Usman Mohseni *¹; Priyanka K. Jat ¹; Vempati Siriteja ¹

1, Civil Engineering Department, Indian Institute of Technology Roorkee, India, Roorkee, Uttarakhand, India

 

Manuscript link
http://dx.doi.org/10.30493/DAS.2025.490282

Abstract

Floods are among the most common natural disasters worldwide, causing significant harm to human lives, socio-economic situations, and the environment. Consequently, it is essential to define and categorize flood risk zones to efficiently manage floods and make educated decisions regarding the mitigation of associated risks. This study assesses flood hazard and vulnerability independently before integrating them to evaluate cyclone-induced pluvial flooding risk in the Chennai districts of Tamil Nadu. For that purpose, various remote sensing satellite data, including the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) and Landsat 8 Operational Land Imager (OLI), were integrated with geomorphological, hydrological, and socio-economic data within a Geographic Information System (GIS) framework. The integration was accomplished utilizing two distinct multi-criteria analysis (MCA) methodologies: the analytical hierarchy process (AHP) and weighted overlay (WO) to develop a comprehensive flood risk map for the Chennai district. Both techniques effectively produced a satisfactory flood risk map delineating four zones of flood hazard, vulnerability, and risk. According to AHP flood risk map, the southern and central northern districts, representing 33.85% of the total area under investigation, are at very high flood risk. The results indicate that the Geographic Information System-Multi-Criteria Analysis (GIS-MCA) framework is an effective and practical instrument for flood catastrophe management, providing a dependable basis for flood mitigation and decision-making. The research highlights the importance of combining various data sources and analytical methods to generate practical insights for policymakers and disaster management officials.

Keywords: Pluvial flooding, Analytical hierarchy process, Flood hazard map, Flood vulnerability map, Flood risk map

Introduction

Floods constitute a continual and intensifying global threat, marked by increased frequency, severity, and significant repercussions across multiple domains. Floods, recognized as an unavoidable natural phenomenon, result in significant consequences, such as population displacement, fatalities, property destruction, and environmental degradation [1-3]. Cyclone-induced flooding represents a significant hazard to communities and ecosystems in India, historically resulting in extensive destruction. Official reports indicated thousands of annual fatalities and significant economic losses attributable to cyclones during the past two decades. Significant storms like Fani in 2019 and Amphan in 2020 wreaked devastation on the eastern coast, highlighting the critical need for comprehensive disaster planning and mitigation policies to combat susceptibility to cyclone-induced floods.

Recently, several methodologies and models have been developed and implemented for flood risk zoning, with satellite imagery and geographic information systems (GIS) increasingly dominating due to advancements in satellite technology [4]. Multi-criteria analysis (MCA) is a crucial decision-making method in the GIS context, particularly in research pertaining to natural risks [5][6]. The amalgamation of GIS with MCA has been thoroughly investigated for flood hazard evaluation in many studies [7-9]. An assessment by [10] underscored the Analytical Hierarchy Process (AHP) as the principal multi-criteria analysis (MCA) method in flood risk management, owing to its effectiveness in measuring spatial flood hazard variability and related concerns. [11][12]. The Ordered Weighted Averaging (OWA) approach has increasingly been utilized in flood hazard studies [13][14].

AHP demonstrates efficacy in creating automated techniques to measure the spatial variability of flood threats and related issues [10][11]. A previous study [12] superimposed seven criteria and employed the resulting index along with the AHP model to delineate the regional flood risk in Greece. The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) has been progressively utilized in flood risk management because to its capacity to address several attributes [10]. Previously, TOPSIS was employed to enhance flood management by evaluating flood mitigation strategies according to multiple parameters [15]. In recent years, OWA method has been progressively and efficiently utilized in flood hazard research. A GIS-OWA model was developed to generate flood potential maps for Meshkinshahr city, Iran, serving as an effective instrument for identifying and categorizing flood-prone regions inside Meshkinshahr and its adjacent drainage basin [13]. An MCA framework that integrates GIS, fuzzy AHP, and OWA methods, utilizing geographical, hydrological, and other aspects as criteria, was proposed on the lower Han River region [14].

Taking into consideration the aforementioned applications, this research aimed to design and test GIS-MCA methodology in the Chennai districts of Tamil Nadu, India, specifically aimed at cyclone-induced pluvial flooding. This research represents a pioneering initiative to develop flood hazard, vulnerability, and risk maps for cyclone-induced pluvial flooding, signifying a vital advancement in comprehensive disaster management techniques.

Material and Methods

Study area

The study concentrates on the essential task of flood risk mapping for the Chennai districts of Tamil Nadu, India. The district is situated in the northern region of Tamil Nadu. The Chennai district is bordered by the Bay of Bengal to the east, Tiruvallur to the north, and Kancheepuram to the south. The overall geographical area is 424 Km2, divided into 15 administrative districts (taluks) (Fig. 1).

Data description

SRTM DEM data with a 30-meter resolution was utilized to generate slope and drainage density maps. A Landsat 8 OLI/TIRS picture from December 2023 was employed for land use and land cover (LULC), normalized difference vegetation index (NDVI), topographic wetness index (TWI), and distance maps for rivers and roads. Soil, geology, and geomorphology data were acquired from FAO/UNESCO and Bhukosh. Rainfall data from TNSMART for the creation of a rainfall map. Socio-economic data, including population density, gender distribution, literacy rates, houses, and hospitals, were aggregated into spatial layers for the evaluation of flood hazard, susceptibility, and risk in the research area. The data included in this investigation have been obtained from various sources (Table 1).

Methodology

The approach for developing the flood risk map was segmented into three components. Preparation of themed maps for hydrogeomorphic hazards and socio-economic vulnerability indicators. Developing flood hazard and vulnerability maps with two distinct multi-criteria analysis (MCA) methodologies: the analytical hierarchy process (AHP) and weighted overlay (WO). Developing a flood risk map by merging flood hazard and vulnerability maps. Figure 2 delineates the comprehensive technique employed in this investigation.

Preparation of thematic layers

The primary challenge in developing flood hazard and vulnerability zoning maps lies in identifying the factors within the study area and regional flood characteristics that lead to flood disasters. Previous literature on flood hazard and vulnerability [8][9][15] identifies eleven hydrogeomorphic hazard indicators (elevation, slope, rainfall, soil, geology, geomorphology, drainage density, NDVI, TWI, distance to river, and storm surge) and seven socio-economic vulnerability indicators (population density, female density, literacy rate, number of households, land use and land cover, distance to road, and number of hospitals) for this study. Detailed information regarding the hydrogeomorphic hazard indicators and socio-economic vulnerability indicators utilized in this research is available in the supplementary material (Supplementary Tables 1 and 2).

Multi-Criteria Analysis (MCA) techniques

Weighted Overlay (WO)

The weighted overlay method, supported in numerous publications [16-18], demonstrates a clear and thorough technique for assessing areas susceptible to flooding. This analysis utilized 11 hydrogeomorphic hazard indicators and 7 socio-economic vulnerability indicators to analyze flood hazard and vulnerability mapping. For analytical purposes, each cell in the distinct map layers necessitated categorization into a standardized preference scale, ranging from 1 to 9, where 9 indicates the utmost favorability. The integration of all thematic maps was conducted utilizing the Weighted Overlay Model (WOM):

where, 𝑊𝑖 is the weight ith factor map, 𝑆𝑖𝑗 is the ith spatial class weight of jth factor map, 𝑆 is the spatial unit value in output map.

Analytical Hierarchy Process (AHP)

The AHP method is the most widely utilized MCA technique that organizes complex decision problems into a hierarchical structure, evaluates criteria weights through pairwise comparisons with numerical importance scales [19], and offers a systematic approach for assessing and integrating the influences of various factors in flood risk management, thereby simplifying the analysis and management of intricate risk-related decisions [14][20].

At each tier of the decision-criteria hierarchy framework, a control process including pairwise comparisons of the criteria is conducted using Saaty’s comparative scale [11]. The given weights were normalized using the eigenvector method and subsequently evaluated for consistency by calculating a consistency ratio (CR), which must be below 0.1 to be deemed acceptable [9][20]. Alternatively, value assessments must be re-evaluated. The final stage is integrating the weights from the attribute map layers obtained using AHP with the attribute map layers within the GIS environment.

Flood risk

The amalgamation of flood risks and vulnerabilities in a certain locale delineates flood risk, requiring a methodical evaluation, collection, and analysis of relevant data. Due to its ability to generate inundation maps, GIS has become an essential instrument for flood mapping and analysis.

The study initially evaluated hazards and vulnerabilities independently using the GIS-MCA approach. Thereafter, by amalgamating hazard and vulnerability data, the study identified and categorized flood risk zones. The flood risk assessment map was ultimately generated with the Tabulate Area tool of ArcGIS utilizing the Flood Risk equation [21][22]:

Flood Risk = Flood Hazard × Flood Vulnerability

Results and Discussion

Thematic layers of hazard and vulnerability indicators

The current study has generated 11 hydrogeomorphic danger levels and 7 socio-economic vulnerability layers utilizing the available datasets. The map of rainfall and storm surge is produced for hydrological analysis. Likewise, slope, geology, geomorphology, distance to stream, elevation, NDVI, TWI, and drainage density maps are produced for geomorphic analysis, while population density, female density, literacy rate, land use/land cover, number of hospitals, number of households, and distance to road maps are created for socio-economic vulnerability analysis. Thematic layers are categorized into five classes utilizing natural breaks, which are subsequently classed. The weights allocated to Hydrogeomorphic Hazard Indicators and socioeconomic vulnerability indicators utilizing WO and AHP methodologies are included in the supplemental material (Supplementary Tables 3 and 4).

The comprehensive AHP analysis of 11 hydrogeomorphic hazards and 7 socio-economic vulnerabilities, including the pairwise comparison matrix among the theme layers and the consistency ratio (CR), was assessed and documented in Supplementary Tables 5-10. Figures 3 and 4 illustrate the theme maps for hydrogeomorphic hazards and socioeconomic vulnerability indicators in the Chennai district.

Flood hazard map (FHM)

The incorporation of hydrogeomorphic thematic layers via Weighted Overlay (WO) and Analytic Hierarchy Process (AHP) approaches has yielded significant insights into the spatial distribution of flood dangers within the study area. The findings demonstrate that the southern and eastern areas of the region are most vulnerable to high and very high flood hazards, as illustrated by both WO and AHP flood hazard maps (Fig. 5). This spatial pattern corresponds with the region’s hydrogeomorphic attributes, such as high precipitation, significant storm surges, low elevation, and the occurrence of Regosols soil type (Fig. 3). Previous studies have universally acknowledged these characteristics as essential contributors to flood susceptibility [23][24].

Overall, 24.41% of the research area was designated as very high flood hazard zones according to AHP, although this proportion was minimal according to WO. Furthermore, a significant portion of the research region (55.87%) was identified as being at high flood risk according to the Analytic Hierarchy Process (AHP), whereas a comparable percentage (68.59%) was categorized as moderate flood risk areas according to Weighted Overlay (WO) (Table 2). These observations highlight the sensitivity of the chosen methodology to the weighting of hazard indicators. which is consistent with studies that have demonstrated the effectiveness of AHP in capturing nuanced flood risk patterns [15].

Flood vulnerability map (FVM)

The flood vulnerability maps were formulated by employing seven thematic layers derived from socio-economic vulnerability analysis. Within these layers, population density significantly contributes to the region’s vulnerability to flooding compared to Land Use and Land Cover (LULC). AHP and WO flood vulnerability maps were almost identical, except for the northern part of the study area where minor classification differences were observed (Fig. 6). The high similarity in classification between both methods reflected on the areal extent of each vulnerability category (Table 2).

Flood risk map (FRM)

The final risk maps were generated by utilizing the Raster Calculator tool to multiply the classified flood hazard and vulnerability maps. The resultant map is classified into five basic categories—Very Low, Low, Moderate, High, and Very High—utilizing the Reclassify tool. The Tabulate Area tool (Zonal tool) is utilized to ascertain the area of each category. Figure 7 illustrates the flood risk maps for the sub-districts of Chennai, developed utilizing both methodologies. The generated risk map has been juxtaposed with the risk map created by Enviraj© Consulting for the Chennai district (Fig. 7 c) (https://oer.Enviraj.com/maps/chennai-flood-risk-map.html). The flood risk map generated using AHP approach was extremely similar to the pre-existing map (Fig. 7 B and C) as they both identify similar very high and high-risk zones. The lowland regions adjacent to Pallikaranai marsh (south) have a historical tendency to experience flooding during each rainy season [25], exhibiting a significant concentration of susceptible areas according to 2020 data (Fig. 7 C). This region was prominently identified by the AHP flood risk map as a zone of very high flood risk (Fig. 7 B).

The ROC curves for the two flood risk maps (FRMs) created in this work visualized in order to offer a better understanding of their efficacy compared to the Enviraj© Consulting reference map (Fig. 8). The data unequivocally demonstrate that the AHP-based FRM surpasses the WOA-based FRM in evaluating flood risk. Both models exhibit high true positive rate (TPR) and true negative rate (TNR) values, as indicated by the closeness of their curves to the top-left corner of the ROC space. Nonetheless, the AHP-based FRM demonstrates enhanced performance with an Area Under the Curve (AUC) value of 0.84, underscoring its efficacy in precisely assessing flood risk relative to the WOA-based method (AUC=0.71).

A variety of strategies have been employed in flood risk assessment. The remarkable accuracy of MCA approaches in flood risk evaluation is significantly beneficial for decision-making in flood management in vulnerable areas. This method calculates weights according to the relative significance of flood indicators, considering expert opinion, published literature, and the inter-correlation among the indicators, to produce more reliable predictions, which have

been applied in environmental management. AHP has previously exhibited greater accuracy compared to other models, including WO. Similarly, the current study revealed that the combined GIS-MCA methods, particularly AHP, displayed superior accuracy compared to WO, highlighting the importance of this approach in improving the prediction of FHM, FVM, and FRM. A comparative review of the two methodologies indicated identical estimates for flood-prone locations within the research area, providing a crucial basis for the development of successful flood management policies. According to AHP flood risk map, the southern and central northern sub-districts are at very high flood risk (Fig. 7 B) representing 33.85% of the total area under investigation (Table 2). The results indicated that AHP is a dependable and effective approach for evaluating and delineating flood risk within a GIS framework. Therefore, this methodology may be recommended for application in different regions.

Conclusion

The paper presents a case study utilizing WO and AHP as GIS-MCA frameworks to delineate cyclone-induced flood risks in Chennai District, Tamil Nadu. This novel method provides significant assistance to decision-makers and policymakers in evaluating and thoroughly reviewing flooding events in the region. This study provides essential information by classifying Chennai sub-districts into specific risk categories. The maps produced in this study are based not on historical inundation data but on a comprehensive synthesis of various elements contributing to flooding. This method includes hydrological, geomorphological, and socio-economic data, providing a thorough perspective. The use of high-resolution pictures and DEMs markedly improves the efficiency of mapping processes. Analytical hierarchy process (AHP) proved to be more efficient in capturing hydrogeomorphic details associated with flood hazard compared to weighted overlay (WO). Consequently, AHP-MCA framework can be suggested in flood susceptibility and flood hazard assessment under similar study settings.

References

1. Sani Y, Noordin A, Ranya FA. Multicriteria analysis for flood vulnerable areas in Hadejia-Jama’are river basin. Nigeria. Eur. J. Sci. Res. 2010;42(1):71-83.
2. Wang Y, Li Z, Tang Z, Zeng G. A GIS-Based Spatial Multi-Criteria Approach for Flood Risk Assessment in the Dongting Lake Region, Hunan, Central China. Water Resour. Manag. 2011;25(13):3465–84. DOI
3. Tsakiris G. Flood risk assessment: concepts, modelling, applications. Nat. Hazards Earth Syst. Sci. 2014;14(5):1361–9. DOI
4. Asare-Kyei D, Forkuor G, Venus V. Modeling Flood Hazard Zones at the Sub-District Level with the Rational Model Integrated with GIS and Remote Sensing Approaches. Water. 2015;7(7):3531–64. DOI
5. Jozaghi A, Alizadeh B, Hatami M, Flood I, Khorrami M, Khodaei N, Ghasemi Tousi E. A comparative study of the AHP and TOPSIS techniques for dam site selection using GIS: A case study of Sistan and Baluchestan Province, Iran. Geosciences. 2018;8(12):494. DOI
6. Nefeslioglu HA, Sezer EA, Gokceoglu C, Ayas Z. A modified analytical hierarchy process (M-AHP) approach for decision support systems in natural hazard assessments. Comput. Geosci. 2013;59:1–8. DOI
7. Chen H, Ito Y, Sawamukai M, Tokunaga T. Flood hazard assessment in the Kujukuri Plain of Chiba Prefecture, Japan, based on GIS and multicriteria decision analysis. Nat. Hazards. 2015;78(1):105–20. DOI
8. Kourgialas NN, Karatzas GP. A national scale flood hazard mapping methodology: The case of Greece – Protection and adaptation policy approaches. Sci. Total Environ. 2017;601–602:441–52. DOI
9. Shadmehri Toosi A, Calbimonte GH, Nouri H, Alaghmand S. River basin-scale flood hazard assessment using a modified multi-criteria decision analysis approach: A case study. J. Hydrol. 2019;574:660–71. DOI
10. de Brito MM, Evers M. Multi-criteria decision-making for flood risk management: a survey of the current state of the art. Nat. Hazards Earth Syst. Sci. 2016;16(4):1019–33. DOI
11. Dahri N, Abida H. Monte Carlo simulation-aided analytical hierarchy process (AHP) for flood susceptibility mapping in Gabes Basin (southeastern Tunisia). Environ. Earth Sci. 2017;76(7). DOI
12. Kazakis N, Kougias I, Patsialis T. Assessment of flood hazard areas at a regional scale using an index-based approach and Analytical Hierarchy Process: Application in Rhodope–Evros region, Greece. Sci. Total Environ. 2015;538:555–63. DOI
13. Ghaffari Gilandeh A, Sobhani B, Ostadi E. Combining Arc-GIS and OWA model in flooding potential analysis (case study: Meshkinshahr city). Nat. Hazards. 2020;102(3):1435–49. DOI
14. Xiao Y, Yi S, Tang Z. Integrated flood hazard assessment based on spatial ordered weighted averaging method considering spatial heterogeneity of risk preference. Sci. Total Environ. 2017;599–600:1034–46. DOI
15. Pathan AI, Girish Agnihotri P, Said S, Patel D. AHP and TOPSIS based flood risk assessment- a case study of the Navsari City, Gujarat, India. Environ. Monit. Assess. 2022;194(7). DOI
16. Intarawichian N, Dasananda S. Analytical hierarchy process for landslide susceptibility mapping in lower Mae Chaem Watershed, Northern Thailand. Suranaree J. Sci. Technol. 2010;17(3).
17. Ahmed MF, Rogers JD, Ismail EH. A regional level preliminary landslide susceptibility study of the upper Indus river basin. Eur. J. Remote Sens. 2014;47(1):343–73. DOI
18. Khatun M, Hossain ATMS, Sayem HMd, Moniruzzaman M, Ahmed Z, Rahaman KR. Landslide Susceptibility Mapping Using Weighted-Overlay Approach in Rangamati, Bangladesh. Earth Syst. Environ. 2022;7(1):223–35. DOI
19. Saaty TL. Decision making with the analytic hierarchy process. Int. J. Serv. Sci. 2008;1(1):83-98. DOI
20. Stefanidis S, Stathis D. Assessment of flood hazard based on natural and anthropogenic factors using analytic hierarchy process (AHP). Nat. Hazards. 2013;68(2):569–85. DOI
21. Abdelkarim A, Al-Alola SS, Alogayell HM, Mohamed SA, Alkadi II, Ismail IY. Integration of GIS-Based Multicriteria Decision Analysis and Analytic Hierarchy Process to Assess Flood Hazard on the Al-Shamal Train Pathway in Al-Qurayyat Region, Kingdom of Saudi Arabia. Water. 2020;12(6):1702. DOI
22. Abdrabo KI, Kantoush SA, Saber M, Sumi T, Habiba OM, Elleithy D, Elboshy B. Integrated methodology for urban flood risk mapping at the microscale in ungauged regions: a case study of Hurghada, Egypt. Remote Sens. 2020;12(21):3548. DOI
23. Mahmoud SH, Gan TY. Multi-criteria approach to develop flood susceptibility maps in arid regions of Middle East. J. Clean. Prod. 2018;196:216-29. DOI
24. Mishra K, Sinha R. Flood risk assessment in the Kosi megafan using multi-criteria decision analysis: A hydro-geomorphic approach. Geomorphology. 2020;350:106861. DOI
25. Azeez PA, Bhupathy S, Ranjini J, Dhanya R, Raj PN. Management Plan for the Eco-restoration of Pallikaranai Reserve Forest. Salim Ali Centre for Ornithology and Natural History, Coimbatore. 2007.

Cite this article:

Mohseni, U., Jat, P. K., Siriteja, V. Multi-criteria analysis-based mapping of the cyclone-induced pluvial flooding in coastal areas of India. DYSONA – Applied Science, 2025; 6(2): 309-321. doi: 10.30493/das.2025.490282