Peri-Urban Floodscapes: Identifying and Analyzing Flood Risk Areas in North Bhubaneswar in Eastern India

Abstract

Peri-urban catchment areas are increasingly susceptible to floods due to rapid land use transformations and unplanned urban expansion. This study comprehensively examines flood vulnerability in the rapidly developing peri-urban areas of North Bhubaneswar, focusing on significant changes in Land Use/Land Cover (LULC) and hydrological dynamics from 2004 to 2024, utilizing Geographic Information System (GIS) tools. The analysis reveals substantial shifts in land use patterns, with the urban footprint expanding by 71.8%, cropland decreasing by 21.7%, and forest areas by 13.6%. These changes have led to increased impervious surfaces, resulting in higher surface runoff and decreased groundwater recharge, thereby exacerbating flood risks in the region. The GRID-based vulnerability analysis classifies 90 villages within the catchment area based on their vulnerability levels, identifying 20 villages as high-risk areas requiring urgent attention, 44 villages as medium vulnerable, and 26 villages as low vulnerable. These classifications are based on factors such as proximity to drainage networks, slope, geomorphology, and LULC characteristics, with areas near drainage channels and low-lying regions being prone to flooding. The analysis integrates multiple factors to provide a comprehensive assessment of flood risk, highlighting the need for sustainable land use planning, conservation of vegetated areas, and the implementation of advanced flood prevention strategies in the peri-urban areas. Extending this research to other fringe regions could offer further valuable insights, guiding flood prevention and sustainable development strategies for areas undergoing significant land use transformations to effectively mitigate future flood risks.

1. Introduction

Peri-urban areas are experiencing growing pressures due to population growth, infrastructure development, and economic expansion, resulting in significant land use changes with poor land use planning [1,2,3,4]. These changes involve converting agricultural land, natural vegetation, and wetlands into developed areas, leading to ecosystem degradation and reduced ecosystem services [4,5,6,7]. The consequences of these transformations include blockages and disconnection in natural water flow systems, encroachment on natural drains, and inadequate drainage infrastructure, resulting in pluvial flooding [8,9,10].

The physiographic characteristics of drainage basins, such as their size, shape, slope, drainage density, and the dimensions of contributory streams, are vital factors that can be correlated with various important hydrological phenomena [9,11]. Urbanization alters local hydrological processes significantly by increasing impervious surfaces, modifying natural flow paths of water, leading to concentrated flows, localized flooding, and causing changes in the timing and volume of runoff in urban and peri-urban river catchments [12,13,14]. Increased runoff and sedimentation can lead to erosion, sedimentation of water bodies, and deterioration of water quality, affecting both aquatic ecosystems and water supply [15,16]. In addition, the increased imperviousness of catchments reduces soil infiltration, leading to lower baseflow. This is compounded by reduced evaporation rates in these areas compared with other land covers. Furthermore, modifications in soil permeability due to topographic changes and soil compaction, along with inter-basin water transfers and groundwater pumping, add complexity to this impact [17]. At the catchment scale, which ranges from 10 to 10,000 km², the hydrological impact of urbanization becomes more intricate due to the significant spatial variability of impervious areas [18,19,20], thereby reducing the natural buffering capacity of landscapes [21]. These factors collectively heighten the potential flood risk, especially exacerbated by the increasing imperviousness and population density in fringe areas [22].

The loss of vegetation also results in decreased infiltration rates since there are fewer plants and trees to aid water absorption into the soil [23]. Land use changes can have a substantial impact on catchment characteristics, influencing soil conditions, soil quality, and the risks of urban flooding, both on a large and small scale [22]. Reduced surface roughness alters flow dynamics, accelerating water movement across urban areas. This leads to faster runoff response times and increased runoff volumes during precipitation events [15,24,25]. The modification of flow pathways due to infrastructural development and land use changes further compounds flood risks in peri-urban environments [20,22]. These altered pathways can redirect water and increase the likelihood of flooding in certain areas. Additionally, these regions often experience elevated levels of pollutants in water systems due to runoff carrying contaminants from roads, industrial areas, and urban activities, posing challenges for water quality management [20,26].

The simultaneous occurrence of urbanization and climate change is expected to intensify hydrologic and hydraulic impacts at the urban and peri-urban scale with significant repercussions across various sectors, including floodplain delineation and urban drainage systems. These systems are already being altered by urban expansion, greatly elevating the risk of floods in both urban areas and their surrounding fringes in the future [27,28]. Peri-urban watersheds are susceptible to ongoing hydrological shifts due to rapid urbanization, land use modifications, and soil sealing, which often leads to flash floods, surface water inundation, and riverine flooding, posing risks to infrastructure, properties, and human safety [24,29].

The advancements in remote sensing techniques have made them invaluable for generating ground-truth data, especially in catchments with limited or no gauging. Catchment morphology plays a pivotal role in runoff generation and routing [30,31]. Remote sensing and Geographic Information Systems (GIS) are extensively used to capture detailed geomorphological data with a high level of accuracy [32,33,34,35,36]. Urban flooding is a prevalent issue across watersheds of varying sizes, particularly in areas where communities have settled in flood-prone locations. In smaller watersheds, urban flooding often occurs due to cyclonic or storm rainfall affecting local areas within or near urban settlements. The process of urban development significantly modifies the runoff-producing characteristics of these catchments, exacerbating the risk of flooding [37].

Heavy rainfall events in cities such as Mexico City, Rome, and Beijing result in floods that concentrate water in low-lying areas, leading to waterlogging [38,39]. Cities like Mumbai, Chennai, Delhi, Gurugram, Noida, Kolkata, Bharuch, Surat, and Hyderabad in India have experienced numerous instances of flooding due to urbanization during the past few decades [40]. In Bengaluru, the conversion of lakes into urban areas has disrupted the natural drainage system, leading to isolated lake networks and increased flood risk [41]. Loss of vegetation, disruptions in hydrological drainage networks, and threats to remaining wetlands from pollution and invasive species further exacerbate the flooding challenges [42]. In Kolkata, Chennai, and Mumbai, flood issues stem from a combination of factors due to rampant urbanization and inadequate planning, including the decline in green cover, water bodies, wetlands, and marshlands, caused by other land use transformations, uncontrolled construction, and infrastructure development on floodplains and canals, which disrupt natural stormwater flow [7,40,41,43]. Overall, urbanization-induced changes, loss of natural drainage areas, disruptions in hydrological systems, and inadequate stormwater management practices contribute to the flood vulnerabilities observed in these Indian cities.

The methodologies for flood risk assessment employed across various study areas reveal a diverse range of approaches. Ouma and Taeshi (2014), in Eldoret Municipality in Kenya, utilized an Urban Flood Risk Index combining multi-criteria AHP and GIS, focusing on parameters such as rainfall distribution, elevation, slope, and land use [44]. Lin et. al. (2019) applied a composite Flood Risk Index (FRI) using GIS spatial analysis and AHP to assess flood vulnerability, hazard factors, and resilience capacity in Zhengzhou City in China [45]. In Greece’s Attica Region, Feloni et. al. (2020) used GIS-based multi-criteria analysis with AHP and FAHP techniques to evaluate flood risk through different scenarios and criteria weighting, using software like SAGA-9 and ArcMap 10.3 [46]. Desalegn and Mulu (2021) employed multi-criteria evaluation within GIS, applying pair-wise evaluation techniques to factors like slope, elevation, and land use for flood vulnerability assessment of the Upper Abbay Basin in Ethiopia [47]. For the Keleghai Basin in West Bengal, Roy and Dhar (2024) used AHP to assess flood risk by analyzing elevation, slope, rainfall, NDVI, and distance from the river [48]. Similarly, Kaaviya and Devadas (2021), integrated GIS, remote sensing, and AHP to conduct a multi-criteria decision analysis, assigning weights to parameters and using weighted overlay analysis to depict water resilience in Chennai City, Tamil Nadu [49].

This situation is similarly becoming worse in Bhubaneswar, the capital city of Odisha, where the development process has overlooked the natural drainage and hydrology of the region, resulting in situations of pluvial floods in many densely populated areas of the city during heavy rainfall [33]. The establishment of educational and institutional hubs in the northern part of the city has significantly spurred urban growth, expanding into surrounding peri-urban areas [26] and pressuring both urban and peri-urban catchments. The swift development to meet land demands in these areas poses risks of flooding in the coming years, potentially causing ecological imbalance and compromising essential environmental services. Many studies predominantly focus on urban flooding, often overlooking the specific dynamics and challenges faced by peri-urban regions. There is a notable gap in research that deeply explores the unique factors contributing to flooding in these transitional zones. Furthermore, most flood risk prevention actions over the past decades have focused on corrective rather than preventive measures [28].

Given these gaps in the literature, this study aims to analyze the drainage patterns and land use dynamics within a watershed in northern Bhubaneswar and identify the flood-vulnerable areas in this region. Several areas within this watershed are susceptible to flooding, yet they have not yet been identified or mapped. This study leverages GIS and remote sensing technologies to generate flood vulnerability maps and identify flood-prone zones in fringe regions experiencing rapid land use changes. This region is vulnerable to flooding due to the encroachment of natural drainage systems and wetlands. The resulting data will provide policymakers and planners with significant insights into hydrological changes reflecting the complex interactions between urban development and environmental sustainability. This research further emphasizes the necessity for integrated urban planning and watershed management strategies to implement preventive measures and devise sustainable development plans for fringe areas, thereby reducing the need for corrective actions in the future.

2. Materials and Methods

2.1. Understanding the Study Area: Characteristics and Features

Bhubaneswar, the capital city of Odisha situated in the Eastern Coast of India, has garnered recognition as a smart city, heralding new opportunities and growth prospects. The city spans from 20°5′ N to 20°26′ N and from 85°30′ E to 85°59′ E. Bhubaneswar, known as the Temple City, is renowned for its abundant cultural heritage, highlighted by iconic ancient temples like Lingaraj Temple and Mukteshwar Temple, which showcase remarkable architecture and historical importance. Bhubaneswar has emerged as a major education and information technology hub in recent years, attracting students and professionals from across the country. The eastern and southern regions of the city feature gently sloping terrains with fertile alluvial soil, sculpted by the rivers Daya and Kuakhai, tributaries of the Mahanadi River. In contrast, the northern and western regions feature hard red lateritic soil and scattered hillocks made up of Upper Gondwana shale/sandstone sequences in the western part. The growth of the city is constrained by the presence of a reserve forest in the northwest and floodplains in the east. Bhubaneswar has a humid tropical climate, characterized by an average annual rainfall of 1498 mm. The city experiences hot and humid summers, with temperatures soaring to 46 °C from late March to mid-June, along with dry and cool winters. Monsoons typically occur from June to September, bringing the bulk of the annual precipitation. The population of the city has reached 1,195,000 in 2024. There is a network of ten main natural drains in the city, with nine of them flowing into the Gangua Nala, while Budhi Nala empties into the Kuakhai River.

Figure 1 depicts the location of the study area, encompassing the catchment area located north of Bhubaneswar, covering an area of 287.7 km².

2.2. Data and Methodology

2.2.1. Watershed Delineation and Analysis

Automated extraction techniques were utilized to evaluate morphometric parameters, specifically the delineation of watershed boundaries and the extraction of stream networks. A Cartosat DEM with a 10 m spatial resolution was acquired from BHUVAN, ISRO, and processed using ArcGIS 10.3 software.

Table 1. Satellite imageries and DEM.

Sl. No	Layers	Data	Sources/Method
1	LULC	LISS-IV, 5.8 m/LANDSAT-8), 30 m resolution	NRSC/ISRO/ https://www.usgsearthexplorer.com (accessed on 10 June 2024)
2	Elevation	Cartosat, 10 m resolution	https://bhuvan.nrsc.gov.in/home/index.php (accessed on 10 June 2024)

provides details of the satellite imagery and DEM products used. The Spatial Analyst tool, including sub-tools such as Hydrology, Fill, Flow Direction, and Flow Accumulation, was employed to extract all stream networks and perform stream ordering to determine the drainage hierarchy. Morphometric analysis, such as calculating drainage density, was performed to assess drainage characteristics. Additionally, contour and slope maps were generated from the DEM dataset to evaluate elevation susceptibility [35,50]. Various methods exist for filling voids in DEMs, including Triangular Irregular Network (TIN) for large voids in flat areas, Spline Interpolation for small to medium-sized voids in high-altitude and dissected terrain, and Kriging or Inverse Distance Weighting (IDW) for small to medium-sized voids in relatively flat low-lying areas [51,52]. In the present context, the IDW method was applied. Figure 2 portrays a schematic representation of the methodology used for stream analysis in the study area.

2.2.2. Land Use/Land Cover Analysis

The LISS-IV satellite imagery for the year 2024 was procured from NRSC/ISRO and LANDSAT-8 for the year 2000 was obtained from USGS website and processed using the ArcGIS software. Image enhancement techniques were employed to improve image quality for interpretation purposes. Visual image interpretation methods were utilized to delineate LULC classes within the study area. Ground truthing was conducted in uncertain areas and integrated into the final LULC classes. Spatial statistics were calculated to gain insights into terrain distribution [8]. Figure 3 illustrates the schematic representation of the methodology used for LULC analysis of the study area for the years 2004 and 2024.

2.2.3. Data Validation

The findings were substantiated and triangulated through key officials from various institutions, including the municipal corporation, town planners, builders, and other relevant stakeholders, along with 30 key informant interviews with residents who have lived in the study area for 15 years or more.

3. Results

3.1. Drainage Analysis

The study area encompasses a total area of 287.7 km². Morphometric parameters play a vital role in drainage analysis for watershed planning. They provide essential information about the slope, topography, soil conditions, runoff behaviour, and surface water potential of the watershed, all of which are critical for making informed decisions and formulating effective management strategies [43,53]. Additionally, morphometric analysis aids in understanding drainage basin responses to climate change, assessing drainage characteristics, evaluating flash flood hazards, and studying hydrological processes [54,55].

3.1.1. Stream Order

Stream order analysis is performed using the method introduced by Strahler in 1964. This method provides a hierarchical classification of streams within a watershed, beginning with the smallest and most upstream streams, designated as first-order or fingertip streams When two first-order streams converge, they form a second-order stream. This hierarchical merging continues, with two second-order streams forming a third-order stream, and so on. This numbering system helps identify the order of drainage-network elements and counts segments of each order. Stream length indicates the area’s contribution to the watershed, the steepness of drainage, and the degree of drainage. Steep areas with good drainage typically have many small tributaries, while plains with deep, permeable soils often have longer tributaries, typically perennial streams [56]. In the study area, the highest stream order identified is fifth order, and the watershed is thus classified as a fifth-order watershed. Each stream segment is assigned an order, starting from the first order up to the maximum order present in the area.

Figure 4 illustrates the stream order map, and

Table 2. Stream length.

Sl. No	Stream Order	Length (m)	Total Length (m)
1	1st order	176,335.97	321,585.35
2	2nd order	93,164.5
3	3rd order	52,084.88
4	4th order	24,614.5	45,394.62
5	5th order	20,780.12
	Total		366,979.97

details the length of streams corresponding to each order within the watershed. It is observed that the total stream length is minimum in the fifth order at 20.78 km, and maximum in the first order at 176.33 km. The high frequency of first-order streams indicates significant surface runoff, erosion, sparse vegetation, and a potential for flooding. The combined length of first- to third-order streams is 321.59 km, whereas the total length of higher order streams is 45.39 km. First-order streams show the highest stream frequency. On an average, there is an increasing trend in individual stream length from first order to third order. The total length of stream segments tends to decrease with increasing stream order. However, deviations from this trend suggest high relief and/or moderately steep slopes, along with varying lithology and potential uplift across the watershed [55,57].

3.1.2. Stream Direction

Stream direction analysis is conducted to gain insights into the flow pattern of surface water for potential water resource development. Using GIS tools, the length and direction of each drainage line are computed and then visualized through plotting (Figure 5). The prevailing direction for maximum stream order is northeast to southwest.

3.1.3. Drainage Density

Drainage density is the cumulative length of channels within a drainage basin. It is an important parameter of a basin as it provides a measure of topographic control on the flow of water. Figure 6 illustrates the drainage density map of the study area. Drainage density, which refers to the stream length per unit area of a basin or watershed, is a key aspect of drainage analysis that provides a quantitative measure for dissecting and understanding landforms, influenced by climate, lithology, structural features, and relief history [55,58,59]. The calculated drainage density is 0.24 km/km², indicating moderate drainage densities. Low drainage density is associated with a coarse drainage texture, whereas high drainage density indicates a fine drainage texture and weak impervious subsurface materials and limited vegetation cover, resulting in increased runoff and a high probability of flooding across the basin area [60].

3.1.4. Slope/Elevation Analysis

Slope refers to the steepness or incline of the terrain, typically measured as the angular inclination between a hill’s top and the valley’s bottom. It is influenced by factors such as relief, landform, geological structure, and vegetation cover [61].

The Digital Elevation Model (DEM) of the study area provides a detailed representation of the terrain’s elevation at a high spatial resolution. It captures the elevation data of the land surface and is a crucial component in hydrological and morphometric analyses. The DEM aids in delineating watersheds, analysing slope characteristics, and understanding the topographic features of the area. By visualizing elevation variations and low-lying areas prone to flooding, the terrain’s suitability for different land uses is assessed, and infrastructure development considering the topographic constraints can be identified [62,63,64].

Slope plays a crucial role in morphometric analysis and significantly impacts surface runoff in terms of its speed, quantity, and direction. The steepness of the slope correlates positively with soil erodibility and negatively with infiltration capacity. A higher slope percentage leads to faster runoff, increased soil erosion, and reduced infiltration capacity, assuming other factors remain constant [57,65]. Conversely, lower slopes encourage greater infiltration compared with steeper slopes. Slope percentage was determined from DEM data using a Cartosat-DEM [35], using the tangent function shown in

Table 3. Slope in degrees and distribution across the study area.

Sl. No	Slope (Degree)	Area (km2)	Percentage (%)
1	0–1.6	266.47	94.11
2	1.7–5.1	14.33	5.06
3	5.2–10	2.22	0.78
4	11–27	0.14	0.05

and Figure 7a,b. This study reveals that the majority of the catchment area (94.11%) consists of slopes ranging from 0° to 1.6°. Lower slopes (1.7–5.1°) cover 5.06% of the region, while higher slopes (5.2–10°), with the highest being 11–27°, are mainly concentrated in the western areas, which constitute only 0.78% and 0.05% of the total area, respectively. These lower slope areas are ideal for agriculture and human settlement due to their high nutrient content and groundwater recharge potential. Proper management of sediment in these regions could substantially enhance agricultural productivity.

3.2. Land Use/Land Cover and NDVI

The LULC statistics from 2004 to 2024 reveal several significant changes in the study area, as presented in

Table 4. Land Use/Land Cover area statistics.

Sl. No	Class	2004		2024
		(km2)	%	(km2)	%
1	Cropland	84.53	29.38	62.80	21.83
2	Forest	61.95	21.53	53.55	18.61
3	Playground/Stadium	0.12	0.04	1.01	0.35
4	Railways	0.73	0.25	0.89	0.31
5	Road Network	0.02	0.01	6.94	2.41
6	Scrubland	76.93	26.74	62.70	21.79
7	Settlement	37.17	12.92	63.88	22.20
9	Vegetation/Plantation	2.30	0.80	7.50	2.61
10	Wasteland	7.34	2.55	17.75	6.17
11	Waterbodies	11.62	4.04	8.56	2.98
12	Wetlands/Marshy/Waterlogged area	4.98	1.73	2.12	0.74
	Total	287.70		287.70

and Figure 8a. Cropland and forest areas have both experienced significant reductions, with cropland decreasing from 84.53 km² (29.38%) to 62.80 km² (21.83%) and forest areas decreasing from 61.95 km² (21.53%) to 53.55 km² (18.61%). This decline in agricultural and forested areas is accompanied by a notable increase in settlement areas, which have grown from 37.17 km² (12.92%) to 63.88 km² (22.20%), reflecting urban expansion and possibly population growth. Additionally, the area dedicated to road networks has surged from a negligible 0.02 km² (0.01%) to 6.94 km² (2.41%), indicating significant infrastructure development. Similarly, playgrounds and stadiums have seen a substantial increase, from 0.12 km² (0.04%) to 1.01 km² (0.35%). While scrubland has decreased from 76.93 km² (26.74%) to 62.70 km² (21.79%), wasteland has increased markedly from 7.34 km² (2.55%) to 17.75 km² (6.17%), suggesting changes in land management or environmental degradation. The area covered by vegetation/plantation has also increased from 2.30 km² (0.80%) to 7.50 km² (2.61%). Meanwhile, waterbodies have decreased slightly from 11.62 km² (4.04%) to 8.56 km² (2.98%), and wetlands have also seen a reduction from 4.98 km² (1.73%) to 2.12 km² (0.74%), indicating potential impacts on local hydrology. These changes highlight significant shifts in land use priorities and environmental conditions over the two decades.

The Normalized Difference Vegetation Index (NDVI) map provides insights into vegetation density and health in the study area. The NDVI map plays a crucial role in assessing the extent of vegetative cover in the catchment area. Vegetation plays a significant role in mitigating flooding by absorbing and slowing down surface runoff, enhancing infiltration rates, and stabilizing soil. Higher NDVI values often correspond to areas with better water absorption capacities, which can reduce the volume and velocity of runoff during heavy rainfall events [7,66].

In Figure 9, distinct variations in NDVI values are observed, with the western parts of the region showing considerably higher values. This indicates dense and healthy vegetation cover in these areas, likely comprising forests, abundant croplands, or well-maintained green spaces. The NDVI values gradually decrease towards the eastern parts of the study area. This decrease could be attributed to factors such as land use changes, urbanization, or natural variations in vegetation density. Notably, the lowest NDVI values are observed in the settlement areas, which typically have limited vegetation due to infrastructure development and human activities. The correlation between NDVI and land cover types from the LULC map is significant. Areas with high NDVI values often correspond with forested regions or areas with dense vegetation cover, as confirmed by the LULC classification. On the other hand, regions with lower NDVI values align with categories like settlements or built-up areas, where vegetation cover is relatively sparse.

3.3. Geomorphology

The geomorphological condition of an area (Figure 10) plays a crucial role in identifying flood-prone regions, as it significantly influences the intensity and impact of flood events. Permeable rock formations, such as sedimentary rocks, allow for water percolation and groundwater infiltration, reducing surface runoff and the likelihood of flooding. On the other hand, crystalline rocks, which are impermeable, hinder water infiltration, thereby promoting surface flow [67,68]. This surface flow can exacerbate the severity of floods, as water is less likely to be absorbed into the ground. Backswamp and flood plains are naturally prone to flooding, alluvial soils are moderate to high flooding risks, shallow weathered pediplains have moderate flood risk due to less permeable soil, and lateritic soils are less prone to flooding due to their elevation and the fact that they are typically well drained. However, their impermeability can lead to surface runoff, contributing to flooding in lower lying areas.

3.4. Drainage Network Proximity

The drainage network proximity map (Figure 11) represents the distances of various areas from the nearest drainage channels. The map categorizes the proximity into four distinct zones: within 200 m, 500 m and 1000 m, and beyond 1000 m from the drainage network. Areas within 200 m are in close proximity to the drainage network, thus likely to experience higher exposure to water flow and potential flooding. The areas within 500 m may be moderately affected by drainage patterns. Those within 1000 m are less directly influenced by the drainage network but may still be impacted during heavy rainfall or flooding events. Areas beyond 1000 m are at the greatest distance from the drainage network, with minimal direct influence from drainage channels, making them the least likely to be affected by water-related issues [67].

3.5. Flood Vulnerability Analysis

Flow accumulation, slope, elevation, proximity to drainage networks, land use, rainfall intensity, and geology are key factors influencing flood risk. High flow accumulation in low-lying areas with dense drainage networks often leads to flooding, especially in flat regions with gentle slopes. Proximity to rivers and streams increases flood susceptibility, with areas within 200 m being at the highest risk. Land use also plays a crucial role, where urbanization and deforestation promote surface runoff, heightening flood potential. Intense rainfall exacerbates this risk, particularly in regions with impermeable geological formations that hinder infiltration [45,67].

3.5.1. GRID Analysis

The primary objective is to develop a comprehensive database that serves as a vital resource for local planners, decision makers, and emergency responders during crises. This database supports effective resource allocation for risk mitigation, precautionary measures, and strengthening community preparedness. The assessment integrates both primary and secondary data sources, utilizing a combination of quantitative and qualitative methods to evaluate hazard ratings within the study area. The vulnerability analysis focuses on the direct impacts on specific sites and the broader effects on the surrounding environment. It also considers strategies for mitigating these impacts and enhancing the capacity to manage future incidents.

Different rates are assigned to each class under various parameters, along with corresponding weights, to determine the most suitable rates and weights that yield optimal results [69,70,71]. After preparing all the parameters and their individual classes, user-defined rates (R) on a scale of 1 to 10 are assigned based on their significance or influence on flood risk. A rate of 10 indicates extreme vulnerability to flood risk, while a rate of 1 signifies almost no risk.

A 1 km × 1 km grid is generated using GIS, with each layer intersected to calculate the average weightage value for each grid [72]. The layers are assigned aggregate scores based on thematic significance and categorized into low, medium, and high vulnerability classes. Individual thematic maps are prepared, leading to a final hazard zonation that classifies the area into zones of high, medium, and low vulnerability [73]. The detailed methodology is illustrated in Figure 12.

3.5.2. Weightage of LULC

The interpreted LULC classes are assigned different ranks based on their vulnerability characteristics, with the detailed weightage information presented in

Table 5. Land Use/Land Cover weightage.

Land Use/Land Cover	Weightage
Wetland/marshy/waterlogged area	10
Settlement	9.5
Road network	9
Railway	8
Playground/Stadium	7
Wasteland	6
Scrubland	4
Cropland	4
Vegetation/Plantation	3
Forest	2
Waterbodies	0

[74].

Each interpreted polygon is assigned a rank according to the ranking table, and a corresponding LULC vulnerability map is generated, as illustrated in Figure 13.

3.5.3. Weightage of Geomorphology

The interpreted Geomorphologic features have been assigned different ranks based on their nature of vulnerability, and the weightage details are shown in

Table 6. Geomorphic weightage.

Geomorphology	Weightage	Geomorphology	Weightage
Abandoned channel (upper deltaic plain)	10	Intermontane valley/structural valley (small)	5
Abandoned channels (lower deltaic plain)	9	Shallow weathered/shallow buried pediplain	5
Back swamp (lower deltaic plain)	9	Valley fill/filled-in valley	5
Channel bar (lower deltaic plain)	9	Moderately weathered/moderately buried pediplain	4
Deltaic plain upper	9	Pediment/valley floor	4
Natural levee (lower deltaic plain)	9	Denudational hills (large)	2
Natural levee (upper deltaic plain)	8	Denudational hills (small)	2
Channel bar (flood plain)	8	Lateritic upland	2
Flood plain	8	Residual hill	2
Alluvial plain	7	Inselberg	1
Alluvial plain younger/lower	7

.

Based on the ranking table, each interpreted polygon has been assigned a rank and specified map geomorphology vulnerability has been prepared, as shown in Figure 14.

3.5.4. Weightage of Proximity

Proximity to the drainage network is crucial in vulnerability studies, as the distance from the drainage influences the exposure to hazards. However, other geomorphic factors also impact vulnerability when considering proximity (

Table 7. Proximity to drainage.

Proximity to Drainage	Weightage
200 m	9.5
500 m	6.5
1000 m	4.8
>1000 m	3.5

). For instance, settlements located on higher terrain, even if near a drainage system, are assigned lower to medium weightage due to their elevated position, which reduces their susceptibility [74].

Based on the ranking table, a buffer zone of 200, 500, 1000, and <1000 m from the major drainage has been carried out by using Arc Map and has been superimposed in Arc Map to analyze the flooding vulnerability, as depicted in Figure 15.

3.5.5. Weightage of Drainage Density

Drainage density is a critical factor in assessing the vulnerability of an area. Regions with higher drainage density facilitate smoother runoff, which is beneficial during events like floods, cyclones, or tsunamis, as they help prevent water stagnation and uncontrolled runoff. Conversely, areas with lower drainage density are more prone to severe issues, such as stagnant water, flooding, and difficulties in water management [67]. A higher drainage density leads to increased runoff generation, enabling the catchment area to efficiently collect and transport excess water. The drainage density map, prepared through spatial density analysis, assigns values based on density, which are then correlated with their respective cells. The weightage or ranking of drainage density is detailed in

Table 8. Drainage density weightage.

Drainage Density	Weightage
High (>10 km/km2):	9.2
Medium (5–10 km/km2):	6.5
Low (<5 km/km2):	2.8

, and the vulnerable areas, as determined by drainage density, are illustrated in Figure 16.

3.5.6. Weightage of Slope

In this analysis, grids with greater elevation are assigned lower values, while grids with lesser elevation receive higher weightage (

Table 9. Slope weightage.

Slope Angle	Weightage
Flat or near-flat slope (0–1.6°)	10
Gentle slope (1.7–5.1°)	4
Moderate slope (5.2–10°)	2
Steep slope (11–27°)	1

and Figure 17). This approach helps in assessing the vulnerability of the area, as lower elevations are typically more prone to flooding and other hazards [74].

3.5.7. Focus Group Discussion and Data Validation

Discussions are held with relevant stakeholders and 30 key informants are identified; interviews are conducted with them to validate the findings of the model.

3.6. Flood Vulnerable Zones (Integration Based on Weightage)

Finally, flood zonation analysis is conducted by integrating all the vulnerability layers and calculating the average values for each grid [70,71]. These values are then translated into qualitative categories, such as very high, medium, and low vulnerability, to provide a clear overview of flood risks across the entire study area (Figure 18).

Flood Vulnerable Villages

Based on the flood vulnerability analysis, ninety villages (depicted in

Table 10. List of flood-vulnerable villages.

Vulnerability	District	Block	Gram Panchayat	Villages
High	Cuttack	Barang	Belagachhia	Naranpur
			Dadhapatana	Panchupal
			Kunheipada	Mundamuhan
				Brahmangan
				Patapur
				Fakirpara
				Kunheipara
			Madhupur	Madhupur
			Ramdaspur	Padmalava nagar
				Ratagaralenka Sahi
			Sribantapur	Sribantapur
		Cuttack Sadar	Town	Town
	Khorda	Bhubaneswar	Barimunda	Barimunda
				Rokat
			Dadha	Marchia
			Kalarahanga	Kalarahanga
			Kalyanpur	Kalyanpur
			Paikerapur	Ghangapatna
			Town	Town
				Town
Low	Cuttack	Banki-Dampara	Banara	Banara
			Bhagipur	Bhagipur
				Gayala Bank
			Forest	Forest
		Barang	Dadhapatana	Madhusudanpur
			Mundali	Chakradharpur
			Narajmarthapur	Nuagan
	Khorda	Bhubaneswar	Andharua	Andharua
				Dasapur
			Chandaka	Similipatana
				Anlapatana
			Dadha	Kantunia
			Kalyanpur	Khairapada
			Kantabad	Dalua
				Bhola
				Kantabad
				Bhagabatipur
			Mendhasal	Giringaput
			Paikerapur	Gothapatna
				Malipada
				Nuagan
				Managobindapur
			Town	Town
				Town
				Town
		Jatani	Madanpur	Deuliapatna
Medium	Cuttack	Barang	Belagachhia	Belagachhia
			Dadhapatana	Dadhapatana
			Madhupur	Madhubana
				Chandiprasad
			Mundali	Mundali
			Narajmarthapur	Narajmarthapur
				Talagar
			Ramdaspur	Arilo
				Govindpur
				Ramadaspur
			Sribantapur	Bachhapur
		Cuttack Sadar	Town	Gopalpur
				Belgochhia
				Shrikorua
				Shrikorua
				Shrikorua
	Khorda	Bhubaneswar	Andharua	Jagannathprasad
			Barimunda	Tangibanta
				Singada
				Krushnasaranpur
			Chandaka	Chandaka Jogisahi
				Sundarpur
				Kujimahal
			Dadha	Jhinkardiha
				Dadha
				Padasahi
				Balipada
			Darutheng	Bhalunka
				Chudanga
				Krushnanagar
				Bhuasuni
				Jujhagada
				Tulasadeipur
				Daruthenga
				Shyamsundarpur
			Kalarahanga	Injana
			Kalyanpur	Gandarpur
				Dhaua
				Ostapada
				Nuabanta
			Raghunathpur	Raghunathpur
				Raghunathpurjali
			Town	Town
				Town
	6	10	45	90

) within the catchment area, which fall within the Cuttack and Khorda districts, are categorized into high, medium, and low vulnerability zones. The insights are further validated through focus group discussions and key informant interviews, as the required data for the model are not readily available for the fringe areas. These discussions address data gaps by integrating local knowledge and expertise, ensuring that the model’s findings are aligned with the actual conditions in these regions.

Twenty villages are identified in the high-risk category, requiring significant attention. These include Naranpur, Panchupal, and Sribantapur in Cuttack district and Barimunda, Marchia, Kalarahanga, and Ghangapatna in Khorda district, as they are prone to severe flooding events. Forty-four villages fall into the medium vulnerability category, including Belagachhia, Dadhapatana, and Ramdaspur in Cuttack district, along with Jagannathprasad, Tangibanta, Raghunathpur, and Injana in Khorda district, indicating a moderate level of flood risk.

Twenty-six villages, including Banara, Bhagipur, and Nuagan in Cuttack district and Similipatana, Anlapatana, and Giringaput in Khorda district, fall under the low vulnerability category. While these areas are less susceptible to flooding, they still require monitoring and preparedness measures.

4. Discussion

The analysis of flood vulnerability in North Bhubaneswar has provided valuable insights into the interplay between land use, vegetation, geomorphology, drainage networks, and other factors influencing flood risks. The findings underscore the significant transformations in Land Use/Land Cover (LULC) over the past two decades and highlight the implications for flood susceptibility in the region. The transition in LULC from 2004 to 2024 reveals a marked decline in cropland and forest areas, accompanied by an increase in settlements and infrastructure [75]. Specifically, cropland has decreased by 21.7% and forest areas by 13.6%, while settlement areas have expanded by 71.8%, and road networks have increased significantly. These changes indicate a shift towards urbanization and infrastructure development, which often results in increased impervious surfaces. Impervious surfaces contribute to higher runoff and reduced groundwater recharge, exacerbating flood risks. The increase in wasteland and decrease in waterbodies and wetlands further complicate the hydrological balance. Waterbodies and wetlands play crucial roles in flood mitigation by absorbing excess water and reducing runoff. Their reduction could thus increase flood vulnerability in the area [76].

The geomorphological conditions of the study area significantly influence flood susceptibility. The presence of impermeable crystalline rocks and shallow weathered pediplains contributes to higher surface runoff, increasing flood risks. Conversely, areas with permeable sedimentary rocks or lateritic soils tend to experience lower flood risks due to better water infiltration [49]. The proximity to drainage networks is another critical factor affecting flood vulnerability. Areas within 200 m of drainage channels are at the highest risk of flooding due to their direct exposure to water flow. Conversely, regions beyond 1000 m from drainage networks are less prone to immediate flooding but may still experience issues during severe rainfall events [67].

The GRID-based vulnerability analysis integrates various factors such as flow accumulation, slope, elevation, proximity to drainage networks, land use, and geomorphology. By assigning weightage to each factor and analyzing the combined effects, this study categorized regions into high, medium, and low vulnerability zones [70,71]. In total, 90 villages within the catchment area have been classified based on their vulnerability levels, with 20 villages identified as high-risk areas requiring urgent attention, 44 classified as medium vulnerability, and 26 as low vulnerability. Naranpur, Panchupal, and Sribantapur in Cuttack district and Barimunda, Marchia, Kalarahanga, and Ghangapatna in Khorda district are highlighted as high-risk due to their susceptibility to severe flooding events.

5. Conclusions and Recommendations

Unfortunately, development issues in peri-urban areas are neither seriously addressed by urban local bodies nor by rural local bodies. Moreover, the databases for these areas lack consistency and availability, creating a disconnect between ongoing development and the sustainability of these transitioning areas. This study provides a comprehensive analysis of flood vulnerability in North Bhubaneswar, highlighting several critical factors that contribute to the region’s flood risks. Over the past two decades, significant changes in Land Use/Land Cover, including reductions in cropland and forest areas and increases in settlements and infrastructure, have altered the region’s hydrological dynamics. The expansion of impervious surfaces due to urbanization has led to higher runoff and decreased groundwater recharge, exacerbating flood risks. The NDVI analysis reveals that regions with high vegetation cover, particularly in the western parts of the study area, are better equipped to manage runoff and mitigate flooding. Proximity to drainage networks has proven crucial in assessing flood risk, with areas close to drainage channels being more prone to flooding. The GRID-based vulnerability analysis integrates various factors such as land use, geomorphology, slope, and proximity to drainage networks to provide a detailed flood risk assessment [70,72]. This analysis has identified specific villages in the catchment as high-risk areas, necessitating targeted flood management strategies.

Recommendations include enhanced land use planning, vegetation and erosion control, and policy and planning integration. Establishing monitoring systems to track changes in land use, vegetation cover, and flood risks is crucial. These data should be used to inform adaptive management strategies. Further research into flood risk prevention techniques and technologies, along with the use of advanced modeling techniques, will improve predictions of land use transformations and flood events. Additionally, strategies like low impact development, sustainable urban drainage systems, and water-sensitive urban design should be integrated from the initial development phases. Integrating agricultural activities with planning, implementing effective land use planning in fringe areas, and ensuring efficient planning governance will contribute to sustainable development in peri-urban areas [77]. This proactive approach to flood prevention will effectively address water management challenges and provide socio-economic benefits in these regions.

Understanding the socio-economic status of the affected communities, their coping mechanisms, and access to resources can offer critical insights into how floods impact different population groups. This socio-economic analysis could help tailor flood mitigation strategies to meet the specific needs of these communities. Additionally, including an analysis of groundwater levels, recharge rates, and their interaction with surface water could provide a more comprehensive understanding of the hydrological processes that influence flood vulnerability. Furthermore, extending this study to other fringe regions prone to flooding could yield valuable insights and support the development of comprehensive and adaptive development plans for peri-urban areas on a broader scale.