Index delineated priority intervention zones requiring targeted drainage infrastructure upgrades, institutional strengthening, and community preparedness programs. The findings provide spatially explicit evidence for disaster risk managers, urban planners, and policymakers in Abbottabad and comparable mountainous cities of South Asia.

1 Introduction

Flooding is the most pervasive and economically destructive natural hazard of the twenty-first century. It accounts for around one-third of global natural disaster losses and affects more people worldwide than any other hydro-meteorological event (Hu et al., 2018; Merz et al., 2021). Climate change is intensifying flood frequency and magnitude by altering the global hydrological cycle, producing more intense and irregular precipitation (Bhunya et al., 2020). Rapid urbanization compounds this by expanding impervious surfaces, increasing surface runoff, and degrading natural drainage systems (Zhou Q. et al., 2019). In some vulnerable settings, land subsidence further amplifies inundation extent (He et al., 2025). Together, these drivers are producing a growing urban flood crisis in densely populated cities worldwide. A single flood event can cause catastrophic infrastructure damage, prolonged disruption of essential services, and significant loss of life (Marengo et al., 2020; Chan et al., 2021). Pluvial flooding occurs when intense rainfall overwhelms stormwater drainage capacity. It is most damaging in rapidly expanding cities of developing countries, where infrastructure investment has not kept pace with urban growth (Bertilsson et al., 2019; Bulti and Abebe, 2020). The Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report confirms that the frequency and intensity of heavy precipitation events over Asia have increased since the mid-twentieth century, with human-induced greenhouse gas emissions identified as the likely primary driver (IPCC, 2021). Analyses of long-term station records over Pakistan document a statistically significant increasing trend in extreme precipitation in the northern and north-western regions (Bhatti et al., 2020; Ullah et al., 2021).

Across South Asia, rapid and largely unplanned urbanization, intensifying rainfall, and weak governance have driven a sharp rise in urban flood risk (Sharifi and Yamagata, 2018; Khurram, 2022). Pakistan has experienced some of the most catastrophic flood events on record in recent decades. These events caused widespread loss of life, displacement, and economic damage (Rahman et al., 2017; Khan et al., 2023). Yet disaster management in Pakistan remains predominantly reactive and top-down. Community participation is limited, local risk reduction policies are weakly implemented, and comprehensive multi-hazard preparedness frameworks are largely absent (Maqbool et al., 2017; Alvi and Khayyam, 2020; Shah et al., 2021). Rapid population growth, poverty, and political instability compound these structural deficiencies. Together, they constrain adaptive capacity and leave urban communities exposed to flood impacts (Seidler et al., 2018; Shakoor et al., 2023).

Traditional flood risk management has primarily focused on hazard mapping. This approach identifies flood-prone areas through spatial analysis of topography, rainfall, and land use, relying on GIS-based multi-criteria decision analysis and hydrological models (Sisman and Aydinoglu, 2020; Moazzam et al., 2020; Luu et al., 2021). Risk is conventionally defined as the product of hazard and vulnerability. Vulnerability, in turn, includes both the susceptibility of exposed assets and the coping capacity of affected communities (Suriya and Mudgal, 2012). Such approaches are useful for identifying physically flood-prone zones. However, they do not capture the full range of factors that determine whether a community can withstand, respond to, and recover from flood events. This limitation is increasingly recognized in the disaster risk literature (Xu et al., 2018; Bertilsson et al., 2019).

Research attention has since moved toward urban flood resilience assessment. This approach integrates physical, socioeconomic, institutional, and infrastructural dimensions into a composite evaluation of a city’s capacity to absorb and recover from flood impacts (Zhou W. et al., 2019; Unisdr, 2011). Resilience-oriented frameworks combine remote sensing, GIS, and the Analytical Hierarchy Process (AHP) to weight and spatially integrate heterogeneous indicators into composite indices, including the Urban Flood Resilience Index (UFRI) (Bertilsson et al., 2019; Tayyab et al., 2021; Jiang et al., 2026). Studies in this tradition show that socioeconomic factors, including institutional capacity, income levels, and educational attainment, can be as decisive as physical hazard in determining flood outcomes (Jamali et al., 2023; Gabor and Tibor, 2023). The UFRM, developed by Tayyab et al. (2021) and applied to Peshawar, is a well-validated framework. It operationalizes this multi-dimensional approach through four components: hazard, exposure, susceptibility, and coping capacity.

A clear gap nonetheless remains. Integrated resilience frameworks have rarely been applied to rapidly growing secondary cities in mountainous regions of South Asia, where topographic complexity, data scarcity, and institutional fragility intersect to create distinctive and poorly understood flood risk configurations (Fan et al., 2017; Meng et al., 2019). Abbottabad City, in the Lesser Himalayas of Khyber Pakhtunkhwa, Pakistan, exhibits this challenge. According to the 2017 Population Census, Abbottabad District has a population of 1,332,912. This reflects significant demographic expansion and rising demand on urban infrastructure. The city has steep, complex terrain and a long-term mean annual rainfall of 1,366 mm (Pakistan Meteorological Department historical station records) concentrated in the monsoon season. Urban population growth reaches 3.32% per year, and the drainage system is inadequate for current urban loads. As a result, Abbottabad has experienced recurring pluvial flooding in 2019, 2021, and 2022, causing extensive damage to infrastructure and residential areas (Bahrawi et al., 2020; Khan et al., 2021). Yet no spatially explicit, multi-dimensional flood resilience assessment has been conducted for the city. Policymakers and disaster managers therefore lack the evidence base needed for targeted intervention. This study addresses that gap. It frames urban flood resilience as a human-environment interaction problem, jointly assessing the physical hazard landscape and the socioeconomic capacity of communities to produce spatially explicit, policy-relevant outcomes.

This study applies the UFRM within an integrated GIS-AHP framework to produce the first comprehensive spatial assessment of urban flood resilience for Abbottabad. The UFRM uses sixteen indicators across four components. Hazard includes elevation, slope, land use and land cover, rainfall, curve number grid, and flood damage. Exposure comprises population density, annual income, and health and educational facilities. Susceptibility covers residential, commercial, and other building types. Coping capacity consists of institutional capacity, educational level, and economic capacity. Pairwise comparisons on Saaty’s (1980) AHP scale were used to derive indicator weights. These weights were checked for consistency, then combined through weighted overlay analysis in ArcGIS to generate component maps and the composite UFRI for all 52 union councils. The study has three objectives. First, it maps the spatial distribution of flood hazard, exposure, susceptibility, and coping capacity across Abbottabad. Second, it derives AHP-based indicator weights and computes the UFRI for all union councils. Third, it delineates spatially explicit priority zones of low resilience to derive evidence-based policy recommendations. The findings aim to support disaster risk reduction, urban planning, and climate adaptation in Abbottabad and comparable mountainous cities of South Asia.

2 Materials and methods

2.1 The study area

Abbottabad is a district in the Hazara Division of Khyber Pakhtunkhwa, Pakistan, covering approximately 1,969 km². The city is situated in a valley encircled by the Sarban Hills, about 120 km north of Islamabad. It lies between 34°07′52″to 34°12′28″N and 73°11′15″to 73°16′43″E, with elevation ranges from approximately 1,090 to 1,518 m above mean sea level (Figure 1). Owing to its scenic landscape, moderate climate and strategic location, Abbottabad has historically been an important regional center and served as a hub of British administrative and military activity during the colonial period. Administratively, Abbottabad district comprises two tehsils (Abbottabad and Havellian) and 52 Union Councils. Although the district extends into the Lesser Himalayas, encompassing mountainous terrain with elevations above 2,500 m, the urban area itself lies within a lower-elevation valley. Major streams such as the Harro, Dor and Siran traverse the district, while the River Kunhar flows along its northeastern boundary before joining the River Jhelum. Plains including the Orash Valley, Dor Valley, Dhan Basin and Lora Tract support modest agriculture but are highly susceptible to waterlogging and flooding. Abbottabad experiences intense monsoon rainfall that frequently exceeds the capacity of urban drainage system, resulting in recurrent pluvial flooding. The combination of steep terrain, expanding urbanization and limited drainage infrastructure makes the district highly prone to flood hazards.

FIGURE 1

Abbottabad has a temperate climate with mild summers and cool winters, receiving a long-term average annual rainfall of 1,366 mm (Pakistan Meteorological Department records). Peak precipitations occurs during the monsoon season (July-September), frequently causing urban flooding due to insufficient drainage capacity. The district’s population grew from 880,666 in 1998 to 1,332,912 in 2017, with urban areas expanding at a faster rate (3.32% per year) than rural areas (1.93%). Agriculture occupies about 48% of the district’s land, though only 11% is irrigated. The economy is primarily based on subsistence farming, livestock, tourism and small-scale mining. Low agricultural productivity, limited irrigation, weak industrial growth and scarce employment opportunities contribute to economic vulnerability and reliance on government and private sector jobs. The combination of rapid urban growth, expanding impervious surfaces, high rainfall intensity, complex topography and inadequate drainage infrastructure collectively heightens Abbottabad’s susceptibility to pluvial flooding. These characteristics make the district a well-suited case for assessing urban flood resilience using GIS-based approaches and the Urban Flood Resilience Model (UFRM).

2.2 Methods of data acquisition and analysis

To assess urban flood resilience in Abbottabad, this study relied on a combination of primary and secondary data sources. The integration of field-based information with geospatial datasets enhanced the reliability and applicability of the results. Primary data were collected through field surveys conducted in flood-prone urban areas of Abbottabad. These locations were identified based on historical flood records and GIS-based spatial analysis of high-risk zones. A standardized questionnaire was designed to obtain information on previous flood events, damages, early warning mechanisms and community coping strategies, with emphasis on economic, institutional and educational aspects. The questionnaire was structured to ensure consistency and comparability among responses. In addition, informal interviews and discussions were held with officials from the Water and Sanitation Services Company Abbottabad (WSSCA) to gain insights into drainage infrastructure conditions, maintenance challenges and flood response practices. Field observations were carried out to examine topography, drainage layouts, culverts and flood-prone depressions. The data were processed using Microsoft Excel (MS Excel) and converted into standardized indicators (percentages and density measures) and spatially visualized using ArcGIS. These field-based observations supported the validation of spatial models applied in the analysis.

Secondary datasets were obtained from credible national and international sources to support flood hazard, exposure and susceptibility assessment. Elevation data were derived from ALOS PALSAR Digital Elevation Model (DEM) with a spatial resolution of 12.5 m, accessed via the Alaska Satellite Facility. Land Use Land Cover (LULC) data were derived from Sentinel-2 imagery available through the Copernicus Open Access Hub. Rainfall records were acquired from the Tropical Rainfall Measuring Mission (TRMM), which provide both daily and monthly precipitation estimates. Demographic and socioeconomic data, including population density, income and health facilities and educational institutions, were obtained from the Pakistan Census Report 2017. Infrastructure-related data, such as residential and commercial buildings, were obtained from the Planning and Development Department. All datasets were processed and analyzed using MS Excel and ArcGIS to generate thematic layers required for resilience assessment. To ensure spatial compatibility for weighted overlay analysis, all raster datasets were projected to a similar coordinate system and resampled to a uniform spatial resolution of 12.5 m using the Nearest Neighbor resampling method.

2.3 Urban Flood Resilience Model (UFRM)

The study utilized the Urban Flood Resilience Model (UFRM) formulated and developed by Tayyab et al. (2021), to assess Abbottabad’s capacity to withstand urban flood events. The model comprises two core elements: sensitivity and coping capacity. Sensitivity is represented by three indicators: flood hazard (IH), flood exposure (I.E.,), and flood susceptibility (IS), while coping capacity (ICc) is evaluated through economic capacity, institutional capacity, and educational level (Figure 2). The methodology was implemented in five sequential steps. First, indicators mapping involves the preparation of flood-related indicators in raster format. Second, weight assignment entails assigning weights to each indicator using the AHP. Third, a consistency check is performed using the Consistency Index (CI) and Consistency Ratio (CR) to validate the reliability of the AHP pairwise comparisons. Fourth, all indicators were reclassified using the ArcGIS reclassify tool to produce comparable thematic layers, followed by a weighted overlay analysis in GIS. For the hazard (IH), exposure (I.E.,), and susceptibility (IS) layers, higher classified values represent greater flood risk, whereas for the coping capacity layer (ICc), higher values represent stronger coping capacity. Fifth, the UFRI was computed through raster overlay operations based on Equations 1–3, and the final Urban Flood Resilience Map was generated by classifying resilience levels into five categories: Very Low, Low, Moderate, High, and Very High.

FIGURE 2

The sensitivity component is calculated as:

The coping capacity component is expressed as:

The computation of the Urban Flood Resilience Index (UFRI) is expressed as:

Both terms were assigned identical weights (m₁ = m₂ = 0.5), following the approach of the original model.

2.4 Generation of flood resilience indicators

The analysis incorporated four groups of indicators. The hazard component comprises land use land cover, slope, rainfall, elevation, flood damages and the curve number grid. Exposure is assessed through population density, annual income, health facilities and educational facilities. Susceptibility is evaluated based on residential, commercial and other building types. Coping capacity is measured through economic capacity, educational level and institutional capacity.

2.5 Analytic hierarchy process (AHP)

The AHP was employed to assign relative weights to flood resilience indicators within a GIS framework. Pairwise comparisons were conducted based on expert judgment and relevant literature. The reliability of the comparisons was checked using the Consistency Index (CI) and Consistency Ratio (CR), where CR values less than 0.1 indicated acceptable consistency.

Despite its effectiveness, AHP involves a degree of subjectivity. Additionally, limitations exist in the datasets used, including the coarse resolution of TRMM rainfall data and potential classification errors in Sentinel-2 LULC maps. These limitations should be considered when interpreting the results and point to the need for future use of higher-resolution datasets alongside thorough ground-truthing.

3 Analysis and results

3.1 Urban flood fazard assessment

Flood hazard across Abbottabad was assessed through six spatial indicators: elevation, LULC, curve number, slope, rainfall, and flood damage records. Weighted overlay analysis with AHP-derived weights produced the composite hazard map, classifying the study area into low, medium, and high hazard zones. High hazard is concentrated in the central and northern wards, driven by low elevation, high curve number values, dense built-up cover, and a documented history of flood events in 2019, 2021, and 2022. Medium hazard zones form a transitional belt around these areas, while low hazard extends across the higher-elevation southern and peripheral areas.

Terrain characteristics exert the strongest control on hazard distribution. Elevation, derived from the ALOS PALSAR DEM, was classified into five zones across the study area (Figure 3a). Valley floors and low-lying depressions accumulate runoff from surrounding slopes and record the highest hazard scores, whereas zones above approximately 1,500 m above mean sea level facilitate rapid runoff dispersal and contribute low hazard values. The central and northern urban fabric of Abbottabad, confined to a valley pocket, is disproportionately represented in the lowest elevation class, directly elevating its hazard ranking. Land surface conditions further amplify hazard in the urbanized core. Supervised classification of Sentinel-2 imagery identified three dominant land cover classes: built-up surfaces, vegetated land, and barren land (Figure 3b). Built-up areas in the central and northern wards generate high surface runoff owing to their impervious character, while vegetated areas on hillsides and peripheries retain moisture and dampen runoff.

FIGURE 3

Similarly, the Curve Number (CN) layer was created in ArcGIS. It combined the Land Use/Land Cover map with the Hydrologic Soil Group (HSG) layer, applying the Soil Conservation Service (SCS) Curve Number method (Figure 3c). The HSG data were derived from the FAO soil database and classified using the USDA Soil Conservation Service classification system. The study area was classified mainly as Hydrologic Soil Group B, representing soils with moderate infiltration capacity. The LULC classes were assigned standard SCS Curve Number values to generate the final CN raster. This raster served as one of the flood hazard indicators in the UFRM, consistent with unified runoff generation approaches applied across comparable hydro-meteorological zones (Zhang et al., 2024).

Slope, also derived from the ALOS PALSAR DEM, was classified into five categories (Figure 3d). Gentle gradients of 0°–4.24° dominate the central valley floor and northern urban areas, promoting water stagnation and prolonged inundation. Steeper gradients of 35.6°–67.5°, confined to the surrounding hill ranges, accelerate runoff and increase downstream discharge into the flat valley zones, consistent with observed rainfall-induced seepage and runoff dynamics on steep bedrock slopes.

Empirical validation was provided through questionnaire surveys across flood-prone localities, which recorded damage severity reported by residents as percentage values assigned to administrative units and mapped in ArcGIS (Figure 3e). Bilal Town, Shahzaman Colony, and Sir Syed Colony recorded the highest damage proportions, with over 60% of respondents reporting structural damage or prolonged inundation during the 2019 and 2021 events. Peripheral and higher-elevation localities reported low to negligible damage, consistent with the composite hazard classification. Hydro-meteorological inputs reinforce the hazard gradient established by terrain and land surface. TRMM rainfall data interpolated using the Inverse Distance Weighting method in ArcGIS produced a precipitation intensity map (Figure 3f). The north-eastern portions of the district receive the highest annual rainfall totals within the TRMM-derived surface, reaching 1,472–1,498 mm values that already exceed the city’s long-term mean annual rainfall. This spatial gradient aligns with the monsoon trajectory, reinforcing flood hazard in areas already most exposed by topography and land use.

3.2 Urban flood exposure

Flood exposure was assessed using four socioeconomic indicators: population density, annual income, health facilities, and education facilities (Figure 4). The composite exposure map shows the highest values in the central and northern urban wards, where dense populations, lower household incomes, and limited service coverage converge. Southern and peripheral zones show markedly lower exposure owing to lower densities and better service ratios.

FIGURE 4

Household income and population density constitute the primary demographic dimensions of exposure. Income data from the Pakistan Census 2017 were categorized into five levels and mapped across the 52 union councils (Figure 4a). Low-income areas are concentrated in the eastern and south-eastern wards and account for the largest share of flood-exposed households, while higher-income zones are confined to the central business districts and western residential areas. Annual income received the highest AHP weight (0.5585) among the four exposure indicators, reflecting the strong relationship between financial capacity and flood recovery. Population density, also derived from census data and mapped in five classes (Figure 4b), reinforces this pattern spatially. The highest densities, reaching 10,063–11,299 persons/km², are recorded in the inner urban wards of the central and northern city, which overlap directly with the high hazard zones identified in Section 3.1, placing a large resident population at elevated flood risk. The outer wards, with densities as low as 709 persons/km², show correspondingly lower exposure values in the composite map.

Service infrastructure coverage further compounds exposure in the most hazard-prone areas. Health facilities, including hospitals, clinics, and dispensaries, were mapped across the 52 union councils using census data (Figure 4c). These facilities are concentrated in the central and southern city, whereas the northern and peripheral wards, which face moderate to high flood hazard, have markedly fewer facilities per capita. This spatial mismatch raises the risk of inadequate emergency response in exactly the areas most likely to need it. A comparable pattern appears for education facilities, mapped by union council using Bureau of Statistics data on schools, colleges, and universities (Figure 4d). The central and northern zones have 24–28 institutions per council, while southern and western union councils have only 6–15, the lowest coverage in the district. Low educational facility density limits disaster awareness and reduces communities’ capacity to respond effectively when flooding occurs.

3.3 Urban flood susceptibility

Susceptibility was evaluated using the type and density of built-up structures across the 52 union councils. Three categories were assessed: residential, commercial, and other buildings. Building footprint data were classified using the natural breaks method and weighted through AHP before integration into the susceptibility layer (Figure 5). The highest susceptibility values are concentrated in the central urban belt, where residential and commercial density is greatest and drainage infrastructure is most overstressed.

FIGURE 5

Residential buildings constitute the largest built-up category and carry the highest AHP weight (0.633) within the susceptibility component. Building counts were classified into five density categories and mapped by union council (Figure 5a). The highest densities are recorded in the central and northern wards, where buildings are closely packed on low-lying ground with inadequate drainage. Their spatial overlap with high hazard and high exposure zones produces compounding vulnerability concentrated in the same geographic footprint. Commercial building density, classified into graded intervals across the union councils (Figure 5b), follows a broadly similar pattern, with activity concentrated in the inner-city central wards where market streets and mixed-use developments dominate. Flood inundation in these zones disrupts livelihoods, supply chains, and local revenue simultaneously, making them high-priority targets for drainage improvement and flood-proofing measures.

Other buildings were mapped separately and classified into five density classes (Figure 5c). Their distribution is more dispersed than that of residential or commercial structures, with moderate concentrations in the southern and eastern wards. These facilities carry a dual role: as physical assets they contribute to the susceptibility score, and as functional nodes they serve as evacuation shelters and relief distribution points. Their sparse coverage in the high-hazard northern wards limits emergency support availability where it is most needed.

3.4 Extent and evaluation of urban flood coping capacity

3.4.1 Coping capacity indicators

Coping capacity was evaluated using three indicators drawn from primary field surveys and secondary sources: educational level, institutional capacity, and economic capacity (Figure 6). Together these indicators reflect the degree to which communities and local institutions can prepare for, respond to, and recover from flood impacts. The composite coping capacity map shows the lowest values in the eastern and south-eastern wards and the strongest coping capacity in the south-western urban areas.

FIGURE 6

Economic capacity, derived from questionnaire data on household income, savings, and credit access, was aggregated by union council and classified into low, moderate, and high levels (Figure 6a). The south-western wards returned the highest proportions of moderate and high economic capacity, reflecting stronger formal employment. The eastern and south-eastern wards, by contrast, show predominantly low economic capacity, with most surveyed households reporting no savings and reliance on daily-wage labour. These households face the greatest barriers to flood recovery and received the lowest economic capacity scores in the composite index. Institutional capacity was assessed through survey responses on local government presence, early warning mechanisms, and disaster response infrastructure, then similarly classified and mapped by union council (Figure 6b). The central and northern wards benefit from proximity to the district administration and WSSCA, giving them moderate to high institutional capacity. The eastern and south-eastern wards, however, report weak local governance and no functional early warning systems, producing the lowest institutional scores in the city. Institutional capacity received the highest AHP weight (0.669) within the coping capacity component, followed by economic capacity (0.243) and educational level (0.088), reflecting its critical role in coordinating flood preparedness and response.

Educational attainment, surveyed and classified into low, moderate, and high categories by union council (Figure 6c), reinforces the spatial pattern established by the preceding two indicators. The central and northern wards, which contain the majority of the city’s schools and colleges, show predominantly high attainment among flood-risk households. The south-eastern union councils show the lowest attainment, with most surveyed residents having not completed secondary education. Lower attainment reduces awareness of flood early warning signals and the capacity to take protective action, compounding the vulnerability already indicated by weak economic and institutional scores in those wards.

3.4.2 AHP weighting and consistency verification

An integrated GIS based framework combining flood hazard, exposure, susceptibility and coping capacity was employed in this study to evaluate overall urban flood resilience. Each component was analysed individually and then integrated to generate the Urban Flood Resilience Index (UFRI). To assign appropriate weights to all indicators, the study employed AHP to ensure a systematic and objective assessment. The final resilience maps provide a spatial overview of flood resilience levels across the study area, showing zones with high adaptive capacity alongside areas that remain highly vulnerable. These results support informed decision-making by identifying priority locations for flood mitigation, infrastructure improvement and resilience planning. Saaty (1980) developed the AHP, which is a multi-criteria decision-making method for evaluating complex problems involving multiple criteria. AHP was applied in this research to assess the relative weighting of flood resilience indicators by categorizing them and conducting pairwise comparisons on a 1–9 scale. A comparison matrix was created and normalized to generate final weights and the Consistency Ratio (CR) was evaluated to ensure logical consistency, with values less than 0.1 being acceptable. The validated weights were then combined with a GIS-based overlay analysis to generate the Urban Flood Resilience Index.

The pairwise comparison matrix was constructed for the four main components of the AHP framework: hazard, exposure, susceptibility and coping capacity, to derive relative weights for each criterion. Saaty’s (1980) standard 1–9 scale (Table 1) was applied, where higher values indicate greater relative importance. The pairwise comparisons were developed based on an extensive review of the relevant literature and the conceptual importance of each indicator within the Urban Flood Resilience Model (UFRM), together with structured consultation with a panel of experts. The expert panel was purposively selected to represent the range of institutions engaged in flood risk management in Abbottabad. It comprised representatives from the Provincial Disaster Management Authority (PDMA) Khyber Pakhtunkhwa, humanitarian organizations active in local disaster response, the Tehsil Municipal Administration (TMA) Abbottabad, and academic experts from institutions within the city. Each expert independently scored the pairwise comparisons using Saaty’s 1–9 scale, and the resulting judgments were averaged to derive the final priority weights used in the AHP model. The resulting weights were subsequently integrated with the GIS-based overlay analysis to generate the Urban Flood Resilience Index (UFRI). The following subsections describe the weighting process for each component.

To confirm the consistency of the pairwise comparisons, the study conducted a reliability evaluation using the Consistency Index (CI) and Consistency Ratio (CR). The CR was calculated following the standard AHP formula:

According to Saaty’s criterion, a CR value equal to or smaller than 0.1 indicates acceptable consistency; values exceeding this threshold suggest the need for revision of judgments.

The Consistency Index was computed using:

In this equation, n denotes the number of criteria. To determine λmax, multiply the column sum of each criterion by its respective priority vector weight, and then sum all these products to obtain the λmax value, as presented in Table 2.

To compute the Consistency Index for all four parameters, the λmax values from Table 2 were substituted into Equation 5. Subsequently, to determine the Consistency Ratio for each parameter, the calculated CI value along with the corresponding Random Index (RI) based on the number of criteria were applied in Equation 4 (CR = CI/RI). The calculated CR values for hazard, exposure, susceptibility and coping capacity were 0.06, 0.061, 0.04 and 0.01, respectively. As all values fall below the acceptable threshold of 0.1, the pairwise comparisons are deemed consistent and suitable for subsequent analysis.

Pairwise comparisons were conducted for all four UFRM components and the derived AHP weights are consolidated in Table 3. For the Hazard component, Elevation received the highest weight (0.4221) followed by Slope (0.2294), reflecting their dominant influence on surface runoff and flood accumulation. For Exposure, Annual Income was assigned the greatest weight (0.5585), as low-income communities bear the highest flood burden. For Susceptibility, Residential Buildings carried the largest weight (0.633) given their greater concentration and damage potential. For Coping Capacity, Institutional Capacity received the highest weight (0.669), followed by Economic Capacity (0.243) and Educational Level (0.088), reflecting its central role in preparedness, response and recovery consistent with the ranking reported for the coping capacity component of the original UFRM (Tayyab et al., 2021), in which institutional capacity was likewise assigned the highest importance. The CR and CI values for all components are presented in Table 3 and full λmax calculations are shown in Table 2.

3.5 Application of Urban Flood Resilience Model

In this study, the UFRM is structured around two key components: sensitivity and coping capacity. Sensitivity integrates hazard, exposure and susceptibility to identify areas most vulnerable to urban flooding, while coping capacity reflects the ability of communities and institutions to anticipate respond to and recover from flood events. Using the AHP, relative weights were assigned to all indicators. A weighted overlay analysis was conducted within a GIS environment to generate individual component maps, which were subsequently integrated to produce the final Urban Flood Resilience Map for Abbottabad City.

3.5.1 Urban flood sensitivity

Sensitivity represents the degree to which an area is prone to flood impacts and was evaluated by integrating hazard, exposure and susceptibility layers (Figure 7). Each component was analyzed independently and later combined to assess overall flood sensitivity across the city.

FIGURE 7

The urban flood hazard was assessed using six key factors including LULC, elevation, rainfall, slope, historical flood damage and curve number (CN) grid. These parameters capture the hydrological, topographic and land-use conditions influencing flood occurrence. AHP-based weights were applied and the layers were integrated using weighted overlay analysis to produce the hazard map. The flood hazard map classifies the study area into low, medium and high hazard areas (Figure 7a). High hazard zones are concentrated in the northern and central parts of Abbottabad, characterized by dense urban development, steep slopes, high rainfall intensity and a history of flood damage. Medium hazard zones surround these areas, while low hazard zones are largely located in the southern and peripheral regions with relatively higher elevation and less vulnerable land use. The concentration of several hazard-driving factors across the northern and central regions indicates a higher flood risk. However, the final level of urban flood resilience is determined by the combined effects of hazard, exposure, susceptibility, and coping ability. As a result, the hazard map should be interpreted together with the other resilience components, rather than separately.

The urban flood exposure was evaluated using four socio-economic indicators including population density, annual income, educational facilities and health facilities. These factors reflect the concentration of people and critical infrastructure in flood-prone areas. After standardization and weighting through AHP, a composite exposure map was generated (Figure 7b).

Figure 7b indicates that high exposure zones are predominantly located in the central and northern urban areas, where dense populations, lower income groups and essential services are concentrated. These areas are more likely to experience severe human and economic losses during flood events. Medium exposure zones extend around the urban core, while low exposure areas are mainly found in peripheral and semi-rural locations. The spatial concentration of people and vital infrastructure increases the risk of floods in these urban areas. However, exposure alone does not define overall flood resilience and must be understood together with hazard, susceptibility, and coping capability to fully comprehend the spatial variation in resilience over Abbottabad.

The urban flood susceptibility was assessed based on the distribution of residential, commercial and other buildings. These factors represent the physical vulnerability of built-up areas to flood damage (Figure 7c). Each building category was evaluated based on its flood sensitivity and analyzed using AHP-derived weights. Figure 7c indicates that highly susceptible zones are concentrated in the central urban belt, where dense residential and commercial developments dominate and drainage infrastructure is inadequate. Medium susceptibility areas occur in transitional urban zones, while low susceptibility areas are found in newly developed or planned neighborhoods toward the south-western outskirts of the city. The concentration of densely built-up regions raises the risk of physical damage during floods, especially where drainage capacity is restricted. However, susceptibility is only one dimension of flood resilience and should be understood in conjunction with hazard, exposure, and coping ability to better understand the overall resilience pattern in the studied area.

The urban flood coping capacity was assessed using three indicators: economic capacity, institutional capacity and educational level. These parameters represent the social and institutional ability of communities to manage and recover from flood impacts. AHP was used to assign relative importance to each indicator and a weighted overlay analysis was applied. The coping capacity map reveals that areas with low coping capacity are mainly located in the eastern and southeastern parts of Abbottabad (Figure 7d), where limited economic resources, weak institutional support and lower education levels reduce flood resilience. Medium coping capacity is observed across much of the city, while high coping capacity zones are concentrated in the southwestern urban areas, benefiting from better access to education, municipal services and economic opportunities. These findings indicate that coping capacity varies greatly across the city and has a significant impact on overall urban flood resistance. Even in areas with significant physical flood risk, those with stronger institutional, educational, and economic capacities are better able to survive and recover from flood catastrophes.

3.5.2 Urban flood resilience of Abbottabad City

The final Urban Flood Resilience Map was generated through a raster overlay approach, integrating the four principal components of the UFRM: hazard, exposure, susceptibility and coping capacity. Each component was created using weighted overlay analysis with AHP-derived weights and then integrated through Equations 1–3 to produce a composite Urban Flood Resilience Index (UFRI). The resulting raster values ranged from 0 to 1, where higher UFRI values indicate higher flood resilience and lower values represent lower resilience. To enhance interpretation, the resilience index was reclassified into five classes (very high, high, medium, low and very low) using the ArcGIS Reclassify tool. Accordingly very high resilience areas correspond to the highest raster values. The spatial distribution of resilience across Abbottabad is illustrated in Figure 8. Areas classified as low to very low resilience are mainly concentrated in densely populated residential and commercial zones characterized by high exposure and limited coping capacity. In contrast, medium to very high resilience areas are observed in regions with lower hazard intensity and stronger coping mechanisms, including better infrastructure, institutional preparedness and socio-economic conditions. The resilience map provides a spatial decision-support tool for identifying priority areas for flood risk reduction and urban resilience planning.

FIGURE 8

The northern and central parts of Abbottabad are more prone to hazard, exposure, and susceptibility. However, they also have a significantly higher coping capacity. This is evidenced by improved drainage infrastructure, institutional preparedness, emergency response services, road accessibility, and socioeconomic conditions. The UFRM assigns equal weights to sensitivity and coping capacity when determining flood resilience (m₁ = m₂ = 0.5). As a result, the higher coping capacity in these areas offsets the effects of increased flood sensitivity, giving comparatively higher overall resilience. This shows that resilience, as captured by the UFRI, depends not only on the magnitude of flood risk but also on the ability of communities and institutions to prepare for, respond to, and recover from flood events.

3.5.3 Validation

Model validation was executed by field observations at three indicative flood-prone sites in Abbottabad during the monsoon period. These areas were deliberately picked as they are consistently impacted by urban flooding and are often emphasized by local authorities and news coverage. The noted field conditions were contrasted with the conclusive Urban Flood Resilience Map, accompanied by example images in Figure 9. However, the validation is qualitative in nature. Quantitative methods such as ROC analysis, confusion matrix analysis, or comparison with historical flood inundation data could not be applied because appropriate flood inventory data for Abbottabad were not available. The findings of this study should be considered as qualitative evidence rather than a thorough quantitative accuracy assessment.

FIGURE 9

4 Discussion

Urban flooding poses a serious threat to environmental sustainability, infrastructure stability, and public health in rapidly urbanizing cities of developing countries (Bertilsson et al., 2019; Chan et al., 2021). The findings of this study are consistent with the broader literature. Unplanned urban expansion, increased impervious surfaces, and inadequate drainage infrastructure intensify flood risk by altering natural hydrological processes (Zhou Q. et al., 2019; Bulti and Abebe, 2020). These pressures are compounded by an observed intensification of monsoon precipitation extremes across northern Pakistan, driven by anthropogenic climate change. This trend is projected to continue, increasing the frequency of high-magnitude rainfall events in the region (Bhatti et al., 2020; IPCC, 2021). The replacement of permeable land with concrete surfaces accelerates surface runoff, and climate change further aggravates flood hazard through more intense and unpredictable rainfall (Bhunya et al., 2020; Marengo et al., 2020). These combined pressures are well documented globally (Hu et al., 2018; Merz et al., 2021) and are clearly evident in Abbottabad. Integrated flood management strategies are therefore needed, extending beyond infrastructure upgrades to land-use planning and governance reform.

These challenges are clearly evident in Abbottabad, where rapid urban growth and encroachment on natural drainage channels, such as Jab Nullah, have increased the frequency and severity of urban flooding. Major flood events in 2019, 2021 and 2022 caused extensive damage to roads, residential areas and critical facilities, including Ayub Teaching Hospital. The July 2021 flood, triggered by intense rainfall over a short duration, resulted in widespread inundation, underscoring the consequences of poor drainage maintenance and unregulated development in vulnerable neighborhoods such as Bilal Town, Shahzaman Colony and Sir Syed Colony.

The findings indicate that urban flood resilience in Abbottabad depends on both physical flood conditions and communities’ ability to cope with flood episodes, not on flood threat alone. The strong weighting of elevation and slope shows that topography plays a major role in flood behaviour in mountainous settings, consistent with earlier research (Suriya and Mudgal, 2012; Bahrawi et al., 2020). At the same time, the strong contribution of annual income and institutional capacity points to the relevance of socioeconomic conditions and effective governance in mitigating flood damage. Similar patterns have been reported elsewhere in Pakistan, where communities with better institutional support and economic resources proved more resilient to flood hazards (Maqbool et al., 2017; Shah et al., 2021; Khan et al., 2023; Shakoor et al., 2023). These findings suggest that enhancing urban flood resilience in Abbottabad requires more than structural measures such as drainage improvements. Long-term investment in institutional preparedness, community awareness, and socioeconomic growth is also needed.

This study’s conclusions can be better understood by comparing them with earlier GIS-based flood resilience studies. Tayyab et al. (2021) applied the UFRM in Peshawar and found that flood susceptibility was most concentrated in low-lying urban areas. By contrast, the current study highlights the role of mountainous terrain, where steep slopes and complex topography strongly shape flood behaviour and resilience patterns. Similarly, Gabor & Tibor (2023) found that densely built metropolitan areas with weaker institutional support had lower resilience, underlining the role of governance and urban design in mitigating flood risk. The current findings also align with Jamali et al. (2023), who found that institutional capacity is often a stronger contributor to flood resilience than economic factors alone. Bertilsson et al. (2019) and Luu et al. (2021) likewise stressed the value of incorporating physical, social, economic, and institutional variables into GIS-based frameworks for evidence-based urban flood management. Overall, the present findings are consistent with this literature. They confirm that multi-criteria spatial techniques provide a sound basis for identifying priority locations and informing flood resilience planning.

Despite these structural vulnerabilities, Abbottabad shows relatively stronger coping capacity in certain areas, notably in educational awareness and community response, compared with similar high-density zones documented by Tayyab et al. (2021) in Peshawar. This supports the argument advanced by Sharifi and Yamagata (2018) that social capital and educational attainment can partially offset physical vulnerability. Seidler et al. (2018) similarly found that communities with higher literacy levels and stronger social networks recovered more rapidly from flood events in South Asian cities. This underscores the value of educational investment as a resilience-building strategy alongside physical infrastructure. The qualitative field validation conducted in this work revealed a strong correlation between the mapped resilience patterns and the observed flood conditions in Bilal Town, Shahzaman Colony, and Sir Syed Colony. However, this validation relied on field observations and photographic evidence rather than quantitative accuracy assessment. It should therefore be viewed as qualitative support for the UFRI’s applicability, not as a thorough test of model accuracy.

From a methodological standpoint, the application of the UFRM proved effective in capturing both physical and socioeconomic dimensions of urban flood resilience, consistent with the multi-criteria framework advocated by Bertilsson et al. (2019) and Jamali et al. (2023). The AHP-derived weights, all with CR values below 0.1, provide a transparent and reproducible basis for indicator prioritization, addressing a key criticism of subjective expert-based weighting raised by Sisman and Aydinoglu (2020). However, several limitations must be addressed. Although all spatial datasets were harmonized to a common spatial resolution of 12.5 m before GIS analysis, the TRMM rainfall dataset has an inherently coarse resolution (0.25°, approximately 27 km). It may therefore miss localized rainfall variability and short-duration convective storms in Abbottabad (Fan et al., 2017). Similarly, the exposure analysis relied on the 2017 population census, which may not reflect the city’s current population distribution given its rapid growth. These sources of uncertainty are consistent with broader challenges in process-based hydrological and hydraulic modeling under data-scarce conditions (Zhao et al., 2023). Additionally, the susceptibility component relied solely on building type and density without incorporating structural quality data, which could refine damage estimates. Floods in Abbottabad also vary seasonally, with most incidents occurring during high monsoon rainfall. The resilience evaluation therefore reflects the available datasets and prevailing conditions rather than seasonal variation in flood behaviour. In addition, the selected indicators were drawn from the Urban Flood Resilience Model (UFRM) and validated by the literature, but adding or removing indicators could affect the resulting resilience patterns. Future research should test alternative indicator sets and weighting methods to reduce this uncertainty further. To improve the robustness and accuracy of urban flood resilience assessments, future studies should include higher-resolution rainfall datasets (e.g., CHIRPS or local meteorological station data), updated population datasets, and dynamic flood simulation models such as SWMM (Xu et al., 2019) and HEC-RAS (Mustaffa and Talib, 2022). The UFRM framework demonstrated clear methodological suitability for Abbottabad’s mountainous context and provides a transferable template for flood resilience assessment in comparable cities across Pakistan and South Asia.

5 Conclusion

This study applied the Urban Flood Resilience Model (UFRM) within an integrated GIS-AHP framework. It produced the first spatially explicit, multi-dimensional assessment of urban flood resilience for Abbottabad City, Khyber Pakhtunkhwa, Pakistan. Sixteen indicators across four components (flood hazard, exposure, susceptibility, and coping capacity) were quantified and weighted using AHP (all CR values <0.1), then integrated through weighted overlay analysis to generate the UFRI for each union council. The UFRI map reveals pronounced spatial heterogeneity in flood resilience across the city. The densely built-up central and northern wards face the highest flood hazard, population exposure, and physical susceptibility in older, informally developed neighborhoods, yet their stronger institutional preparedness, drainage infrastructure, and socioeconomic capacity offset much of this risk, producing comparatively higher overall resilience. By contrast, the eastern and south-eastern wards combine considerable hazard exposure with critically weak coping capacity, arising from limited institutional presence, low incomes, and lower educational attainment, and consequently register the lowest resilience in the city. These zones, including Bilal Town, Shahzaman Colony, and Sir Syed Colony, are the highest-priority areas for targeted intervention. The south-western and peripheral zones show medium to high resilience, reflecting lower hazard intensity, less dense development, and stronger institutional capacity. Qualitative field validation during the monsoon season successfully supported the spatial correspondence between model-classified low-UFRI zones and observed waterlogging, drainage failure, and infrastructure stress. While quantitative inventory data for Abbottabad are currently limited, the GIS-AHP-UFRM framework proved to be a highly reliable and transferable tool for flood resilience assessment in data-scarce, mountainous urban environments. The findings provide a practical, spatially explicit decision-support tool to guide localized disaster risk reduction, infrastructure upgrades, and urban planning efforts in Abbottabad. Priority interventions should target three areas: (i) rehabilitation and expansion of stormwater drainage infrastructure in high-hazard central and northern wards; (ii) institutional strengthening of local disaster management bodies, with emphasis on early warning systems; and (iii) community-based preparedness programs and literacy campaigns in the low-coping-capacity south-eastern union councils. Future research should incorporate dynamic flood simulation models, higher-resolution datasets, quantitative validation methods, and participatory community surveys, both to refine the UFRM and to extend the framework to comparable growing cities across Khyber Pakhtunkhwa and South Asia.

References

• 1

  AlviS.KhayyamU. (2020). Mitigating and adapting to climate change: attitudinal and behavioural challenges in south Asia. Int. J. Clim. Change Strategies Manag.12 (4), 477–493. 10.1108/ijccsm-08-2019-0054

  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  BahrawiJ.EweaH.KamisA.ElhagM. (2020). Potential flood risk due to urbanization expansion in arid environments, Saudi Arabia. Nat. Hazards104 (1), 795–809. 10.1007/s11069-020-04190-7

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  BertilssonL.WiklundK.de Moura TebaldiI.RezendeO. M.VerólA. P.MiguezM. G. (2019). Urban flood resilience–A multi-criteria index to integrate flood resilience into urban planning. J. Hydrology573, 970–982. 10.1016/j.jhydrol.2018.06.052

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  BhattiA. S.WangG.UllahW.UllahS.Fiifi Tawia HaganD.Kwesi NooniI.et al (2020). Trend in extreme precipitation indices based on long-term in situ precipitation records over Pakistan. Water12 (3), 797. 10.3390/w12030797

  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  BhunyaP. K.KumarS.GurrapuS.BhuyanM. K. (2020). A review of case studies on climate change impact on hydrologic cycle: an Indian perspective. Int. J. Sci. Res. Sci. Eng. Technol.7 (5), 249–266. 10.32628/ijsrset207548

  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  BultiD. T.AbebeB. G. (2020). A review of flood modeling methods for urban pluvial flood application. Model. Earth Systems Environment6, 1293–1302. 10.1007/s40808-020-00803-z

  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  ChanF. K. S.YangL. E.ScheffranJ.MitchellG.AdekolaO.GriffithsJ.et al (2021). Urban flood risks and emerging challenges in a Chinese Delta: the case of the pearl river Delta. Environ. Sci. and Policy122, 101–115. 10.1016/j.envsci.2021.04.009

  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  FanY.AoT.YuH.HuangG.LiX. (2017). A coupled 1D–2D hydrodynamic model for urban flood inundation. Adv. Meteorology2017, 1–12. 10.1155/2017/2819308

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  GáborK.TiborL. (2023). Urban flood resilience analysis using GIS-based AHP and modified S-FRESI approach: a case study of Gyor city, Hungary. Water15 (6), 1072.

  • Google Scholar
  • View reference in article

• 10

  HeL.YinY.CaiJ.WuL.TangY.HeR.et al (2025). Flood inundation amplified by large-scale ground subsidence funnel under the ongoing global climate change. Commun. Earth and Environ.6 (1), 718. 10.1038/s43247-025-02669-4

  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  HuP.ZhangQ.ShiP.ChenB.FangJ. (2018). Flood-induced mortality across the globe: spatiotemporal pattern and influencing factors. Sci. Total Environ.643, 171–182. 10.1016/j.scitotenv.2018.06.197

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  Intergovernmental Panel on Climate Change (IPCC) (2021). Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press. 10.1017/9781009157896

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  JamaliA.RobatiM.NikoomaramH.FarsadF.AghamohammadiH. (2023). Urban resilience assessment using hybrid MCDM model based on DEMATEL-ANP method (DANP). J. Indian Soc. Remote Sens.51 (4), 893–915. 10.1007/s12524-023-01670-8

  • CrossRef
  • Google Scholar

• 14

  JiangJ.YangY.ChenY. (2026). An integrated approach to evaluating and prioritizing socio-physical flooding mitigation planning to enhance resilience in a community. Reliab. Eng. and Syst. Saf.271, 112303. 10.1016/j.ress.2026.112303

  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  KhanI.LeiH.ShahA. A.KhanI.MuhammadI. (2021). Climate change impact assessment, flood management and mitigation strategies in Pakistan for sustainable future. Environ. Sci. Pollut. Res.28, 29720–29731. 10.1007/s11356-021-12801-4

  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  KhanA.GongZ.ShahA. A.HaqM. (2023). A multi-criteria decision-making approach to vulnerability assessment of rural flooding in Khyber Pakhtunkhwa province, Pakistan. Environ. Sci. Pollut. Res.30 (19), 56786–56801. 10.1007/s11356-023-25609-1

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  KhurramA. (2022). Problem Identification: Examining the Factors that Resulted in Islamabad's Unsustainable Development (Master’s thesis, Ankara, Turkey: Middle East Technical University (METU), Graduate School of Social Sciences).

  • Google Scholar
  • View reference in article

• 18

  LuuC.BuiQ. D.CostacheR.NguyenL. T.NguyenT. T.Van PhongT.et al (2021). Flood-prone area mapping using machine learning techniques: a case study of Quang Binh province, Vietnam. Nat. Hazards108 (3), 3229–3251. 10.1007/s11069-021-04821-7

  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  MaqboolM. Y.HussainS.KhanM. B. (2017). National framework of disaster risk management in Pakistan: issues, challenges and policy recommendations. Abasyn J. Soc. Sci.10, 182–192.

  • Google Scholar
  • View reference in article

• 20

  MarengoJ. A.AlvesL. M.AmbrizziT.YoungA.BarretoN. J.RamosA. M. (2020). Trends in extreme rainfall and hydrogeometeorological disasters in the metropolitan area of São Paulo: a review. Ann. N. Y. Acad. Sci.1472 (1), 5–20. 10.1111/nyas.14307

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  MengX.ZhangM.WenJ.DuS.XuH.WangL.et al (2019). A simple GIS-Based model for urban rainstorm inundation simulation. Sustainability11 (10), 2830. 10.3390/su11102830

  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  MerzB.BlöschlG.VorogushynS.DottoriF.AertsJ. C.BatesP.et al (2021). Causes, impacts and patterns of disastrous river floods. Nat. Rev. Earth and Environ.2 (9), 592–609. 10.1038/s43017-021-00195-3

  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  MoazzamM. F. U.LeeB. G.RahmanA. U.FaridN.RahmanG. (2020). Spatio-statistical analysis of flood susceptibility assessment using bivariate model in the floodplain of river swat, district charsadda, Pakistan. J. Geoscience Environ. Prot.8 (5), 159–175. 10.4236/gep.2020.85010

  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  MustaffaM. H. B.TalibS. H. A. (2022). Flood mapping by using hydrologic engineering center river analysis system (HEC-RAS) in kota tinggi, Johor, Malaysia. Recent Trends Civ. Eng. Built Environ.3 (1), 1839–1849.

  • Google Scholar
  • View reference in article

• 25

  RahmanG.AnwarM. M.AhmedM.AshrafH.ZafarU. (2017). Socio-economic damages caused by the 2014 flood in Punjab province, Pakistan. Proc. Pak. Acad. Sci. B. Life Environ. Sci.54 (4), 365–374.

  • Google Scholar
  • View reference in article

• 26

  SaatyT. L. (1980). The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation. New York: McGraw-Hill.

  • Google Scholar
  • View reference in article

• 27

  SeidlerR.DietrichK.SchweizerS.BawaK. S.ChopdeS.ZamanF.et al (2018). Progress on integrating climate change adaptation and disaster risk reduction for sustainable development pathways in south Asia: evidence from six research projects. Int. Journal Disaster Risk Reduction31, 92–101. 10.1016/j.ijdrr.2018.04.023

  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  ShahA. A.WuW.GongZ.PalI.KhanJ. (2021). Multidimensional six-stage model for flood emergency response in schools: a case study of Pakistan. Nat. Hazards105, 1977–2005. 10.1007/s11069-020-04386-x

  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  ShakoorU.ShahS. A.UllahI. (2023). Investigating spatial patterns and contributing factors of rural poverty and vulnerability to environmental threats in Khyber Pakhtunkhwa, Pakistan. J. Soc. Econ. Dev.26, 1–20. 10.1007/s40847-023-00271-y

  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  SharifiA.YamagataY. (2018). “Resilience-oriented urban planning,” in Resilience-Oriented Urban Planning: Theoretical and Empirical Insights, 3–27.

  • Google Scholar
  • View reference in article

• 31

  SismanS.AydinogluA. C. (2020). Using GIS-Based multi-criteria decision analysis techniques in the smart cities. Int. Archives Photogrammetry, Remote Sens. Spatial Inf. Sci.44, 383–389. 10.5194/isprs-archives-xliv-4-w3-2020-383-2020

  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  SuriyaS.MudgalB. V. (2012). Impact of urbanization on flooding: the thirusoolam sub Watershed–A case study. J. Hydrology413, 210–219. 10.1016/j.jhydrol.2011.05.008

  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  TayyabM.ZhangJ.HussainM.UllahS.LiuX.KhanS. N.et al (2021). GIS-Based urban flood resilience assessment using urban flood resilience model: a case study of peshawar city, Khyber Pakhtunkhwa, Pakistan. Remote Sens.13 (10), 1864. 10.3390/rs13101864

  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  UllahW.WangG.LouD.UllahS.BhattiA. S.UllahI.et al (2021). Large-scale atmospheric circulation patterns associated with extreme monsoon precipitation in Pakistan during 1981–2018. Atmos. Res.253, 105489. 10.1016/j.atmosres.2021.105489

  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  Unisdr. (2011). Global Assessment Report on Disaster Risk Reduction 2011: Revealing Risk, Redefining Development. United Nations International Strategy for Disaster Reduction (UNISDR), Geneva, Switzerland.

  • Google Scholar
  • View reference in article

• 36

  XuH.XuK.LianJ.MaC. (2018). Compound effects of rainfall and storm tides on coastal flooding risk. Stoch. Environ. Res. Risk Assess.33 (7), 1249–1261. 10.1007/s00477-019-01695-x

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  XuZ.XiongL.LiH.XuJ.CaiX.ChenK.et al (2019). Runoff simulation of two typical urban green land types with the stormwater management model (SWMM): sensitivity analysis and calibration of runoff parameters. Environ. Monitoring Assessment191, 1–16. 10.1007/s10661-019-7445-9

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  ZhangQ.ZhangK.ChaoL.ChenX.WuN. (2024). A unified runoff generation scheme for applicability across different hydrometeorological zones. Environ. Model. and Softw.180, 106138. 10.1016/j.envsoft.2024.106138

  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  ZhaoY.WangH.SongB.XueP.ZhangW.PethS.et al (2023). Characterizing uncertainty in process-based hydraulic modeling, exemplified in a semiarid Inner Mongolia steppe. Geoderma440, 116713. 10.1016/j.geoderma.2023.116713

  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  ZhouW.CaoF.WangG. (2019). Effects of spatial pattern of forest vegetation on urban cooling in a compact megacity. Forests1 (1), 1–24. 10.3390/f10030282

  • CrossRef
  • Google Scholar
  • View reference in article

• 41

  ZhouQ.LengG.SuJ.RenY. (2019). Comparison of urbanization and climate change impacts on urban flood volumes: importance of urban planning and drainage adaptation. Sci. Total Environ.658, 24–33. 10.1016/j.scitotenv.2018.12.184

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article