Abstract

Urban flooding presents a critical challenge worldwide, driven by rapid urbanization, inadequate drainage systems, and climate change. In developed countries, aging infrastructure and increasing precipitation intensities worsen flood risks. In contrast, developing nations struggle with uncontrolled urban sprawl, poor stormwater and waste management, and limited financial resources for flood mitigation. Pakistan’s major cities, including Peshawar, frequently experience urban flooding, resulting in infrastructure damage, economic loss, and disrupted livelihoods. This study explores the role of Green Infrastructure (GI) in enhancing urban flood resilience in Peshawar, using Khamosh Colony as a case study which faces frequent flooding due to poor drainage, and encroachments. Climate projections (2015–2100) were analyzed for two Shared Socioeconomic Pathways (SSPs) using the best-performing Global Climate Models (GCMs), revealing a significant increase of 10–40% in precipitation for smaller return periods (2–10 years) and up to 61.66% and 81.35% increases for 200-year events under SSP2-4.5 and SSP5-8.5, respectively, consistent with severe flooding observed in 2022. Using SWMM, standalone and combined GI strategies were assessed for their flood mitigation potential under 2- and 5-year return period rainfall events, including future climate scenarios. Economic feasibility was evaluated using 2024 market rates while community preferences were assessed through surveys conducted in Peshawar via Kobo Toolbox. Among the options, permeable pavement (standalone) and permeable pavement with bioretention (combined) emerged as the most effective and preferred solutions. A novel aspect of this study is the integration of 1D hydrodynamic modeling with 3D visualization in SWMM and GIS, providing stakeholders with an intuitive understanding of pre- and post-GI flood scenarios. This enhances public awareness, facilitates participatory decision-making, and supports investment prioritization. By addressing the lack of integrated flood modeling and decision-support tools in Pakistan, this research fills a key gap and offers a replicable framework for evidence-based, climate-adaptive, and community-inclusive urban planning to build flood-resilient cities.

1 Introduction

Rapid global population growth continues to drive large-scale urbanization. According to the United Nations Department of Economic and Social Affairs, by 2050, an additional 2.6 billion people are expected to live in urban areas (UNDESA 2020). This expansion places immense pressure on urban infrastructure and water resources. As natural landscapes are replaced by impervious surfaces, surface runoff from precipitation events increases, groundwater recharge declines, and water quality deteriorates (Rahman et al. 2019). Moreover, abrupt land cover changes disturb natural hydrological processes, increasing the likelihood and severity of urban flooding (Akhter and Hewa 2016).

Urban flooding is among the most destructive natural hazards affecting millions worldwide and causing enormous damage to infrastructure and livelihoods (Khan et al. 2020; Loc et al. 2020, 2023). Between 1980 and 2008, approximately 100 urban flood events occurred annually, claiming an average of 6,700 lives each year (UNISDR 2015). Climate change intensifies this problem by increasing the variability and intensity of rainfall (Pindyck 2013; Miller and Hutchins 2017). Driven by greenhouse gas emissions, changing climate patterns disrupt long-established weather norms affecting ecosystems, agriculture, water resources, and urban environments globally (Farauta et al. 2012; Obubu et al. 2021).

To assess these changes researchers increasingly rely on Global Climate Models (GCMs), often downscaled with Regional Climate Models (RCMs) to simulate climate trends with greater spatial resolution (Karamouz et al. 2011; Teutschbein and Seibert 2012; Prudhomme et al. 2010). Conventional stormwater systems often designed based on historical climate data are ill-equipped to handle the increasing peak flows driven by the expansion of impervious surfaces and climate change (Hu and Zhang 2020; Padulano et al. 2021; Xiong et al. 2019). In many developing countries, outdated drainage infrastructure and a lack of climate-resilient planning exacerbate flood risks (Bibi and Kara 2023). To address these challenges, cities around the world are adopting Green Infrastructure (GI), a nature-based approach that restores hydrological function while delivering ecological and social benefits (IUCN 2016). These strategies have been widely implemented across the globe particularly in developed regions under various names and conceptual frameworks. In North America, they are known as Low Impact Developments (LIDs) and Best Management Practices (BMPs) (Fletcher et al. 2015); in the United Kingdom, they are referred to as Sustainable Urban Drainage Systems (SUDs) (CIRIA 2015); Australia promotes them under the term Water Sensitive Urban Design (WSUD) (Water by Design 2014); in Europe, they are often described as Blue-Green Infrastructure (Wilbers et al. 2022); Singapore implements similar practices through its Active, Beautiful, Clean Waters (ABC Waters) Programme (PUB 2018); and in China, the concept is advanced under the Sponge City initiative (Song 2022).

GI includes practices such as green roofs, rain gardens, permeable pavements, vegetative swales, and bioretention systems, which mimic natural processes to manage stormwater at its source (Dagenais et al. 2018; Si et al. 2022; Vijayaraghavan et al. 2021). These solutions reduce surface runoff, improve water quality, and lower flood risks while offering cost savings and co-benefits (Arjenaki et al. 2021; Kim et al. 2022; Latifi et al. 2023; Suresh et al. 2023; Ahmed et al. 2024). However, successful GI implementation requires evaluating not only its technical effectiveness but also its economic feasibility and social acceptability.

Hydrological and hydraulic modeling is critical to understanding urban flood dynamics. The U.S. EPA’s Storm Water Management Model (SWMM) is widely used for simulating stormwater flow and drainage system performance (Rossman and Huber 2017). Although SWMM5 improves user accessibility, it remains highly technical for general use (Elliott and Trowsdale 2007). Numerous studies have used SWMM to evaluate stormwater systems and GI interventions (Qin et al. 2013; Saadatpour et al. 2020; Wang et al. 2021). When integrated with two-dimensional surface modeling as seen in Personal Computer Storm Water Management Model (PCSWMM) it can visualize complex urban flooding scenarios and support decision-making (Hsu et al. 2002; Masseroni and Cislaghi 2016).

In Pakistan’s arid and semi-arid regions rainfall is highly variable with nearly 70% of stations showing increasing monsoon trends (Chaudhari 1994). Several recent extreme weather events such as the catastrophic 2020 floods in Karachi that caused nearly $1 billion in damages and 20 deaths (Ali 2021) and the 2022 floods in Lahore that halted city functions for days (Gabol 2023) highlight the growing vulnerability of Pakistan’s urban centers. Peshawar, one of Pakistan’s oldest and fastest-growing cities, is highly vulnerable to urban flooding. The city frequently experiences intense monsoon rainfall that, when combined with rapid urbanization and limited drainage capacity, leads to frequent inundation of homes, roads, and businesses. Despite the proven effectiveness of GI in various international contexts, its applicability, efficiency, and community acceptance have not been systematically studied in Peshawar. There is a critical need for a comprehensive, context-specific evaluation that incorporates hydrological modeling, cost analysis, and local perspectives to support climate-resilient urban development.

This study addresses that gap by assessing multiple GI options in the urban catchment of Khamosh Colony, one of Peshawar’s most flood-prone neighborhoods due to deteriorating drainage infrastructure, and encroachment of drains as a result of urbanization. Through the integration of high-resolution climate projections, SWMM modeling, market-based cost assessments, and household-level surveys, this study aims to evaluate the performance of GI across technical, economic, and social dimensions. Hence, the specific objectives of the study are to:

1. Analyze the impacts of projected climate change on urban flooding using high-resolution downscaled GCMs;
2. Evaluate the effectiveness and contextual suitability of selected GI interventions in mitigating urban flood risks under both historical and future climate scenarios; and
3. Develop an integrated, evidence-based decision-support framework that informs policy and planning by combining hydrological modeling, economic assessment, and stakeholder input for resilient urban flood management.

While this study is rooted in the local context of Peshawar, its methodological framework offers broader relevance to other rapidly urbanizing and flood-prone cities across South Asia and beyond. By integrating climate projections, hydraulic modeling, and community engagement, the findings contribute to the global discourse on sustainable urban water management and the mainstreaming of GI for climate adaptation.

2 Materials and Methods

The study was conducted following the methodology illustrated in the subsequent flow chart (Figure 1). Data obtained from Objective I, specifically the projected increase in flow volume under climate change scenarios were used as input for the PCSWMM model. This facilitated the assessment of the performance of various GI strategies in reducing flood nodes and peaks. Each GI option was then evaluated across the three previously discussed dimensions, technical effectiveness, economic feasibility, and social acceptability. Based on these criteria, scores were assigned, and a ranking system was applied to identify the most effective GI interventions. Finally, an evidence-based decision-making model was developed to support policymakers and key stakeholders in formulating and implementing strategies for building flood-resilient cities.

Figure 1 Methodology flow chart.

2.1 Study area

Peshawar City was selected as the study area for this research due to its rapid urban expansion and high vulnerability to climate-related hazards particularly urban flooding. As of the 2017 census, the city had a population of 4,267,198, with a population density of approximately 3,394.75 people per km² and an average annual growth rate of 3.99% from 1997 to 2017. Peshawar is a rapidly developing urban center that faces recurring and severe flooding events. According to the Water and Sanitation Services Peshawar (WSSP), 27 flooding hotspots have been identified across the city, which experience significant inundations following major rainfall events.

The total area of District Peshawar is approximately 1,215 km², of which 176.9 km² or 14.6% constitutes the study area (Figure 2). This area includes the Water and Sanitation Services Peshawar’s (WSSP’s) five designated operational zones: Zone A, B, C, D, and E, encompassing 51 Union Councils that define the urban extent of Peshawar (WSSP 2025). In addition, planned residential areas such as Hayatabad, Regi Lalma Township, and the Defense Housing Authority (DHA) have been included in the study boundary.

This research focuses on the Khamosh Colony catchment, one of the 27 identified flooding hotspots located in the city’s core urban zone, which is projected to experience significant growth in built-up land between 2025 and 2050. This hotspot was selected based on its flood vulnerability, rapid land-use changes, and critical urban infrastructure exposure.

2.2 Data and sources

The data used in this study were collected from a variety of reliable sources to ensure accuracy in the modeling and analysis of urban flooding in Peshawar (Table 1). Precipitation data from 1961 to 2024 were obtained from the Pakistan Meteorological Department (PMD), providing a robust historical baseline for trend analysis and model calibration. To assess future climate impacts, Global Climate Models (GCMs) from the CMIP6 dataset were used, covering the period from 1980 to 2100. This allowed the study to evaluate the implications of climate change under different Shared Socioeconomic Pathways (SSPs).

Table 1 Data and sources.

Sr No	Data	Period	Source
1	Precipitation	1961–2024	Pakistan Meteorological Department (PMD)
2	GCMs	1980–2100	CMIP6
3	Digital Elevation Model (DEM)	2020	Project Management Unit (PMU) of local government KPK
4	Buildings	2023	Google Open Buildings
5	Drainage	2024	Field survey
6	GI economic aspects	2024	GKP 2024, literature

Topographical and spatial analysis was supported by a 1-m resolution Digital Elevation Model (DEM) from 2020, sourced from the Project Management Unit (PMU) of the Local Government, Khyber Pakhtunkhwa (KPK), which was crucial for delineating flow paths and sub-catchments. Building footprint data for 2023 were retrieved from Google Open Buildings, aiding in accurate land use representation and 3D visualization. Information on the existing drainage network was gathered through a field survey conducted in 2024 to ensure up-to-date and site-specific details for hydraulic and hydrological modeling. Finally, economic data for evaluating GI options were compiled from both Government of Khyber Pakhtunkhwa (GKP) 2024 and relevant literature, enabling a well-rounded assessment of cost-effectiveness and feasibility.

Figure 2 shows the study area map showing the WSSP jurisdictional boundary within Peshawar, delineated into five operational zones (A–E). Red markers indicate the eight identified flood hotspot locations used in the analysis.

Figure 2 Study area.

2.3 Climate change

GCMs and Bias correction

Based on an extensive literature review, nine Global Climate Models (GCMs) from the CMIP6 platform namely BCC-CSM2-MR, ACCESS-ESM1-5, MIROC6, CNRM-CM6-1, MRI-ESM2-0, CNRM-ESM2-1, EC-EARTH-VEG, IPSL-CM6A-LR, and MPI-ESM1-2-HR were selected for precipitation projection. GCM outputs often exhibit systematic biases due to inherent simplifications in the modeling of complex atmospheric processes, which can lead to discrepancies between simulated and observed data (Hemanandhini 2023).

To improve the reliability of the climate projections and align them more closely with observed historical data, a bias correction process was applied. This study employed the quantile mapping technique, which adjusts the statistical distribution of modelled data by aligning the cumulative distribution function (CDF) of the GCM outputs with that of observed data (Humphries et al. 2024). This approach effectively corrects biases across the distribution, particularly for extreme values.

The quantile mapping method was implemented using the “qmap” package in the R programming language (Gudmundsson 2016; Venables et al. 2012), which provides robust tools for bias correction of climate variables. The mathematical formulation of the quantile mapping technique used in this study is presented in Equations 1–2, below.

(1)

(2)

Where:

P	=	precipitation,
d	=	daily,
m	=	monthly,
*	=	bias corrected,
his	=	raw GCM data,
obs	=	observed data,
sim	=	raw GCM future data,
F	=	Cumulative Distribution Function (CDF), and
F−1	=	inverse of CDF.

GCM Ranking

To effectively analyze climate trends, it is essential to identify the best-performing GCMs from the selected set. This is accomplished by comparing the statistical performance of each model against observed historical data using evaluation metrics such as Root Mean Square Error (RMSE), Percentage Bias (PBIAS), correlation coefficient (r), and other relevant indicators (Jose and Dwarakish 2022). In this study, RMSE, PBIAS, and Standard Deviation have been selected as the primary metrics for ranking the GCMs. These indicators are particularly effective in evaluating model accuracy, quantifying errors, and assessing variability between simulated and observed data. The mathematical formulations of these metrics used in the study are provided in Equations 3–5 below.

(3)

(4)

(5)

Where:

Xobserved and Xmodel	=	observed and model values of precipitation,
N	=	number of data points, and
X̅	=	mean value of observed data.

The model with the smallest RMSE value, percent bias closer to zero, and the standard deviation closer to the observed data’s standard deviation gets the first rank among all.

Ensemble and climate change scenarios

To address the inherent uncertainties in climate projections generated by GCMs, this study employed an ensemble approach by integrating multiple models with differing structures, assumptions, and parameterizations. This method improves the reliability of future climate forecasts by averaging the outputs of multiple GCMs, thereby reducing the influence of individual model biases and minimizing the effects of outliers. Based on a ranking process using performance metrics, the top five models BCC-CSM2-MR, CNRM-CM6-1, MIROC6, ACCESS-ESM1-5, and MRI-ESM2-0 were selected for further analysis. The ensemble derived from these models was used to project future precipitation trends in the study area, focusing on variations expected through the year 2100.

For temporal analysis, the projection period from 2015 to 2100 was divided into three standard timeframes, 2015–2040 (near future), 2041–2070 (mid future), and 2071–2100 (far future). This segmentation, commonly adopted in climate impact assessments, allows for a structured evaluation of temporal trends. The near future reflects short-term responses to current emissions, the mid-future captures the increasing influence of socioeconomic and emission pathways, and the far future serves as a critical window for examining long-term, high-risk climate scenarios (Enyew et al. 2024).

Climate projections were analyzed under two Shared Socioeconomic Pathways (SSPs) namely SSP2-4.5 and SSP5-8.5. SSP2-4.5 represents a "middle-of-the-road" scenario, characterized by balanced challenges to both mitigation and adaptation. In contrast, SSP5-8.5 depicts a high-emission pathway associated with limited mitigation efforts, high fossil fuel dependency, and low adaptive capacity (Hausfather 2018). Among the selected models, BCC-CSM2-MR identified as the top-performing GCM was specifically used to investigate the projected changes in rainfall intensity under both scenarios.

2.4 SWMM model

The Storm Water Management Model (SWMM) is a widely used and flexible simulation tool for modeling rainfall-runoff processes, applicable to both single-event and long-term continuous storm simulations. Designed primarily for urban hydrology, SWMM simulates runoff quantity and quality by analyzing precipitation over defined sub-catchments, which generate surface runoff and pollutant loads. The model then routes this runoff through a drainage network consisting of pipes, channels, storage units, pumps, treatment facilities, and flow regulators. Throughout the simulation period, SWMM tracks dynamic changes in flow rates, water depths, and pollutant concentrations in each sub-catchment.

At the core of SWMM is an unsteady flow solver based on the EXTRAN algorithm developed by Roesner et al. (1988), which remains central to SWMM 5.1 due to its simplicity and robustness, despite ongoing enhancements for improved numerical stability (Rossman and Huber 2017). The model employs a link-node framework and solves the full Saint-Venant equations to simulate unsteady, free-surface flow in conduits and channels (Roesner et al. 1988; Rossman 2006). These equations, mass and momentum conservation form the foundation of SWMM’s hydraulic computations, enabling detailed and accurate simulation of stormwater behavior under a wide range of hydrological conditions (Equations 6 and 7).

(6)

(7)

Where:

A	=	cross-sectional area,
t	=	time,
Q	=	flow rate,
x	=	distance,
H	=	hydraulic head of water in the conduit,
g	=	gravity, and
Sf	=	friction slope.

This was implemented using Manning’s equation (Rossman 2006).

In this study, the PCSWMM was employed to simulate urban flooding in Khamosh Colony and to evaluate the technical suitability of various GI interventions. PCSWMM is specifically designed for modeling urban runoff during intense rainfall events and serves as a comprehensive tool for planning, analyzing, and designing drainage networks in metropolitan areas (Surwase and Manjusree 2019).

Model development

To simulate the rainfall-runoff processes, the hydrological model in PCSWMM was developed for Khamosh Colony catchment using a structured and systematic approach. The modeling process began with the delineation of watershed boundaries and sub-catchments, using spatial datasets derived from GIS, including the Digital Elevation Model (DEM), land use land cover, and soil characteristics. The stormwater drainage network was then constructed in the form of conduits and junctions, with designated outfalls. The physical dimensions of these conduits were incorporated based on actual field data collected during site visits.

PCSWMM utilizes rainfall data to perform hydrological simulations and evaluate system responses. In line with the study’s objective of conducting a climate-inclusive assessment of GI, both historical and projected precipitation data were used. Projected rainfall data were sourced from selected GCMs to capture future climate variability. This enabled a comparative evaluation of GI performance across different rainfall return periods under both current and future climate scenarios, supporting a more comprehensive assessment of their effectiveness in mitigating flood risks under changing climatic conditions.

Model calibration and validation

Model calibration was performed using multiple input parameters derived from available spatial and hydrological data. These included the DEM for calculating catchment slopes, intensity-duration-frequency (IDF) curves developed from precipitation data provided by the meteorological department, drainage network dimensions obtained through field surveys, and imperviousness percentages for each sub-catchment, extracted from GIS-based land cover maps.

Given the absence of detailed historical flow data, the model was calibrated using the design assumption that drains were constructed for a 2-year return period event, in line with the standards of the Water and Sanitation Agency (WASA) Lahore, which prescribes a 2-year return period for secondary and tertiary drains and a 5-year return period for primary drains. (WASA 2025) This assumption-based approach is commonly used in urban flood modeling where observed calibration datasets are limited or unavailable (Leandro et al. 2011).

The calibration process was iterative. Initial model runs for the 2-year return period rainfall event resulted in flooding at several junctions within the catchment. Input parameters were then systematically adjusted including those related to subcatchments until simulations showed no flooding. This indicated successful calibration enabling the model to be used for assessing the effectiveness of GI in reducing runoff volume and the number of flooding junctions. A similar approach was followed by Rosenberger et al. (2021), who calibrated their urban stormwater model using input subcatchments parameters applying a 5-year return period storm for calibration.

2.5 Green infrastructure

To evaluate the impact of GI on runoff within the study area, multiple implementation scenarios were developed and integrated into the PCSWMM model. These scenarios included both standalone applications of individual GI types and combined configurations tailored to the site-specific conditions of the selected catchments. The effectiveness of each GI scenario was assessed based on three key dimensions, technical performance in reducing runoff and flooding, economic feasibility through cost analysis, and social acceptability determined via community feedback. This comprehensive evaluation approach ensures a contextually relevant and practically implementable GI strategy for enhancing urban flood resilience.

Economic assessment

A comprehensive economic assessment of various GI options was carried out, encompassing both capital expenditure (CAPEX) related to construction and operational expenditures (OPEX) associated with maintenance. Typical cross-sections and design specifications for each GI type were adapted from relevant literature and previous research studies. Capital costs were estimated using the latest Market Rate System (GKP 2024), while maintenance costs were derived as a percentage of the initial construction costs, based on values reported in existing studies.

These cost estimation scenarios serve multiple purposes: they support the selection of appropriate GI types, inform the evaluation of specific design features, provide insight into how costs vary with site complexity or constraints, and offer preliminary estimates of annual operation and maintenance costs for each GI alternative (Bernagros et al. 2021). The detailed cost estimates for the different GI interventions considered in this study are presented in Table 2, below.

Table 2 GI cost estimates.

S No	GI	Level of Intervention	CAPEX (PKR)	OPEX (PKR)	Units
1	Green Roof	House Level	7,012	1,207	per m2
2	Rain Barrel	House Level	556	0	per m2
3	Permeable Pavement	Street Level	4,481	224	per m2
4	Infiltration Trench	Street Level	11,513	2,303	per m2
5	Bioretention System	Street Level	18,555	742	per m2
6	Rain Garden	Street Level	12,078	725	per m2
7	Vegetative Swale	Street Level	1,622	180	per m2

Technical assessment and scenario design

The technical suitability of GI options was assessed using the PCSWMM model by evaluating their impact on runoff volume and flooding nodes reduction under various implementation scenarios. These included both standalone and combined applications of GI, with a particular focus on their effectiveness in mitigating flood volumes.

A baseline scenario representing the current condition without any GI intervention was established to reflect the existing urban drainage infrastructure. This “No GI” scenario served as a reference for comparison. Subsequently, individual GI types were simulated separately, including Green Roof (GR), Rain Garden (RG), Bioretention (BR), Rain Barrel (RB), Permeable Pavement (PP), Infiltration Trench (IT), and Vegetative Swale (VS). In addition, several paired and multi-GI combinations were evaluated such as GR+BR, GR+RB, RB+BR, RG+BR, PP+BR, and PP+IT. The final scenario included the integration of all six GI types across the catchment.

Each scenario was applied under both historical and future precipitation conditions, considering 2- and 5-year return period storm events, to comprehensively assess the performance of GI interventions under changing climatic conditions (Table 3). Model outputs including the number of flooding nodes and total flood volume were analyzed to identify the most technically effective GI configuration for urban flood mitigation in the study area.

Table 3 Standalone and combined GI scenarios.

Application	GI Applied	Description	Number of units	Total area (m2)
Standalone	GR	50% houses in each sub catchment with 50% of roof covered with vegetation	2,960	18,451
	BR	5% impervious area in each catchment	664	56,089
	RG	10% pervious area in each catchment	332	49,736
	PP	50% roads and streets	559	75,251
	IT	Along 25% roads and streets	555	34,662
	RB	100% houses	5,907	-
	VS	5% in each sub-catchment	559	2,639
Combination	GR & RB	50% green roof + 100% rain barrel
	GR & BR	50% green roof + 5% bioretention
	RB & BR	100% rain barrel + 5% bioretention
	BR & PP	5% bioretention + 50% porous pavement
	IT & PP	–
	RG + BR	–
	All 6 GI	–

For combined GI scenarios, the number of units and total area are not reported separately, as these scenarios are generated through aggregation of the standalone measures. No additional land or unit requirements are introduced beyond those already specified under the standalone GI options.

Social assessment

The social acceptability of GI and its perceived aesthetic and functional value within the community is a critical component of this study. To assess public perception, structured surveys, interviews, and group discussions were conducted using Kobo Toolbox (Harvard Humanitarian Initiative n.d.), an open-source platform for data collection and analysis. A detailed questionnaire was designed and administered to residents within the study area to capture insights from the primary stakeholders, which is the local community.

The survey covered various aspects, including the frequency and extent of flooding in the respondents’ neighborhoods, the duration of flood-related issues, perceived causes of urban flooding, types of damage experienced, and the community’s willingness to support GI implementation and maintenance—either independently or through government initiatives. Additionally, the questionnaire sought to determine preferences for mitigation measures at the household level versus the street or neighborhood level.

To determine an appropriate sample size, a 95% confidence level and a 5% margin of error were applied, resulting in a minimum required sample size of 385 for a population of 2 million. A total of 443 valid responses were collected, exceeding the required sample size and offering robust data for analysis.

Based on the survey results, the level of community acceptance and potential resistance to different GI options were evaluated. These insights were then incorporated into the final ranking of GI alternatives, with higher preference given to those receiving greater support and acceptability among residents. This approach ensures that the selected GI strategies are not only technically and economically viable but also socially sustainable and aligned with community priorities.

Figure 3 illustrates the geographic distribution of survey locations within the study area where the social assessment was conducted. The survey covered a broad spatial extent, including all five Water and Sanitation Services Company (WSSC) operational zones (A, B, C, D, and E), as well as the residential areas of Regi, DHA, and Hayatabad Township. Individual survey responses are represented as black dots on the map, highlighting the comprehensive coverage and spatial representation of public input across the entire study area.

Figure 3 Kobo survey locations.

2.6 Developing an informed decision-making model for policy makers

To develop a comprehensive PCSWMM model integrated with the most effective GI solutions and to visualize the urban flooding scenario in three dimensions, ArcGIS Pro was used in conjunction with PCSWMM. The 3D visualization incorporated several datasets, including a 1-m resolution DEM, building footprints sourced from open-access building data, flood simulation results from PCSWMM, and a base map for contextual reference.

The DEM was processed in ArcGIS Pro to extract topographic features and delineate flow directions within the catchment. Building footprints were converted into 3D models by extruding polygon geometries based on available height attributes, effectively creating a realistic urban landscape. Flooding data from PCSWMM, particularly the extent and intensity of flooding at individual nodes, was integrated into the ArcGIS environment. The radius of influence for each flooded junction was used to map the spatial extent of inundation across the catchment.

Following the development of the base 3D model, the best-performing GI interventions, both as standalone and combined configurations, were incorporated into the PCSWMM model. These GI-integrated scenarios were then visualized alongside the 3D building models to provide a more intuitive and spatially rich representation of the urban flooding situation.

Simulations were conducted for storm events corresponding to 2-year (under climate-inclusive conditions) and 5-year (under both historical and climate-inclusive conditions) return periods. The results of these simulations are presented in the following section, comparing flood extent and severity in scenarios with and without GI implementation, thereby illustrating the effectiveness of GI in reducing urban flood risks within the catchment.

3 Results

3.1 Future climate projections

The variation in annual precipitation from 2015 to 2100 under both moderate and extreme climate scenarios, SSP2-4.5 and SSP5-8.5 respectively, was projected using the top five climate models individually, as well as through an ensemble of these models. The results, illustrated in Figure 4, highlight a pronounced flooding scenario during the years 2021–2022 under the SSP5-8.5 scenario simulated by the MIROC6 model. This aligns well with observed events, as Pakistan experienced severe flooding during this period. Among the evaluated models, BCC-CSM2-MR emerged as the top performer and was subsequently used to project 24-hour precipitation intensities in the study area for return periods of 2, 5, 10, 25, 50, 100, and 200 years. These projections were compared with estimates derived from observed data spanning 1961–2024.

Figure 5 presents both observed and projected precipitation values under the two SSP scenarios. Higher return periods showed much sharper rises, with 100-year and 200-year events increasing by over 55% and 62% under SSP2-4.5 and by nearly 60% and 69% under SSP5-8.5, whereas lower return periods (2–5 years) showed moderate rise of 11–27%.

Figure 4 Future climate projections under SSP2-4.5 and SSP5-8.5.

Figure 5 Observed vs. projected precipitation under SSP scenarios.

While the percentage increase in daily precipitation for return periods of 2 to 100 years was relatively modest between SSP2-4.5 and SSP5-8.5, the difference became more pronounced for the 200-year return period with around 62% rise under SSP2-4.5 and 81% increase under SSP5-8.5, indicating a potential intensification of extreme precipitation events under the more severe climate scenario.

3.2 SWMM model

The model developed in PCSWMM is illustrated in Figure 6, which represents the existing drainage network within the sub-catchment of Khamosh Colony. Variations in the width and depth of the drainage system are depicted through color-coded drain segments, corresponding to different geometric dimensions as indicated in the legend. Blue markers identify the calibrated model outfalls. Based on the scenarios outlined in the methodology section, each GI intervention is incorporated into the model for the catchment. The effectiveness of these GI measures in reducing flood volumes is then evaluated. For the technical assessment, the number of flooding junctions associated with each GI scenario are simulated and analyzed using the PCSWMM model to determine their potential in mitigating urban flooding under current and projected conditions.

Figure 6 Calibrated SWMM model.

3.3 Technical assessment of green infrastructure

The results of the peak flow reduction are presented in the form of graphs as shown in Figures 7–9. For the climate-inclusive 2-year return period event (Figure 7), the findings highlight the strong potential of GI to mitigate flooding. Standalone GI measures significantly reduce peak flow from 0.1634 m³/s (baseline scenario), with green roofs and permeable pavements eliminating runoff completely, and bioretention systems, infiltration trench and rain gardens achieving 38–49% reductions. In contrast, rain barrels and vegetative swales provide smaller reductions of about 12–13%. Combined GI measures prove most effective, with nearly all combinations reducing flows to zero, except for the RB + BR combination, which showed a residual peak flow of 0.0941 m³/s, confirming that standalone measures are particularly effective for smaller, more frequent storms.

Figure 7 Climate-inclusive 2-year return period at node J887: (a) Peak flow reduction by standalone GI, and (b) Peak flow reduction by combined GI.

Figure 8 Historic 5-year return period at node J459: (a) Peak flow reduction by standalone GI, and (b) Peak flow reduction by combined GI.

For the historic 5-year return period (Figure 8), standalone GI measures with bioretention, rain gardens, permeable pavements, and green roofs reduced peak flows from 0.1303 m³/s to zero, eliminating runoff completely, while infiltration trench lowered it to 0.1 m³/s (≈23% reduction). Rain barrels and vegetative swales showed minimal effect, leaving flows close to the baseline. Combined GI measures performed even better, with most pairings and the “all GI” scenario reducing peak flow to zero, except GR+RB which yields a modest reduction to 0.1183 m³/s (Figure 9).

Figure 9 Climate-inclusive 5-year return period at node J459: (a) Peak flow reduction by standalone GI, and (b) Peak flow reduction by combined GI.

In contrast, the climate-inclusive 5-year return period scenario revealed considerably higher peak flows, driven by intensified rainfall associated with climate change. In this case, standalone GI measures were generally insufficient to fully mitigate flooding. The baseline peak flow of 0.2017 m³/s is reduced most effectively by standalone rain gardens, which lower flows to 0.1299 m³/s (≈36% reduction). Other measures, such as vegetative swales, bioretention, green roofs, and infiltration trenches, achieve modest reductions of 7–14%, while rain barrels have almost no effect. Combined GI measures deliver greater benefits, with RG+BR reducing flows to 0.1 m³/s (≈50% reduction) and the “all GI” configuration completely eliminating runoff, emphasizing the critical role of integrated GI strategies in adapting to future climate extremes.

3.4 Social and economic assessment of green infrastructure

The results of social assessment of green infrastructure show that streel-level GI interventions like permeable pavement, bioretention systems, rain gardens and vegetative swales are preferred over house-level interventions of rain barrels and green roofs, showing resistance to alter private properties. However, based on the economic assessment of the same GI, rain barrels emerge as the most cost-effective interventions with the lowest CAPEX and OPEX, followed by vegetative swales. These results highlight the significance of policies that align affordability with community acceptance. The ranking of different GI scenarios under consideration based on social, economic and technical assessment is shown in Table 4 and Table 5 for standalone and combined categories respectively.

3.5 Performance of green infrastructure

The performance of Green Infrastructure (GI), both as standalone measures and in combination, varies across the different aspects of the assessment. The evaluation is based on a multi-criteria performance ranking system that considers economic, technical, and social factors. In the results presented in the following table, a lower overall score indicates a better-performing GI option across these three dimensions. Among the standalone applications, permeable pavement emerges as the most effective solution due to its cost-effectiveness, social acceptability, ease of implementation, and balanced performance across technical, social, and economic criteria. In terms of combined applications, scenarios involving bioretention with permeable pavement and permeable pavement with infiltration trench demonstrate the best overall performance.

These combinations outperform others by offering a well-rounded solution that addresses all three key evaluation aspects. Tables 4 and 5 present the performance and ranking of standalone and combined GI options, respectively.

Table 4 Standalone GI ranking.

S No	GI	Score			Total Score	Ranking
		Economic Aspect	Technical Aspect	Social Aspect
1	Permeable Pavement	3	1	1	5	1
2	Rain Garden	5	1	1	7	2
3	Green Roof	4	2	2	8	3
4	Rain Barrel	1	5	2	8	3
5	Vegetative Swale	2	6	1	9	4
6	Infiltration Trench	6	4	1	11	5
7	Bioretention System	7	3	1	11	5

Table 5 Combined GI ranking.

S No	GI	Score			Total Score	Ranking
		Economic Aspect	Technical Aspect	Social Aspect
1	Bioretention 5% + Permeable Pavement 50%	4	2	1	7	1
2	Infiltration Trench + Permeable Pavement	2	4	1	7	1
3	50% Green Roof + 5% Bioretention	5	2	2	9	2
4	Green Roof + Rain Barrel	1	6	3	10	3
5	Rain Barrel + 5% Bioretention	3	5	2	10	3
6	Rain Garden + Bioretention	6	3	1	10	3
7	All 6 GI	7	1	2	10	3

3.5 Three-dimensional representation of urban flooding

This study integrates the PCSWMM model with 3D visualization to enhance the understanding of flooding scenarios for policymakers and key stakeholders. The most effective GI options, PP and the combination of BR and PP were identified through a multi-criteria ranking process and applied to the case study area of Khamosh Colony. The resulting 3D maps provide a more intuitive and realistic depiction of buildings and flood inundation, aiding in clearer communication of flood risks.

Under the 2-year return period with climate change considerations, the performance of GI measures was observed to be highly effective (Figure 10). For the 5-year return period under historical conditions (Figure 11), GI performance remained optimal, significantly reducing flood impacts. However, under the 5-year return period with climate change effects (Figure 12), the effectiveness of standalone or partial GI applications diminished, highlighting the sensitivity of GI performance to increased rainfall intensity and volume associated with climate change.

Figure 10 Flooding scenario under climate inclusive 2-year return period event.

Figure 11 Flooding scenario under historic 5-year return period event.

Figure 12 Flooding scenario under climate inclusive 5-year return period.

Complete elimination of flooding, even under the 5-year climate change scenario, was achieved only when all six GI types were implemented simultaneously reducing flood nodes to zero across both return periods and climate conditions. Nevertheless, such a comprehensive application is often unfeasible in dense urban areas like Peshawar, where space constraints limit the large-scale implementation of GI measures.

Therefore, it can be concluded that while GI plays a valuable role in mitigating flood impacts, especially under moderate conditions, it may not be sufficient on its own during high-intensity rainfall events driven by climate change. In such cases, GI should be considered a complementary system to traditional grey infrastructure, rather than a standalone solution, particularly in space-constrained urban environments.

4 Discussion

4.1 Comparative analysis

The conclusions of this study highlight the effectiveness of GI in mitigating urban flooding in Peshawar. While GI solutions such as permeable pavement (PP), bioretention system (BR), and infiltration system (IT) have been widely recognized for their ability to manage stormwater and reduce urban flood risks globally, this study reaffirms that their performance is highly influenced by local factors including climate conditions, urban density, and hydrological characteristics. In this context, PP emerged as the most effective standalone GI option, offering a balanced performance in terms of economic feasibility, technical efficiency, and social acceptance. These findings are consistent with previous research conducted in other flood-prone urban areas, such as Beijing, China (Hu et al. 2018), and Shahrekord City, Iran (Arjenaki et al. 2021), where PP significantly reduced surface runoff and enhanced infiltration. Their performance improves further when combined with bioretention systems, as they provide additional storage and treatment capacity. This dual mechanism enhances infiltration and delays stormwater release, explaining the superior effectiveness of the combined approach.

Moreover, GI combinations, particularly IT with PP and BR with PP outperformed standalone measures in reducing flood volumes. Under historic 5-year return periods, standalone GI (e.g., permeable pavement and bioretention systems) reduced peak flows effectively, but their efficiency decreased under climate-inclusive scenarios. In contrast, 2-year climate-inclusive events showed the best performance of GI, while for 5-year events under climate change, only combined GI applications were able to eliminate flooding nodes. However, applying all GI together is often impractical in congested urban settings like Peshawar, suggesting that GI can complement but not fully replace grey infrastructure for high-intensity precipitation events. This reinforces findings from prior studies (Kim and Kang 2023), emphasizing that GI should be integrated with conventional grey infrastructure to develop a robust and comprehensive urban flood risk management strategy.

The integration of 3D visualization proved critical for bridging the gap between technical modeling and stakeholder understanding. 3D visualizations, as opposed to 1D outputs, offered an understandable spatial representation of the depth of inundation, flow paths, and infrastructure impacts. Research indicates that 3D visualization improves communication with non-technical audiences, stimulates creativity, and strengthens decision-making (Kilsedar et al. 2019; Wang et al. 2019; Herman and Stachoň 2016). Policymakers can test GI scenarios, identify high-risk areas, and better engage the community using such models. Although there are possible negative effects like public anxiety, and our institutional mistrust must be controlled through careful framing and communication, this tool can increase awareness of flood risks and promote preventive measures.

4.2 Practical implications for urban planning and policy

This study has important implications for urban planning and policymaking in Peshawar, as well as in other rapidly urbanizing cities across Pakistan and across the globe. The findings underscore the need to integrate GI into existing urban development plans to enhance flood resilience, particularly in high-risk areas identified as flooding hotspots by the WSSP.

The results highlight the urgency of revising building codes to mandate the inclusion of GI in new developments, for example, the use of PP for roads, parking lots and streets, and GR, particularly on commercial buildings. Peshawar’s current urban drainage systems are based on historical rainfall patterns and do not adequately consider the increased intensity and frequency of rainfall events resulting from climate change. This study recommends the integration of SWMM based flood modeling into infrastructure planning to support climate-resilient urban development.

Given the spatial constraints in dense urban areas like Peshawar, large-scale GI implementation may not be practical. Instead, targeted retrofitting of existing infrastructure such as roads, sidewalks, and public spaces with BR and PP should be prioritized, along with the incorporation of GI into all new construction projects.

To encourage broader adoption, policies should introduce financial incentives such as tax rebates or subsidies for homeowners and businesses that install GI solutions, including RB and GR. Effective implementation and funding of GI initiatives will require collaboration between public agencies, private developers, and non-governmental organizations (NGOs). Given that social acceptance plays a critical role in the success of GI strategies, it is essential to launch public awareness campaigns to educate residents about the long-term benefits of GI, such as reduced flood risk, improved environmental quality, and enhanced urban aesthetics. By adopting these strategies, Peshawar can move towards a climate-resilient and sustainable urban future, mitigating flood damage while promoting greener, healthier urban living environments.

4.3 Broader implications

Beyond Peshawar, the findings of this study hold broader relevance, contributing to global discussions on disaster risk reduction, sustainable water management, and climate-resilient urban planning. The methodology employed specifically the use of PCSWMM modeling to evaluate flood mitigation strategies can be effectively replicated in other flood-prone urban areas across South Asia, and particularly within Pakistan, where rapid urbanization and changing climate patterns present similar challenges.

Moreover, the multi-criteria ranking system used to assess GI performance based on economic, technical, and social factors offers a robust decision-making framework for urban planners and policymakers at both national and international levels. This framework enables informed selection of context-appropriate GI strategies tailored to the specific needs and constraints of urban environments.

The study’s outcomes also align with and support multiple United Nations Sustainable Development Goals (SDGs), particularly, SDG 11 which is “Sustainable Cities and Communities” by promoting urban infrastructure that enhances resilience to flooding and other climate-related hazards, and SDG 13 “Climate Action” through the integration of climate-adaptive infrastructure into urban planning and design to reduce vulnerability to extreme weather events. These contributions reinforce the role of nature-based solutions and modeling tools in building more sustainable and resilient cities in the face of climate change.

4.4 Challenges, limitations, and opportunities

The results of the social survey conducted as part of this study reveal that large-scale implementation of GI often faces resistance due to financial limitations, maintenance concerns, and constraints related to land availability. Additionally, the adoption of GI is significantly hindered by a general lack of awareness among residents regarding its long-term environmental and social benefits. This knowledge gap can be addressed through targeted awareness campaigns and by actively involving communities in the planning, design, and maintenance of GI solutions, thereby fostering a sense of ownership and long-term support.

The SWMM model calibration being based on a 2-year return period assumption introduces potential uncertainties, as it may not fully represent the variability of extreme rainfall or flooding events. This limitation could influence the accuracy of peak flow reductions estimated for different Green Infrastructure (GI) interventions. Similarly, the scope of the social survey may not have captured the full diversity of community perspectives, which restricts the broader aspects of deriving conclusions from the findings. Cost estimates were generalized rather than location-specific, and governance and institutional aspects were not fully covered. But at the same time, the study presents opportunities for future research.

Future research should explore the integration of smart stormwater management systems, such as Internet of Things (IoT)-based sensors for real-time flood detection and response. These technologies can enhance the performance of GI by enabling automated reactions to rainfall events, improving both efficiency and reliability. To effectively manage extreme precipitation events in Peshawar, a hybrid approach combining GI with conventional grey infrastructure should be considered. This integrated strategy can help overcome the limitations of GI under high-intensity storms, especially in dense urban areas with limited space for large-scale interventions. Furthermore, additional cost-benefit analyses are needed to evaluate the long-term economic returns of GI investments. These should account for a range of co-benefits, including reduced flood damages, improved air quality, enhanced urban aesthetics, and increased property values, which can strengthen the case for funding and implementation. Another promising area for future work is the promotion of community-led GI initiatives, where residents manage and maintain small-scale installations such as GR, community RG, and RB. Such grassroots involvement can improve sustainability, reduce implementation costs, and increase public acceptance. Finally, it is important to examine the impact of GI on groundwater recharge, particularly in regions like Peshawar where over-extraction and rapid population growth have led to declining groundwater levels. Understanding the hydrological benefits of GI can further support its role in integrated urban water resource management.

To enhance urban flood resilience, strict zoning and building regulations should be enforced to limit impervious growth, while GI must be integrated into planning codes as a mandatory requirement. A hybrid approach combining GI with grey infrastructure is recommended, with cost-effective measures like rain barrels, bioretention systems and permeable pavements prioritized in high-risk areas. Targeted deployment based on hazard mapping, supported by public–private partnerships, can improve efficiency and financing. Additionally, 3D visualization tools should be used to support policymaking, awareness, and community engagement.

5 Conclusion

Peshawar, one of the most flood-prone cities in Pakistan, faces growing challenges due to rapid urbanization, inadequate drainage infrastructure, and the intensifying effects of climate change. This study underscores the urgent need to incorporate climate-resilient and nature-based solutions into urban planning as a proactive measure to flood risk management. By assessing projected climate change impacts using CMIP6 GCMs under SSP2-4.5 and SSP5-8.5, this study projects substantial increases in extreme precipitation (12–62% under SSP2-4.5 and 11–81% under SSP5-8.5).

Applying hydrological modeling, the research provides critical insights into the rising risk of urban flooding and the potential of GI to mitigate these impacts. Through the application of the PCSWMM, the study evaluates a range of GI strategies including GR, BR, RG, RB, PP, VS and IT and demonstrates their effectiveness in reducing surface runoff and peak flood flows. Permeable pavements emerged as the most effective standalone option, while combined treatments (e.g., BR + PP, IT + PP) achieved the greatest mitigation. Economic and social assessments further indicate strong community support for street-level interventions like BR, PP, and IT, especially when government-funded, while household-level measures face low acceptance. GI measures were found to be particularly effective during low to moderate rainfall events, however, their standalone effectiveness diminishes under extreme precipitation scenarios driven by climate change. This finding emphasizes the necessity of integrating GI with conventional grey infrastructure for comprehensive flood risk management.

A key innovation of this study is the integration of PCSWMM modeling with 3D visualization using GIS, which enhances the communication of complex flood scenarios to policymakers, stakeholders, and the general public. This visual tool supports more informed decision-making and fosters broader public engagement. Additionally, the use of a multi-criteria ranking framework ensures that GI interventions are prioritized based on their technical performance, economic feasibility, and social acceptance, offering a practical guide for implementation in vulnerable urban areas.

The findings advocate for data-driven, climate-responsive, and community-inclusive urban planning approaches to build more flood-resilient cities. By incorporating these insights into urban development policies, climate adaptation strategies, and local capacity-building efforts, cities like Peshawar can significantly enhance their ability to manage urban flooding while promoting sustainable, livable environments for their citizens.

References

1. Ahmed, F., H.H. Loc, M.S. Babel, and J. Stamm. 2024. “A community-scale study on nature-based solutions (NBS) for stormwater management under tropical climate: The case of the Asian Institute of Technology (AIT), Thailand.” Journal of Hydroinformatics 26 (5): 1080–1099. https://doi.org/10.2166/hydro.2024.288
2. Akhter, M.S. and G.A. Hewa. 2016. “The use of PCSWMM for assessing the impacts of land use changes on hydrological responses and performance of WSUD in managing the impacts at Myponga catchment, South Australia.” Water 8 (11): 511. https://doi.org/10.3390/w8110511
3. Ali, F.Z. 2021. Karachi's Urban Flooding: Looking for Solutions. Human Rights Commission of Pakistan. https://oldsite.hrcp-web.org/wp-content/uploads/2020/12/Karachis-Urban-Flooding-_2020.pdf
4. Arjenaki, M.O., H.R.Z. Sanayei, H. Heidarzadeh, and N.A. Mahabadi. 2021. “Modeling and investigating the effect of the LID methods on collection network of urban runoff using the SWMM model (case study: Shahrekord City).” Modeling Earth Systems and Environment 7 (1): 1–16. https://doi.org/10.1007/s40808-020-00870-2
5. Bernagros, J.T., D. Pankani, S. Struck, and M. Deerhake. 2021. “Estimating regionalized planning costs of green infrastructure and low-impact development stormwater management practices: Updates to the US Environmental Protection Agency’s national stormwater calculator.” Journal of Sustainable Water in the Built Environment 7 (2): 04020021. https://doi.org/10.1061/jswbay.0000934
6. Bibi, T.S. and K.G. Kara. 2023. “Evaluation of climate change, urbanization, and low-impact development practices on urban flooding.” Heliyon 9 (1): e12955. https://doi.org/10.1016/j.heliyon.2023.e12955
7. Chaudhari, Q. 1994. “Pakistan’s summer monsoon rainfall associated with global and regional circulation features and its seasonal prediction.” Proceedings of the International Conference on Monsoon Variability and Prediction, Trieste, Italy, May 9.
8. CIRIA. 2015. The SuDS Manual (C753). CIRIA, London, UK.
9. Dagenais, D., J. Brisson, and T.D. Fletcher. 2018. “The role of plants in bioretention systems; does the science underpin current guidance?” Ecological Engineering 120, 532–545. https://doi.org/10.1016/j.ecoleng.2018.07.007
10. Elliott, A. and S. Trowsdale. 2007. “A review of models for low impact urban stormwater drainage.” Environmental Modelling & Software 22 (3): 394–405. https://doi.org/10.1016/j.envsoft.2005.12.005
11. Enyew, F.B., D. Sahlu, G.B. Tarekegn, S. Hama, and S.E. Debele. 2024. “Performance Evaluation of CMIP6 Climate Model Projections for Precipitation and Temperature in the Upper Blue Nile Basin, Ethiopia.” Climate 12 (11): 169. https://doi.org/10.3390/cli12110169
12. Farauta, B., C. Egbule, A. Agwu, Y. Idrisa, and N. Onyekuru. 2012. “Farmers’ adaptation initiatives to the impact of climate change on agriculture in northern Nigeria.” Journal of Agricultural Extension 16 (1): 132–144. https://doi.org/10.4314/jae.v16i1.13
13. Fletcher, T.D., W. Shuster, W.F. Hunt, R. Ashley, D. Butler, S. Arthur, S. Trowsdale, et al. 2015. “SUDS, LID, BMPs, WSUD and more – The evolution and application of terminology surrounding urban drainage.” Urban Water Journal 12 (7): 525–542. https://doi.org/10.1080/1573062X.2014.916314
14. Gabol, I. 2023. Heavy Rain in Lahore Causes Urban Flooding Again. Dawn, Karachi, Lahore, Islamabad. Available online: https://www.dawn.com/news/1776785
15. Government of Khyber Pakhtunkhwa (GKP). 2024. Market Rate System (MRS). Government of Khyber Pakhtunkhwa Finance Department. Available online: https://www.scribd.com/document/785541587/1723456836-Market-Rate-System-2024-1st-Bi-Annual
16. Gudmundsson, L. 2016. “qmap: Statistical Transformations for Post-Processing Climate Model Output.” R package version, 1.0-4. CRAN. https://doi.org/10.32614/CRAN.package.qmap
17. Harvard Humanitarian Initiative, United Nations Office for the Coordination of Humanitarian Affairs and International Rescue Committee. (n.d.). KoboToolbox. https://www.kobotoolbox.org/ 
18. Hausfather, Z. 2018. “Explainer: How ‘Shared Socioeconomic Pathways’ explore future climate change?” CarbonBrief Clear on Climate. Available online: https://www.carbonbrief.org/explainer-how-shared-socioeconomic-pathways-explore-future-climate-change/
19. Hemanandhini, S. and L. Vignesh Rajkumar. 2023. “Performance evaluation of CMIP6 climate models for selecting a suitable GCM for future precipitation at different places of Tamil Nadu.” Environmental Monitoring and Assessment 195 (8): 928. https://doi.org/10.1007/s10661-023-11454-9?urlappend=%3Futm_source%3Dresearchgate.net%26utm_medium%3Darticle
20. Herman, L. and Z. Stachoň. 2016. “Comparison of user performance with interactive and static 3D visualization–pilot study.” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 41, 655–661. https://doi.org/10.5194/isprsarchives-XLI-B2-655-2016
21. Hsu, M., S. Chen, and T. Chang. 2002. “Dynamic inundation simulation of storm water interaction between sewer system and overland flows.” Journal of the Chinese Institute of Engineers 25 (2): 171–177. https://doi.org/10.1080/02533839.2002.9670691
22. Hu, M., X. Zhang, Y.L. Siu, Y. Li, K. Tanaka, H. Yang, and Y. Xu. 2018. “Flood mitigation by permeable pavements in Chinese sponge city construction.” Water 10 (2): 172. https://doi.org/10.3390/w10020172?urlappend=%3Futm_source%3Dresearchgate.net%26utm_medium%3Darticle
23. Hu, S., Y. Fan, and T. Zhang. 2020. “Assessing the Effect of Land Use Change on Surface Runoff in a Rapidly Urbanized City: A Case Study of the Central Area of Beijing.” Land 9 (1): 17. https://www.mdpi.com/2073-445X/9/1/17
24. Humphries, U.W., M. Waqas, P.T. Hlaing, P. Dechpichai, and A. Wangwongchai. 2024. “Assessment of CMIP6 GCMs for selecting a suitable climate model for precipitation projections in Southern Thailand.” Results in Engineering 23, 102417. https://doi.org/10.1016/j.rineng.2024.102417
25. IUCN. 2016. Green Infrastructure: Origin & Definition, Added Value, Implementation/ Practice. Available online: https://iucn.org/sites/default/files/import/downloads/green_infrastructure_lawrence_j_w.pdf
26. Jose, D.M. and G.S. Dwarakish. 2022. “Bias Correction and Trend Analysis of Temperature Data by a High-Resolution CMIP6 Model over a Tropical River Basin.” Asia-Pacific Journal of Atmospheric Sciences 58 (1): 97–115. https://doi.org/10.1007/s13143-021-00240-7
27. Karamouz, M., A. Hosseinpour, and S. Nazif. 2011. “Improvement of urban drainage system performance under climate change impact: Case study.” Journal of Hydrologic Engineering 16 (5): 395–412. https://doi.org/10.1061/(ASCE)HE.1943-5584.0000317
28. Khan, M., A. Uzair, U. Ali, and W. Khan. 2020. “Flood modeling of Naray-Khwar using HEC-RAS.” Journal of Himalayan Earth Sciences 53 (1): 1–11. https://doi.org/10.13140/RG.2.2.21003.92964
29. Kilsedar, C.E., F. Fissore, F. Pirotti, and M.A. Brovelli. 2019. “Extraction And Visualization of 3D Building Models in Urban Areas for Flood Simulation.” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-2/W11, 669–673. https://doi.org/10.5194/isprs-archives-XLII-2-W11-669-2019
30. Kim, J. and J. Kang. 2023. “Development of hazard capacity factor design model for net-zero: Evaluation of the flood adaptation effects considering green-gray infrastructure interaction.” Sustainable Cities and Society 96, 104625.
31. Kim, K., R. Kim, J. Choi, and S. Kim. 2022. “The applicability of LID facilities as an adaptation strategy of urban CSOs management for climate change.” Water Supply 22 (1): 75–88. https://doi.org/10.2166/ws.2021.285
32. Latifi, M., G. Rakhshandehroo, M.R. Nikoo, and M.G. Mooselu. 2023. “Multi-stakeholder stochastic optimization of urban low impact developments for climate consistency under uncertainty.” Journal of Cleaner Production 382, 135259. https://doi.org/10.1016/j.jclepro.2022.135259
33. Leandro, J., S. Djordjević, A. Chen, D. Savić, and M. Stanić. 2011. “Calibration of a 1D/1D urban flood model using 1D/2D model results in the absence of field data.” Water Science and Technology 64 (5): 1016–1024. https://doi.org/10.2166/wst.2011.467
34. Loc, H.H., A. Emadzadeh, E. Park, P. Nontikansak, and R.C. Deo. 2023. “The Great 2011 Thailand flood disaster revisited: Could it have been mitigated by different dam operations based on better weather forecasts?” Environmental Research 216, 114493. https://doi.org/10.1016/j.envres.2022.114493
35. Loc, H.H., E. Park, D. Chitwatkulsiri, J. Lim, S-H. Yun, L. Maneechot, and D. Minh Phuong. 2020. “Local rainfall or river overflow? Re-evaluating the cause of the Great 2011 Thailand flood.” Journal of Hydrology 589, 125368. https://doi.org/10.1016/j.jhydrol.2020.125368
36. Masseroni, D. and A. Cislaghi. 2016. “Green roof benefits for reducing flood risk at the catchment scale.” Environmental Earth Sciences 75 (7): 579. https://doi.org/10.1007/s12665-016-5377-z
37. Miller, J.D. and M. Hutchins. 2017. “The impacts of urbanisation and climate change on urban flooding and urban water quality: A review of the evidence concerning the United Kingdom.” Journal of Hydrology: Regional Studies 12, 345–362. https://doi.org/10.1016/j.ejrh.2017.06.006
38. Obubu, J.P., S. Mengistou, T. Fetahi, T. Alamirew, R. Odong, and S. Ekwacu. 2021. “Recent Climate Change in the Lake Kyoga Basin, Uganda: An Analysis Using Short-Term and Long-Term Data with Standardized Precipitation and Anomaly Indexes.” Climate 9 (12): 179. https://www.mdpi.com/2225-1154/9/12/179
39. Padulano, R., G. Rianna, P. Costabile, C. Costanzo, G. Del Giudice, and P. Mercogliano. 2021. “Propagation of variability in climate projections within urban flood modelling: A multi-purpose impact analysis.” Journal of Hydrology 602, 126756. https://doi.org/10.1016/j.jhydrol.2021.126756
40. Pindyck, R.S. 2013. “Climate change policy: what do the models tell us?” Journal of Economic Literature 51 (3): 860–872. https://doi.org/10.1257/jel.51.3.860
41. Prudhomme, C., R.L. Wilby, S. Crooks, A.L. Kay, and N.S. Reynard. 2010. “Scenario-neutral approach to climate change impact studies: application to flood risk.” Journal of Hydrology 390 (3–4): 198–209. https://doi.org/10.1016/j.jhydrol.2010.06.043
42. PUB. 2018. Active, Beautiful, Clean Waters Design Guidelines, Fourth Edition. Singapore ABC Waters Programme. Available online: https://www.pub.gov.sg/Professionals/Working-on-ABC-Waterways/ABC-Waters-Design-Guidelines
43. Qin, H., Z. Li, and G. Fu. 2013. “The effects of low impact development on urban flooding under different rainfall characteristics.” Journal of Environmental Management 129, 577–585. https://doi.org/10.1016/j.jenvman.2013.08.026
44. Rahman, A-U., A. Khan, N. Haq, Samiullah, and R. Shaw. 2019. “Soil sealing and depletion of groundwater in rapidly growing Peshawar City District, Pakistan.” In: Ray, B., Shaw, R. (eds.) Urban Drought: Emerging Water Challenges in Asia, 289–309. https://doi.org/10.1007/978-981-10-8947-3_17?urlappend=%3Futm_source%3Dresearchgate.net%26utm_medium%3Darticle
45. Roesner, L.A., J.A. Aldrich, and R.E. Dickinson. 1988. Storm Water Management Model, Version 4 Part B: EXTRAN Addendum. U.S. Environmental Protection Agency, Washington, DC, USA. 3, 88.
46. Rosenberger, L., J. Leandro, S. Pauleit, and S. Erlwein. 2021. “Sustainable stormwater management under the impact of climate change and urban densification.” Journal of Hydrology 596, 126137. https://doi.org/10.1016/j.jhydrol.2021.126137
47. Rossman, L.A. 2006. Storm Water Management Model, Quality Assurance Report: Dynamic Wave Flow Routing. US Environmental Protection Agency, Office of Research and Development, Cincinnati, OH, USA. https://nepis.epa.gov/Exe/tiff2png.cgi/P10089TF.PNG?-r+75+-g+7+D%3A%5CZYFILES%5CINDEX%20DATA%5C06THRU10%5CTIFF%5C00000857%5CP10089TF.TIF
48. Rossman, L.A. and W. Huber. 2017. Storm Water Management Model Reference Manual Volume II–Hydraulics. US Environmental Protection Agency, Washington, DC, USA. 2, 190.
49. Saadatpour, M., F. Delkhosh, A. Afshar, and S.S.S. Solis. 2020. “Developing a simulation-optimization approach to allocate low impact development practices for managing hydrological alterations in urban watershed.” Sustainable Cities and Society 61, 102334. https://doi.org/10.1016/j.scs.2020.102334
50. Si, S., J. Li, Y. Jiang, Y. Wang, and L. Liu. 2022. “The response of runoff pollution control to initial runoff volume capture in sponge city construction using SWMM.” Applied Sciences 12 (11): 5617. https://doi.org/10.3390/app12115617
51. Song, C. 2022. “Application of nature-based measures in China’s sponge city initiative: Current trends and perspectives.” Nature-Based Solutions 2, 100010. https://doi.org/10.1016/j.nbsj.2022.100010
52. Suresh, A., S. Pekkat, and S. Subbiah. 2023. “Quantifying the efficacy of Low Impact Developments (LIDs) for flood reduction in micro-urban watersheds incorporating climate change.” Sustainable Cities and Society 95, 104601. https://doi.org/10.1016/j.scs.2023.104601
53. Surwase, T. and P. Manjusree. 2019. “Urban flood simulation-A case study of Hyderabad city.” National Conference on Flood Early Warning for Disaster Risk Reduction, 30–31 May 2019, Hyderabad, 133–143.
54. Teutschbein, C. and J. Seibert. 2012. “Bias correction of regional climate model simulations for hydrological climate-change impact studies: Review and evaluation of different methods.” Journal of Hydrology 456, 12–29. https://doi.org/10.1016/j.jhydrol.2012.05.052
55. UNDESA. 2020. World Social Report, Inequality in a Rapidly Changing World. Department of Economic and Social Affairs, United Nations Secretariat, NY City, USA. Available online: https://www.un.org/development/desa/dspd/wp-content/uploads/sites/22/2020/02/World-Social-Report2020-FullReport.pdf
56. UNISDR. 2015. The Human Cost of Weather-Related Disasters 1996-2015. United Nations Centre for Research on the Epidemiology of Disasters CRED. Belgium. Available online: https://www.preventionweb.net/files/46796_cop21weatherdisastersreport2015.pdf
57. Venables, B., D. Smith, R. Gentleman, and R. Ihaka. 2012. “Notes on R: A Programming Environment for Data Analysis and Graphics.” Dept. of Statistics, University of Adelaide and University of Auckland, Auckland, NZ.
58. Vijayaraghavan, K., B.K. Biswal, M.G. Adam, S.H. Soh, D.L. Tsen-Tieng, A.P. Davis, S.H. Chew, et al. 2021. “Bioretention systems for stormwater management: Recent advances and future prospects.” Journal of Environmental Management 292, 112766. https://doi.org/10.1016/j.jenvman.2021.112766
59. Wang, C., J. Hou, D. Miller, I. Brown, and Y. Jiang. 2019. “Flood risk management in sponge cities: The role of integrated simulation and 3D visualization.” International Journal of Disaster Risk Reduction 39, 101139. https://doi.org/10.1016/j.ijdrr.2019.101139 
60. Wang, M., Y. Zhang, D. Zhang, Y. Zheng, S. Li, and S.K. Tan. 2021. “Life-cycle cost analysis and resilience consideration for coupled grey infrastructure and low-impact development practices.” Sustainable Cities and Society 75, 103358. https://doi.org/10.1016/j.scs.2021.103358
61. WASA. 2025. Water and Sanitation Agency, Punjab. Lahore Development Authority. Accessed online: https://wasa.punjab.gov.pk/
62. Water by Design. 2014. Bioretention Technical Design Guidelines (2014). Healthy Land & Water, Brisbane, AU. https://waterbydesign.com.au/download/bioretention-technical-design-guidelines
63. Wilbers, G-J., K.D. Bruin, I. Seifert-Dähnn, W. Lekkerkerk, H. Li, and M. Budding-Polo Ballinas. 2022. “Investing in Urban Blue-Green Infrastructure—Assessing the Costs and Benefits of Stormwater Management in a Peri-Urban Catchment in Oslo, Norway.” Sustainability 14 (3): 1934. https://doi.org/10.3390/su14031934
64. WSSP. 2025. Zones of Water and Sanitation Services Peshawar. Water and Sanitation Services, Peshawar. Accessed online: https://wsspeshawar.org.pk/
65. Xiong, L., L. Yan, T. Du, P. Yan, L. Li, and W. Xu. 2019. “Impacts of climate change on urban extreme rainfall and drainage infrastructure performance: A case study in Wuhan City, China.” Irrigation and Drainage 68 (2): 152–164.