Assessment of flood vulnerability in a coastal metropolitan city for sustainable environmental using Machine Learning methods

Assessment of flood vulnerability in a coastal metropolitan city for sustainable environmental using Machine Learning methods | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Assessment of flood vulnerability in a coastal metropolitan city for sustainable environmental using Machine Learning methods Nivedita V, Sabarunisha Begum S, Sakthi U, Sellam V, Navaneethan C, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4194276/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Floods are very destructive natural catastrophes that significantly harm people, the environment, and structures. Due to their dynamic nature, flash flood-prone locations are difficult to predict. We used cutting-edge machine learning algorithms for early flash flood detection to overcome this. To evaluate flood vulnerability in Chennai, Tamil Nadu, India, we employed and evaluated two combined ensemble models: the Artificial Neural Network (ANN) and Random Forest (RF). To put these models into practice, we built a GIS framework out of 280 historical flooding sites and twelve flood-related parameters. Using information gain ratio and multicollinearity diagnostic tests, we evaluated the association between flood occurrences and pertinent factors. Statistical criteria like "Freidman" were used to compare the prediction abilities of ANN and RF models. Both ANN and RF models performed better than expected when simulating flood susceptibility. In order to reduce flood-related hazards and create efficient mitigation plans, state and local authorities, as well as policymakers, will benefit from the study's findings and methodology. Flood susceptibility percentages were 18% very low, 16% low, 13% moderate, 22% high, and 31% extremely very high, according to the research of ANN. While this was going on, RF revealed that the likelihood of flooding was 7% very low, 11% low, 34% moderate, 31% high, and 18% extremely very high. Maintaining a strong drainage system, using regulated building techniques in sensitive regions, setting up early warning systems, creating resilient infrastructure, and educating people are all necessary to reduce floods in Chennai. For resource allocation, disaster response, and readiness exercises, effective coordination between government agencies, non-governmental organizations (NGOs), and local populations is essential. Chennai's resistance to floods depends on a multi-pronged approach that includes infrastructure improvement, pro-active planning, and public awareness. Flood Environment ANN Model RF Model Algorithms Machine Learning Vulnerability Risk Assessment Mitigation Plans Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction A flood is a type of natural disaster that happens when an area of typically dry land is submerged by a large volume of water (Douglas et al, 2000). Heavy rainfall, quick snowmelt, dam breaches, or overflowing rivers, lakes, and seas are a few of the possible causes of this phenomena (Hong et al, 2019). Floods may have a catastrophic impact on the environment and populated areas, frequently causing extensive damage and fatalities (Kuncheva et al, 2010). The earth may become saturated when heavy rain falls quickly, overburdening drainage systems and causing surface runoff (Luo et al, 2019). Water levels in rivers and streams rise quickly as a result of this extra water flowing into them (Kuriqi et al, 2018). This might cause floods in locations with insufficient infrastructure to handle the extra water (Masks et al, 2019). Similar to how snowmelt during the spring thaw may boost water flow into rivers and cause floods (Hosni et al, 2019). Depending on their severity, floods can be categorized into a variety of groups. Intense and rapid flash floods can happen minutes or hours after significant rainfall. Due to their sudden beginning and potential to catch individuals off guard, they constitute a serious hazard (Ma et al, 2020). On the other hand, when rivers gradually rise and overrun their banks, river floods occur over a longer period of time (Islam et al, 2021). Sustainability involves satisfying current requirements while preserving the capacity of future generations to fulfill their own. It encompasses the conscientious utilization of resources, the advocacy of environmental well-being, social fairness, and economic prosperity. Sustainable approaches strive to harmonize ecological, societal, and economic aspects for enduring welfare and resilience. Sustainable flood management practices can benefit agricultural land by replenishing soil nutrients, improving water availability, and enhancing crop growth. Proper flood control measures, such as the construction of sustainable drainage systems, can mitigate soil erosion, promote groundwater recharge, and foster ecosystem resilience, supporting long-term agricultural sustainability. When ocean waters surge inland, frequently as a result of powerful storms or tsunamis, coastal floods happen (Nikolaos et al, 2019). A flood may have a significant negative effect on local residents and the environment. Floodwaters can cause economic losses by eroding the soil, destroying infrastructure, and destroying crops (Duque et al, 2020). Homes, businesses, and public buildings may sustain damage or destruction, forcing people to relocate and upsetting everyday life (Oeurng et al, 2011). Floods can result in substantial human suffering, including fatalities and serious injuries, in densely populated places (Hou et al, 2020). Flood impact reduction initiatives frequently include planning, response, and recovery techniques (Pradhan et al, 2010). Zoning for floodplains and building codes assist control where and how development happens in flood-prone regions (Akter et al, 2019). Communities can efficiently prepare for and respond to floods thanks to early warning systems and emergency plans (Pravalie et al, 2013). Levee construction and dam construction are examples of infrastructure upgrades that can assist manage and steer flooding (Falah et al, 2019). In certain areas, the frequency and severity of floods have increased in recent years due to the consequences of climate change (Rumelhart et al, 1986). Higher sea levels are a result of glaciers and polar ice melting due to rising global temperatures (Sahana et al, 2019). The susceptibility of many areas to flooding occurrences has risen as a result of this and increasingly intense weather patterns (Alexander et al, 2019). By replicating the dynamics and impacts of floods, flood modelling helps forecast and lessen their effects (Torcivia et al, 2020). It influences infrastructure development, disaster preparedness, and urban planning (Uthayakumar et al, 2020). Models calculate the possible flood extent, depth, and velocity by examining variables including rainfall, geography, and land use, enabling a well-informed risk assessment (Fang et al, 2020). These models support the development of floodplain management policies, resilient building design, and the optimization of evacuation plans (Bui et al, 2019). They make it easier to make decisions for catastrophe preparation and response, reducing property loss and protecting lives (Ruiz et al, 2008). For the development of models that are susceptible to flooding, many academics have used a variety of techniques (Kumar et al, 2019). Since there are many different types of flood-susceptible models, they can be employed in many different research, but in this research, we used an artificial neural network and random forest (Getahun et al, 2015). Hybrid machine learning techniques are being implemented and developed for several natural hazard models as a result of their great results (Kourgialas et al, 2011). On the optimum technique for simulating different types of natural hazards, such as landslides or flood susceptibility, there was, however, no general agreement. Researchers advise creating and experimenting with novel techniques for modelling natural disasters and mapping flood susceptibility (Cao et al, 2019). In order to simulate flood susceptibility in the Chennai metropolitan area, we also created ensemble machine learning methods including Artificial Neural Network and Random Forest models (Ghasemain et al, 2020). Few applications of the hybrid ensembles of the Random Forest (RF) and Artificial Neural Network (ANN) models for simulating natural disasters have been made (Chen et al, 2014). In this study, we selected the Chennai area, known for experiencing frequent flash floods, as their research case to develop a flash flood susceptibility map (Hong et al., 2019). The study's novelty can be attributed to two main aspects: (i) The production of a more comprehensive flood susceptibility model for the specific study area; and (ii) The introduction of a novel hybrid ensemble model called Random Forest (RF) combined with Artificial Neural Networks (ANN) to model flood susceptibility maps, a method never used before in in this study area. The contribution of this research involves utilizing both traditional models (RF and ANN) and ensemble models to accurately map flood susceptibility and identify flood hazards in the Chennai area (Deng et al., 2020). This innovative approach aims to improve flood prediction and mitigation strategies in the region. The study's findings will help regional and local authorities and policymakers reduce the risks associated with floods and establish effective mitigation strategies to prevent possible harm. Study Area The city of Chennai and the surrounding urban and suburban areas are all included in the Chennai Metropolitan Area (CMA). It is a prominent commercial and cultural center in southern India, distinguished by a variety of businesses, institutions of higher learning, and historical sites. The CMA attracts both domestic and foreign investors and is essential to regional trade, technology, and services. However, the effects of increased urbanization include infrastructural needs, environmental degradation, and traffic congestion. For the Chennai Metropolitan Area to continue growing and prospering, it is imperative to solve these concerns through effective urban planning, transportation management, and sustainable development. Due to its geological history, the Chennai Metropolitan Area is primarily distinguished by a mix of soil types and lithology. A mixture of sedimentary, igneous, and metamorphic rocks makes up the region's geology. Alluvial soils, red soils, black soils, and laterite soils are examples of typical soil types. Red and black soils are common in the hinterlands, whereas alluvial soils are found in riverbanks and along the shore. Major river flows in study area which include kosasthaliayar river, Coocum river, Adyar river and Buckingham canal. Waterlogging occurs when inadequately designed drainage systems are unable to manage significant monsoon rains. Water body encroachment and inappropriate trash disposal make the issue worse. The regular floods are also a result of insufficient upkeep of the current infrastructure and ineffective floodplain management. Comprehensive urban planning, better drainage systems, the restoration of water bodies, public awareness campaigns, and sustainable development practices are necessary to minimize and manage the recurrent flooding difficulties in the Chennai Metropolitan Area in order to meet these challenges. In this study, flood susceptibility models were used. These are the reasons listed above for taking severe measures for proper management plans in the Chennai Metropolitan area using the flood susceptibility analysis by using machine learning along with models like ANN and RF. Material and methods According to the methodological flow chart in Fig. 2 , the current study used the flood inventory map, the development of flood conditioning factors, the evaluation of the flood conditioning factors using information gain ratio and multi-collinearity test, and flood susceptibility models using machine learning techniques (ANN and RF) (Nikolaos et al., 2019). The groundwork of a flood inventory map for the study region is the first stage in creating a map of flood vulnerability. The usual state at the flooded areas was achieved and investigated using the readily available GIS technology. The chance of a flood in the near future will rely on the likelihood of a flood in the recent past, making this a crucial stage in the flood prediction process. The development of the sensitive flood model is frequently extremely precise and difficult due to the requirement for numerous geographic topographical and hydrological parameters. Determining the causes of the flood is therefore crucial, and the systematically designated factors will confirm the correctness of the maps of flood susceptibility. The flood susceptibility literature that is currently available, twelve flood-influencing factors, including elevation, lithology, slope, aspect, topographic wetness index (TWI), topographic roughness index (TRI), sediment transport index (STI), stream power index (SPI), land use/land cover (Lu/Lc), distance to the river, soil type, and rainfall, were selected for the current study area. The conversion of all affecting factors into a raster format with a spatial resolution of 30m was done. Topographic factors must be taken into account while modelling flood studies since they have a direct and indirect impact on the hydrological features of the study region. First, a Digital Elevation Model (DEM) for the research basin was produced in the ArcGIS 10.8 environment using the ASTER GDEM (Version 2). The production of Digital Elevation Models (DEMs) involves the collection of elevation data using various methods, such as airborne LiDAR or satellite-based radar. These data are processed to create a comprehensive representation of the Earth's surface, enabling the visualization and analysis of terrain features for applications in cartography, engineering, and environmental studies. From DEM topographic parameters in the ArcGIS environment, slope, lithology, aspect, TWI, SPI, STI, and TRI have all been produced. Elevation: Flooding and height are inversely associated; the higher the elevation, the lesser likely flooding will occur, and vice versa. It is possible to derive Chennai Metropolitan Areas elevation information using DEM data. Slope: Another crucial factor that influences a flood is the slope, it controls the speed of the water's flow. The likelihood of water stagnation is reduced, infiltration is reduced, and flow velocity is increased with increasing slope angle. The slope for the research area is generated using the Arc toolbox viva spatial analyst tool, and the surface slope is generated using the Arc GIS software's Slope tool. Input data is DEM to generate slope as output. Aspect: Another component, aspect, determines the directions that flooded water moves in addition to maintaining soil humidity. The aspect thus has an indirect impact on flooding. Consider the section of a slope that is shaded, where the soil has a high relative humidity and there is significant runoff. Aspect tool in the surface tool subsection is used to generate aspects such as slope-wise aspects. Rainfall: One of the key elements that influence the likelihood of flooding has been identified as rainfall. because flooding may result after a brief period of heavy rain. We employed IDW interpolation and rainfall data from four Chennai meteorological stations to produce rainfall maps in the ArcGIS 10.8 environment. We interpolated using the IDW approach because we only had data for four locations, despite the fact that this method is strongly advised when there is a very little amount of data. TRI: The TRI is the most important factors impelling flood events. It is based on the neighborhood's topography in the research area. The chance of a flood increasing with decreasing TRI values. By using Focal statistical tool to create Minimum, Maximum and Mean raster file with input data of DEM. To generate the TRI for the study area followed by the equation in the raster calculator tool, \(\text{T}\text{R}\text{I}=\text{M}\text{e}\text{a}\text{n}-\text{M}\text{i}\text{n}\text{i}\text{m}\text{u}\text{m} / \text{M}\text{a}\text{x}\text{i}\text{m}\text{u}\text{m}-\text{M}\text{i}\text{n}\text{i}\text{m}\text{u}\text{m}\) ………………….1 TWI: It expresses the alteration in wetness of a basin spatially, is a significant determinant in the likelihood of flood. This index displays the amount of water present in each individual pixel in the area. Using the following equation, TWI are \(TWI =\frac{In \left(As\right)}{\text{t}\text{a}\text{n}{\beta }}\) …………………………………………….2 As and, respectively, represent the explicit catchment area (m 2 m − 1 ) and slope gradient (in degrees). In general, there is a direct correlation between floods and high TWI readings. SPI: It has a substantial effect on the fluvial system. To determine the SPI, use the equation below, \(SPI =As tan\beta\) …………………………………………3 Whereas, As is the slope gradient stand in for the particular catchment region, and are denoted by (radians). Total SPI is the term used to describe both the bed's erodibility and its ability to transport sediment. STI: Another factor that might cause flooding is the STI, which can increase the frequency of flooding and cause damage to foundations. The following equation is used to derive the STI from the DEM are, \(STI=\left({\frac{AS}{22.13})}^{2}* \right({\frac{sin\beta }{0.0896})}^{2}\) …………………………….4 where the letter A stands in for the area upstream and the symbol designates each pixel of the slope. The hydro-climatic and geomorphologic parameters of the basin region are used to calculate the STI. As sediment is deposited, the channel's bed shifts, reducing the channel's capacity to hold water and leading to flooding. Land use/Land cover: Flood frequency is directly impacted by LULC because it affects sediment transportation and surface runoff. Because the formation and infiltration of surface runoff are directly controlled by the LULC. As a result of these areas' inability to produce surface water and allow for water infiltration, flooding occurs more frequently there. The woodland region, on the other hand, provides for water infiltration that lessens flooding. When comparing hydrological responses at different temporal scales, the connection between flood episodes and vegetation density is inverse. In the present study, a Land Use Land Cover (Lu/Lc) map was created using Landsat and OLI (Operational Land Imager) satellite images. The Artificial Neural Network (ANN) technique was employed for this purpose, and the analysis was conducted using ENVI software (version 5.3) with a spatial resolution of 30 meters. This approach allowed for accurate classification and mapping of different land cover types in the study area, providing valuable information for various environmental and land management applications. The LULC map was divided into five categories: agricultural land, waste land, urban area, grassland, and aquatic body. Distance to the river: The majority of flooded places are typically found close to rivers. Because the distance from the river affects the likelihood of flooding and the ratio of river flow to river, it is a crucial determining element in determining the research area's flood-prone areas. The likelihood of flood events decreases with increasing distance from the river. Flooding is connected to the loading of terrestrial water at the local level. In the current study, we created a map showing the distance to the river using Google Earth Pro, converted the KML file into a shapefile in the Arc GIS environment, and created a buffer for the main river to calculate the distance from the river. Soil: One of the major determining factors that impacts how rainfall-runoff functions is soil. While other factors, like the local climate and the erosion process, also have an impact on how rainfall-runoff forms, the soil possessions sprightly control water penetration. The higher the rate of soil infiltration, the less frequently flooding happens. The National Bureau of Soil Survey provided the soil map that was utilised to digitise the soil for this investigation. Lithology: The study of rock properties known as lithology affects flood behaviour. Sand and other porous lithologies reduce flood risk by absorbing water. Clay-like impermeable rocks can cause surface runoff, which heightens floods. In order to better control floods, flood extent can be predicted and effective drainage systems can be planned by understanding lithology. Lithology map was prepared by using Geological Survey of India lithology map. Method for flood influencing factors using Information gain ratio and multicollinearity test It is essential to evaluate the importance of the flood affecting parameter or the probability for flooding before beginning the model's training sections. Based on each parameter's statistical traits and connection to the floods, its relative importance has been determined. The Information Gain Ratio (InGR) approach has been used to determine the influential factors for FSM prediction. An InGR value is assigned to each influencing element in order to quantify its significance. Higher InGR values are indicative of more pertinent influencing elements. The decision to use the InGR model in the current experiment was based on its simplicity and effectiveness. The InGR model is well-suited for the research objectives and provides valuable insights into the relevant influencing factors, making it an appropriate choice for the study. It is computed using the following equation are, \(Gain Raito \left(x, Z\right)=\frac{Entropy \left(Z\right)- {\sum }_{1}^{n}{\sum }_{i=1}^{n}\frac{Zi}{z} Entropy \left(zi\right)}{-\sum _{i=1}^{n}\frac{Zi}{Z}\text{log}\frac{Zi}{Z}}\) …………………5 If the property x originates from training point Z with subsets Zi1 = 1, 2, 3, etc. Using a range of multi-collinearity tests, such as variance decomposition proportions, conditional index, VIF, and tolerances, influencing factors have been evaluated for all probability models. We utilised the Pearson's relationship coefficient and the VIF to determine the respective weights of the twelve flood training factors in this investigation. The VIF > 9 and incredibly weak correlation serve as indicators of the problem of multicollinearity in the components. Therefore, if the conditioning factor's VIF value is more than 9, it is highly recommended to leave it out of the model. Food susceptibility modelling-ANN The three-layer ANN model (input, hidden, and output layers) utilised in the current work using a Back-Propagation (BP) and error correction learning method has been effectively utilised in the flood susceptibility modelling. The input layers and ten hidden nodes in this investigation were configured with the same numbers as the critical parameters (Table 1 ). The output layer, on the other hand, uses a single node and is coded as 1 for flood occurrences and 0 for non-flood events. Although there are other techniques for training ANN models, BP is the most often used one. Thus, the BP based ANN model has been used to estimate the nonlinear connection between the essential parameters and the flood occurrences. First, BP selects the starting weights at random. There has been a comparison of calculated and observed values. Errors are defined as discrepancies between calculated values and observed values. Several error measurement approaches, including mean squared error (MSE) and root mean square error (RMSE), have been used to analyse it. The initial weights are adjusted based on the generalised delta rule to distribute the entire error across the network's neurons. This method is iterated when the degree of mistake is at a reduced degree. Random Forest The groundbreaking Random Forest (RF) approach combines classification and regression decision trees to make accurate predictions. It is a popular ensemble-learning method. Ho's concept of "random selection features" and Breiman's "random subspace" are essential components of the RF, which can be divided into two subgroups. The Random Subspace is an ensemble machine learning technique that generates multiple classifiers in order to boost the prediction accuracy of a subpar classifier right from the beginning. To predict the data classification, the RF performs numerous regression tree training stages and generates diverse sets of samples via sampling with replacement. The final classification chosen by the RF is based on the voting outcomes of several classifiers, ensuring a significant number of votes from each tree in the forest. During the regression tree's training phase, the observation datasets are categorized using rules based on response parameters until the prediction achieves the lowest possible node deviation. One of the major drawbacks of regression trees is their tendency to overfit the training data, resulting in poor performance when faced with an unknown dataset. However, Random Forest (RF) can help address this issue. In the RF algorithm, during the training of each regression tree, a random subset of input archives and predictor factors is chosen as input. Through various sampling approaches, different sets of regression trees are generated, each trained on a different randomly chosen subset. Using a total sample to train the decision trees is not recommended, as it disregards the importance of local samples. Therefore, RF employs a more robust approach by creating diverse subsets for training individual trees. In flood susceptibility analysis, the RF model serves as a benchmark for comparing outcomes with those of a new hybrid model, highlighting its usefulness in such applications. Table 1 The parameters of the machine learning algorithm used for flood susceptibility modelling S.No Model name Description of parameters 1. ANN Hidden layer-7, learning rate-0.38, momentum-0.25, seed-3, training time-750, validation threshold-15, Normal to binary filter-TRUE 2. Random forest Batch size-100, seed-4, number of iteration-175, max depth-2, calc out of bag-TRUE, Compute attribute importance-TRUE Results Factors for modelling flood susceptibility For analysis of flood susceptibility in Chennai Metropolitan area by using machine learning models which include several factors which are elevation, slope, aspect, rainfall, TRI, TWI, SPI, STI, LULC, distance from the river, soil and lithology are prepared for the major input for training sets factors for flood susceptibility models (Vafakhah et al, 2020; Wu et al, 2020; Xie et al, 2019). Elevation: Elevation for Chennai metropolitan area are under the value of low from surface of the land ranges are − 26 to high rate of 163 (m), entire study are falls under the zone of low and medium due to coastal zone and western side part of study area cover most high range of elevation part. Due to the huge coverage of low elevation is favourable for flood susceptibility most occur near the coast surrounding areas and river bank side and high elevation is not so favourable. So he estimate of flood risk must take elevation into a major account, Low-lying places are at risk because floods can occur when water levels rise during periods of heavy rain. The Fig. 3 A shows the elevation map for Chennai metropolitan area. Slope: Slope for Chennai metropolitan area are under the slope ranges from 39.42 to 0, most of the region covers by low slope category because of Nearby coastal zone and some of the zone shows high slope value in the part of study area, 0 slope values represents water body. As like as elevation slope also plays a major role for flood analysis, because nearly and very low slope prefers flood vulnerable due to insufficient of water flow velocity is very low. This causes water to quickly build up, which inundates low-lying areas and causes floods. The Fig. 3 B shows the slope map for Chennai metropolitan area. Aspect: Aspect for Chennai metropolitan area generate from flat − 1 followed by north 0–22.5, northeast 22.5–67.5, east 67.5–112.5, southeast 112.5–157.5, south 157.5–202.5, southwest 202.5–247.5, west 247.5–292.5 and northwest shows 292.5–360. Over all study area facing towards eastern direction due to flow river ends in ocean and some part of study area shows some other direction too. Riverbanks and drainage systems are overloaded by heavy rain, which causes water to quickly accumulate. Deforestation and urbanisation together worsen runoff, lowering natural absorption. Rainfall frequency and intensity increase due to climate change, which also intensifies weather patterns. In low-lying places, these elements combine to produce destructive flooding. These can be identity the flow direction of above mentioned by using aspect analysis. The Fig. 3 C shows the aspect map for Chennai metropolitan area. Rainfall: Rainfall for Chennai metropolitan area are under the rainfall ranges from 249 to 544 mm, most of the region covers by high and very high rainfall pattern due to coastal processes, very low annual rainfall pattern receiving from north side and south west direction zone. Low-lying areas are flooded by excessive rain, which also overwhelms drainage systems and swells waterways. Flooding occurs when water builds up quickly, engulfing crops, buildings, and roadways. Rescue operations and evacuations that follow emphasise the devastation caused by natural flood. The Fig. 3 D shows the Rainfall map for Chennai metropolitan area. TRI: Chennai metropolitan area are under the roughness surface ranges from 0.1111 to 0.8888, most of the region covers by low and medium roughness terrain due to coastal zone, very low roughness surface shows from water body side. Due to the heavy rainfall receiving Chennai metropolitan area, as high topographic roughness index implies uneven terrain, which hinders water movement. Flooding is brought on by water accumulating in depressions and overtaxing drainage systems. Sharper water velocity and a higher risk of flooding are caused by steeper slopes, which intensify runoff. The Fig. 4 A shows the TRI map for Chennai metropolitan area. TWI: Chennai metropolitan area is under the wetness surface ranges from − 7.7246 to 11.5742, most of the region covers by low wetness terrain due to huge amount of urbanization around the study area, very low wetness surface shows from water body side. It evaluates landscape features that influence water accumulation and might cause flooding when high values indicate areas vulnerable to water accumulation. Due to inadequate drainage systems and elevated soil saturation, these areas are more susceptible to excess water during periods of heavy precipitation, raising the danger of flooding. The Fig. 4 B shows the TWI map for Chennai metropolitan area. SPI: Chennai metropolitan area are falls under very low stream power index zone, but in this study river part showing maximum amount of SPI at the rate of 14.5003. and other parts of study shows very low range up to 0. Increased water flow and sediment transport are indicated by high stream power index values, overwhelming natural channels. Flooding results from excessive rainfall or quick snowmelt. Deposition and erosion change the landscape and increase the chance of infrastructure damage. The Fig. 4 C shows the SPI map for Chennai metropolitan area. STI: Chennai metropolitan area are cover very low amount of sediments transportations, along the side of water bodies shows high and very high sediments transportations zones. STI falls under the ranges of 0 to 339.016. When there has been a lot of rain, rivers that have a high Sediment Transport Index, which indicates increased sediment movement, may become blocked and have less channel capacity. By reducing water flow and increasing the possibility of overflow and flooding in downstream areas, this increases the risk of flooding. The Fig. 4 D shows the STI map for Chennai metropolitan area. Land use/Land cover: Over all study area are been analysis the Landsat 8 OLI/TIRS data consists of 30m resolution and though Artificial neural network and classified their output in the form NRSC level 1 classification and shows 5 major classes which includes like, Water Body, Agriculture Land, Barren land, Built-up land and grass land. On the study area covers mostly covered by built-up land. Rapid urbanization due to the modification of others lands into built up land resulting more vulnerable zones for flood susceptibility zonation. Urbanisation and deforestation, for example, can disturb natural drainage patterns, lowering water absorption and raising runoff. These Land Use and Land Cover (LULC) changes. By clogging up local rivers and creating fast water collection during periods of high rainfall, this change to natural landscapes can raise the danger of flooding. Figure 5 A shows the land use and land cover map for the study area. Distance to the river: In the study area shows major river flow continuously in the boundary, buffer zone is considered in order of 1000m interval from the major river flow, 5 major classes are classified buffer zones are distance from the road shows nearby river zone shows most vulnerable for flood susceptibility zone. Far distance from the river zone shows very less vulnerable zones. North and southwest zones are along falls under the above 5000m distance from the river side. Flooding is considerably exacerbated by river proximity. The river's water level rises during periods of intense precipitation, overflowing its banks and flooding neighbouring areas. Infrastructure and human populations nearby are more at risk. Figure 5 B shows the distance from the river map for the study area. Lithology: Chennai metropolitan area falls under the lithology class are boulders beds, charnockite, fluvial, garnet, marine, pyroxene granulite, quartz – conglomerate shingles and shale with limestone. Over all the study area cover by the lithology features are marine and fluvial features. Some of the zones covers by charnockite and shale with limestone. When there is insufficient lithology, such as impermeable rock strata, there may be surface runoff during periods of excessive rainfall. This runoff overwhelms drainage systems, limiting adequate groundwater penetration and resulting in floods. Figure 5 C shows the lithology map for the study area. Soil: Chennai metropolitan area falls under the soil class are calcareous clay soil, calcareous cracking clay soil, cracking clay soil, gravelly clay soil and sandy soil. Over all the study area cover by the soil features are cracking clay soil and gravelly clay soil. Some of the zones covers by calcareous clay soil. Because compacted soil is less capable of absorbing water, heavy rains increase surface runoff. Because the runoff overwhelms drainage systems, water builds up and travels overland rather than penetrating the impervious soil, resulting in floods. Figure 5 D shows the soil map for the study area. Factors for flood susceptibility by InGR and multicollinearity tests Using a 12-fold cross-validation procedure, the analyses' results are given in Fig. 6 , where each parameter's InGR values were calculated. The most significant impact factors, as shown by the InGR data, are the high InGR value LULC (0.48), DR (0.54), elevation (0.61), and slope (0.58). The total roughness indicator TRI (0.45), the silt transport index STI (0.42), and the stream power index SPI (0.44) all demonstrate significant importance. The total wetness index (TWI), rainfall (0.52), lithology (0.35), and soil (0.38), all show little significance. The fact that the aspect factor has a value of InGR = 0.00, showing that it has minimal bearing on FSM prediction, is crucial. The results of the multi-collinearity test of the flood conditioning components and how it affects the capacity to foresee FSMs are shown in Table 2 . The Pearson's correlation coefficient for the association between floods and their potential sources is shown in Table 3 . Table 2 presents the results of a multi-collinearity test (VIF) examination of the flood conditioning components and influences the ability to anticipate FSMs. Parameters Elevation Slope Aspect DR Rainfall LULC TRI TWI SPI STI Soil Lithology Multicollinearity test (VIF) 2.948 2.328 1.930 2.186 2.046 2.043 2.414 2.470 2.165 2.149 3.520 1.950 Table 3 shows the Pearson's correlation coefficient for the relationship between floods and its possible causes. Parameters Elevation Slope Aspect D R Rainfall LULC TRI TWI SPI STI Soil Lithology Correlation (Pearson) 0.985 0.875 0.692 0.524 0.425 0.324 0.234 0.082 0.042 0.021 0.009 0.003 The analysis of flood susceptible models For each pixel in the Chennai metropolitan region, maps of flood vulnerability were computed using the two machine learning models, such as ANN and RF. Based on the aforementioned experiment findings, it has been determined that the random forest model is the highest-performing benchmark model for the geographical datasets. For reclassifying the flood sensitive models, ArcGIS 10.8 provides a variety of techniques. Quantile and natural break approaches are two that have been extensively described in the literature that is currently accessible on flood susceptibility investigations. The quantile is a well-known reclassification method that has mostly been used to reclassify flood-prone maps since it can produce better results than other reclassification methods. This is exactly the quantile reclassifying strategy was used for the current investigation. Five categories, including extremely low, low, moderate, high, and very high, were created from the reclassification of the flood susceptibility models. Figure 7 shows the flood susceptibility maps from 2 different derived models (ANN & RF). Table 4 and Fig. 8 shows Two models that are vulnerable to flooding had their accuracy evaluated using various error measures for training and testing data. Table 5 and Fig. 9 shows the results of the Friedman test for models that are prone to flooding in ANN and RF with a p value = 0.05. The results of flood susceptibility model show derived from 2 different models like ANN and RF. In results of ANN represents entire coastal zones are falls under the very high flood susceptibility zone, nearby covers by high and moderate flood susceptibility zones. North side and southwest zones facing very low and low category flood susceptibility. In the model of RF shows half of the coastal zones are covered by very high and nearby surrounding areas are high flood susceptibility. Over all high area covered by moderate zones and little bit coverage near the southwest and northeast zones are falls under low and very low flood susceptibility zones. On the comparability analysis of both models ANN and RF models shows most of the zones are falls under moderate, high and very high flood susceptibility zones, especially Chennai, T. Nagar, Adyar, Lakshmi Nagar, Thoraipakkam, Sholinganallur and Tiruvottiyur zones are falls under very high flood susceptibility zones. Padianallur, Ennore, Avadi and Porur are falls under the zones of high flood susceptibility but in ANN method shows very high flood susceptibility. Karanodai, New vellanur, Thirumudivakkam, chromepet, Tambaram and Perungalathur are zones falls under low flood susceptibility zones but in ANN model shows very low flood susceptibility zones. Over all comparability shows more moderate zones occurs in RF model and less amount of area occupancy in ANN model. Table 6 shows the area classification of ANN model and RF model flood susceptibility zones, along the Chennai metropolitan area. Table 4 Two models that are vulnerable to flooding had their accuracy evaluated using various error measures for training and testing data. S.NO Methods ANN Model RF Model Training 1. RMSE 0.256 0.298 2. MAE 0.175 0.196 3. R 2 0.856 0.924 Table 5 The results of the Friedman test for models that are prone to flooding in ANN and RF with a p value = 0.05. S.NO Model Mean Ranks Chi-Square Significance 1. ANN 1.65 198.56 0 2. RF 2.87 368.43 0 Table 6 Flood susceptibility zones area classification by using ANN and RF S.NO Flood Susceptibility Zones ANN (Area Sq.km) RF (Area Sq.km) 1 Very Low 20 18 2 Low 18 16 3 Moderate 15 13 4 High 25 22 5 Very High 35 31 113 Discussion The Chennai Metropolitan Area (CMA) has already had various flooding incidents, with significant examples in recent years. Old Mahabalipuram Road (OMR), Anna Nagar, Tambaram, Guindy, Velachery, Mogappair, and other places in and around Chennai have historically been prone to flooding. These flooding episodes are primarily the result of a confluence of circumstances, including high rainfall, insufficient stormwater drainage infrastructure, encroachment on water bodies, and urban growth without effective land-use planning. No data values factors are taken in to model that’s particular zones represent very low and low vulnerability zones. The overall comparatively ANN model & RF model shows more number area occupancy in the ANN model than the RF model, So more approximately accurate flood model susceptibility for Chennai metropolitan area. Major location like Chennai, T. Nagar, Adyar, Thoraipakkam and Lakshmi nagar falls under very vulnerability condition zone and some of the zones like Karanodai, chromepet, Tambaram and Perungalathur falls under low vulnerability zones. An accurate assessment of flood susceptibility is a key step for the protection of people and the creation of practical and effective mitigation solutions (Sahana et al., 2019). As a result, the main necessity is to control flooding to avoid it and lessen the damage it causes (Youssef et al, 2011; Zaharia et al, 2020). As a result, the FSM has evolved into the global management pillar (Sahana et al., 2019). In order to get findings with a very high degree of precision and accuracy, researchers always strive to employ innovative and reliable procedures (Youssef et al., 2011). For this reason, we also tried to employ several cutting-edge models, such ANN and RF algorithms, to create maps of Chennai's flood vulnerability. Modelling the region's flood vulnerability is essential since nationwide flooding has been a common occurrence for a very long time. Among the applied approaches, ANN and RF have been widely used for landslide susceptibility modelling (Xu, 2013; Wu et al, 2020; Luo et al, 2019; Hong et al, 2019; Ghasemain et al, 2020;Fang et al, 2020; Chen et al, 2014; ), flood susceptibility modelling (Islam et al, 2021; Bui et al, 2019; Dano et al, 2019; Falah et al, 2019; Kourgialas et al, 2011; Nikolaos et al, 2019; Pradhan, 2010), and forest fire modelling (Vafakhah et al, 2020; Luo et al, 2019), but scarce studies applied random subspace and Dagging model for different natural hazards prediction including flood susceptibility. Additionally, the information gain ratio was used to evaluate how the flood conditioning settings affected the situation. Since ANN and RF performed better in the current study for both training and testing datasets, it can be concluded that these models are also very competent of modelling flood vulnerability. From the model results validation taken place based on the 2 model’s outcomes similar vulnerability shows the validation process of low and high vulnerability zones. Therefore, in order to anticipate such natural disasters with great precision, it is strongly advised to use the RF model. Conclusion This research introduces two novel hybrid ensemble models, ANN and RF, integrated with machine learning models for flood risk mapping in the Chennai metropolitan area. Twelve variables, including elevation, slope, lithology, aspect, SPI, TWI, STI, land use and land cover (LULC), rainfall, distance from river, TWI, and soil types, were utilized to identify 280 flooding spots. Feasibility of flood conditioning parameters was assessed using VIF, IGR, and Pearson's correlation matrix, with feature selection techniques such as information gain ratio employed to gauge their impact. The RF and ANN models exhibited superior flexibility and predictive capacity, warranting their use in flood susceptibility mapping. However, a combination of both models was utilized for its reliability. The study concluded that RF is one of the most effective methods for predicting flood susceptibility. The Chennai Metropolitan area spans 113 sq.km, with 35 to 20 sq.km considered particularly vulnerable to floods. Limitations in modeling temporal changes, such as SPI and land use, necessitate future studies with temporal datasets. Sensitivity analysis on significant variables can further enhance model performance. The RF model offers advantages over other models due to its reduced parameter count, optimization power, and rapid convergence, facilitating efficient flash flood vulnerability mapping. Addressing flood threats in Chennai requires upgraded drainage systems, strict building standards in flood-prone areas, technology-driven early warning systems involving the community, promotion of green spaces, and construction of flood-resistant infrastructure. Community education on flood preparedness and disaster response is crucial. Effective flood control necessitates collaboration between governmental and non-governmental organizations and local communities. Resilience can be enhanced through drills, resource allocation, and ongoing monitoring of flood-prone regions. Sustainable flood control demands a comprehensive strategy integrating infrastructure development, community involvement, and emergency readiness. Declarations Authors' contributions: V. Nivedita - Manuscript writing, analysis, data processing, review and editing. S. Sabarunisha Begum - Data processing, writing, review and editing. U. Sakthi - Analysis, editing and corrections. V. Sellam - Data processing, analysis and editing. C. Navaneethan - Analysis, editing and corrections. Mohamed Yacin Sikkandar - Analysis, editing and corrections. T. Jayasankar - Analysis, editing and corrections. S. Vivek - Data processing, analysis and editing. Acknowledgements : The authors would like to express their gratitude to the anonymous referees and the editor/associate editor for their constructive comments and valuable suggestions. Funding: This work does not have any funding. Availability of data and materials : All data generated or analysed during this study are included in the manuscript. Conflict of Interest The authors declare no conflict of interests. Ethics approval Not applicable. Consent to participate Not applicable. Consent for publication Not applicable. References Abu Reza Md Towfiqul Islam, Swapan Talukdar, Susanta Mahato, Sonali Kundu, Kutub Uddin Eibek, Quoc Bao Pham, Alban Kuriqi, Nguyen Thi Thuy Linh (2021), Flood susceptibility modelling using advanced ensemble machine learning models https://doi.org/10.1016/j.gsf.2020.09.006. Akter, J., Das, L., Meher, J.K., Deb, A., (2019). Evaluation of different large-scale predictor-based statistical downscaling models in simulating zone-wise monsoon precipitation over India. Int. J. Climatol. 39, 465–482. Alexander, K., Hettiarachchi, S., Ou, Y., Sharma, A., (2019). Can integrated green spaces and storage facilities absorb the increased risk of flooding due to climate change in developed urban environments? J. Hydrol. 579, 12420 Bui, D.T., Tsangaratos, P., Ngo, P.T.T., Pham, T.D., Pham, B.T., (2019) b. Flash flood susceptibility modeling using an optimized fuzzy rule-based feature selection technique and tree-based ensemble methods. Sci. Total Environ. 668, 1038–1054. Cao, Q., Mehran, A., Ralph, F.M., Lettenmaier, D.P, (2019). The role of hydrological initial conditions on Atmospheric River floods in the Russian River basin. J. Hydrometeorol. Chen, W., Li, W., Hou, E., Zhao, Z., Deng, N., Bai, H., Wang, D., (2014). Landslide susceptibility mapping based on GIS and information value model for the Chencang District of Baoji, China. Arab. J. Geosci. 7 (11), 4499–4511 Dano, U.L., Balogun, A.L., Matori, A.N., Wan Yusouf, K., Abubakar, I.R., Said Mohamed, M.A,Aina, Y.A., Pradhan, B.,(2019). Flood susceptibility mapping using GIS-based analytic network process: A case study of Perlis, Malaysia. Water 11 (3), 615. Deng, S., Liu, S., Mo, X., (2020). Assessment of Three Common Methods for Estimating Terrestrial Water Storage Change with Three Reanalysis Datasets. J. Clim. 33 (2),511–525 Douglas, E.M., Vogel, R.M., Kroll, C.N., (2000). Trends in floods and low flows in the United States: impact of spatial correlation. J. Hydrol. 240 (1–2), 90–105. Duque, E.L., Aquino, P.T, (2020). Anthropometric Analysis in Automotive Manual Transmission Gearshift Quality Perception. CTI SYMPOSIUM 2018. Springer Vieweg, Berlin, Heidelberg, pp. 97–109. Falah, F., Rahmati, O., Rostami, M., Ahmadi sharaf, E., Dimakopoulos, I.N., Pourghasemi, H.R., (2019). Artificial neural networks for flood susceptibility mapping in data-scarce urban areas. Spatial modeling in GIS and R for Earth and Environmental Sciences. Elsevier, pp. 323–336 139. Fang, Z., Wang, Y., Peng, L., Hong, H., (2020). Integration of convolutional neural network and conventional machine learning classifiers for landslide susceptibility mapping. Comput. Geosci., 104470 García-Ruiz, J.M., Regüés, D., Alvera, B., Lana-Renault, N., Serrano-Muela, P., Nadal Romero, E., ... Arnáez, J., (2008). Flood generation and sediment transport in experimental catchments affected by land use changes in the central Pyrenees. J. Hydrol.356 (1-2), 245–260. Getahun, Y.S., Gebre, S.L., (2015). Flood hazard assessment and mapping of flood inundation area of the Awash River Basin in Ethiopia using GIS and HEC-GeoRAS/HEC-RAS model. J. Civil Environ. Eng. 5 (4), 1. Ghasemain, B., Asl, D.T., Pham, B.T., Avand, M., Nguyen, H.D., Janizadeh, S., (2020). Shallow landslide susceptibility mapping: A comparison between classification and regression tree and reduced error pruning tree algorithms. Vietnam J. Earth Sci Hong, H., Miao, Y., Liu, J., Zhu, A.X., (2019). Exploring the effects of the design and quantity of absence data on the performance of random forest-based landslide susceptibility mapping. Catena 176, 45–64. Hong, H., Shahabi, H., Shirzadi, A., Chen, W., Chapi, K., Ahmad, B.B., ... Bui, D.T., (2019). Landslide susceptibility assessment at the Wuning area, China: A comparison between multi-criteria decision making, bivariate statistical and machine learning methods. Nat. Hazards 96 (1), 173–212. Hosni, M., Abnane, I., Idri, A., de Gea, J. M. C., & Alemán, J. L. F. (2019). Reviewing Ensemble Classification Methods in Breast Cancer. Comput. Methods Prog. Biomed. Hou, G., Shao, Q., Zou, B., Dai, L., Zhang, Z., Mu, Z., Zhai, J., 2020. A Novel Underwater Si-multaneous Localization and Mapping Online Algorithm Based on Neural Network. ISPRS Int. J. Geo Inf. 9 (1), 5. Kourgialas, N.N., Karatzas, G.P., (2011). Flood management and a GIS modelling method to assess flood-hazard areas—a case study. Hydrol. Sci. J. 56 (2), 212–225. Kumar, R., Nayak, P.C., Patra, J.P., Mani, P., (2019). Chapter-10-Impact of climate change on floods. National Institute of Hydrology. Kuncheva, L.I., Rodríguez, J.J., Plumpton, C.O., Linden, D.E., Johnston, S.J., (2010). Random subspace ensembles for fMRI classification. IEEE Trans. Med. Imaging 29 (2), 531–542. Kuriqi, A., Ardiçlioǧlu, M.,(2018). Investigation of hydraulic regime at middle part of the Loire River in context of floods and low flow events. Pollack Periodica 13 (1),145–156. Luo, X., Lin, F., Chen, Y., Zhu, S., Xu, Z., Huo, Z., ... Peng, J., (2019). Coupling logistic model tree and random subspace to predict the landslide susceptibility areas with considering the uncertainty of environmental features. Sci. Rep. 9 (1), 1–13. Ma, J., Ding, Y., Cheng, J.C., Jiang, F., Tan, Y., Gan, V.J., Wan, Z.,(2020). Identification of high impact factors of air quality on a national scale using big data and machine learning techniques. J. Clean. Prod. 244, 118955. Masks, D.L.G., Starr, N.B., Brady, M.A., Gaylord, N.M., Driessnack, M., Duderstadt, K., (2019). Burns' Pediatric Primary Care (7th Edition). Elsevier Nikolaos, S., Kleomenis, K., Elias, D., Panagiotis, S., Panagiota, L., Vagelis, P., Christos, C., (2019). A Robust Remote Sensing–Spatial Modeling–Remote Sensing (RMR) Approach for Flood Hazard Assessment. Spatial Modeling in GIS and R for Earth and Environmental Sciences. Elsevier, pp. 391–410. Oeurng, C., Sauvage, S., Sánchez-Pérez, J.M., (2011). Assessment of hydrology, sediment and particulate organic carbon yield in a large agricultural catchment using the SWAT model. J. Hydrol. 401 (3-4), 145–153. Pradhan, B., (2010). Flood susceptible mapping and risk area delineation using logistic regression, GIS and remote sensing. J. Spat. Hydrol. 9, 2. Prăvălie, R., Costache, R., (2013). The vulnerability of the territorial-administrative units to the hydrological phenomena of risk (flash-floods). Case study: the subcarpathian sector of Buzău catchment. Analele Universității din Oradea–Seria Geografie 23 (1),91–98. Rumelhart, D.E., Hinton, G.E., McClelland, J.L., (1986). A general framework for parallel distributed processing. Parallel distributed processing: Explorations in the microstructure of cognition. 1, p. 26 45-76. Sahana, M., Patel, P.P., (2019). A comparison of frequency ratio and fuzzy logic models for flood susceptibility assessment of the lower Kosi River Basin in India. Environ.Earth Sci. 78 (10), 289 Torcivia, C.E.G., López, N.N.R., (2020). Preliminary Morphometric Analysis: Río Talacasto Basin, Central Precordillera of San Juan, Argentina. Advances in Geomorphology and Quaternary Studies in Argentina. Springer, Cham, pp. 158–168. Uthayakumar, J., Metawa, N., Shankar, K., Lakshmanaprabu, S.K., (2020). Financial crisis pre-diction model using ant colony optimization. Int. J. Inf. Manag. 50, 538–556. Vafakhah, M., Loor, S.M.H., Pourghasemi, H., Katebikord, A., (2020). Comparing performance of random forest and adaptive neuro-fuzzy inference system data mining models for flood susceptibility mapping. Arab. J. Geosci. 13, 417. Wu, Y., Ke, Y., Chen, Z., Liang, S., Zhao, H., Hong, H., (2020) b. Application of alternating decision tree with AdaBoost and random subspace ensembles for landslide susceptibility mapping. Catena 187, 104396. Xie, H., Dong, J., Shen, Z., Chen, L., Lai, X., Qiu, J., ... Chen, X., (2019). Intra-and inter-event characteristics and controlling factors of agricultural nonpoint source pollution under different types of rainfall-runoff events. Catena 182, 104105. Xu, C., (2013). Assessment of earthquake-triggered landslide susceptibility based on expert knowledge and information value methods: a case study of the 20 April 2013 Lushan, China Mw6. 6 earthquakes. Disaster Advances 6 (13), 119–130 Youssef, A.M., Pradhan, B., Hassan, A.M., (2011). Flash flood risk estimation along the St. Katherine Road, southern Sinai, Egypt using GIS based morphometry and satellite imagery. Environ. Earth Sci. 62 (3), 611–623. Zaharia, L., Ioana-Toroimac, G., Perju, E.R.,(2020). Hydrological Impacts of Climate Changes in Romania. Water Resources Management in Romania. Springer, Cham, pp. 309–351 Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4194276","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":340812953,"identity":"842ec77d-4a9a-4382-bac1-ac9c6f3d31b6","order_by":0,"name":"Nivedita V","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9klEQVRIiWNgGAWjYDACZiDmMQCxDh988AFIsbETpwWoh/FYsuEMkBZmYmziYQBqYT5jJs0DMwQfMGfnMXzwpuCPvMGxA8bGNr+2yfMxMzB++JiDW4tlM4+x4RwDA8MNZw4kPs7tu23YxszALDlzG24tBofZ0qSBfmGcOePAYePcntuMQC1szLz4taT/Bmqxnzn/YZu0Zc9teyK0MB9jBmpJ7Gc4zCbN8ON2IjFaDkvOMTBO7mc4xmzY23A7uY2ZsRm/X84fbPzw5o+cbRvD+Y8Pfvy5bTu/vfngh494tKACxjYw2UCsehD4Q4riUTAKRsEoGCkAAM3CUD+7pT2TAAAAAElFTkSuQmCC","orcid":"","institution":"SRMIST: SRM Institute of Science and Technology (Deemed to be University)","correspondingAuthor":true,"prefix":"","firstName":"Nivedita","middleName":"","lastName":"V","suffix":""},{"id":340812954,"identity":"bfdc04a8-0b04-477a-9bad-91d1d315c3d3","order_by":1,"name":"Sabarunisha Begum S","email":"","orcid":"","institution":"PSR Engineering College","correspondingAuthor":false,"prefix":"","firstName":"Sabarunisha","middleName":"Begum","lastName":"S","suffix":""},{"id":340812955,"identity":"203d1497-10ef-44b9-8bec-b6d0a6e580a2","order_by":2,"name":"Sakthi U","email":"","orcid":"","institution":"SRM Institute of Science and Technology (Deemed to be University)","correspondingAuthor":false,"prefix":"","firstName":"Sakthi","middleName":"","lastName":"U","suffix":""},{"id":340812956,"identity":"6fa7b53f-8e32-4948-a495-eb6e59120fee","order_by":3,"name":"Sellam V","email":"","orcid":"","institution":"Vellore Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Sellam","middleName":"","lastName":"V","suffix":""},{"id":340812957,"identity":"97b55b09-d1c9-48d3-81f7-77983c07707c","order_by":4,"name":"Navaneethan C","email":"","orcid":"","institution":"Vellore Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Navaneethan","middleName":"","lastName":"C","suffix":""},{"id":340812958,"identity":"c58c5a42-1514-4ae4-ba43-f6cb05826944","order_by":5,"name":"Mohamed Yacin Sikkandar","email":"","orcid":"","institution":"Majmaah University","correspondingAuthor":false,"prefix":"","firstName":"Mohamed","middleName":"Yacin","lastName":"Sikkandar","suffix":""},{"id":340812959,"identity":"06623cfb-b2b7-497d-95a8-d2f9c8a14138","order_by":6,"name":"Jayasankar T","email":"","orcid":"","institution":"University College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Jayasankar","middleName":"","lastName":"T","suffix":""},{"id":340812960,"identity":"ef3241aa-3ea6-4d40-9c8d-18969b8abecf","order_by":7,"name":"Vivek S","email":"","orcid":"","institution":"GMRIT: GMR Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Vivek","middleName":"","lastName":"S","suffix":""}],"badges":[],"createdAt":"2024-03-31 05:16:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4194276/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4194276/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":64364085,"identity":"93f500a2-3d3f-447f-a31e-3bf587b2e767","added_by":"auto","created_at":"2024-09-12 07:41:00","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":556899,"visible":true,"origin":"","legend":"\u003cp\u003eThe study area location extraction Map for Chennai Metropolitan Area\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/fe934f225c99261f3de4bfb3.jpg"},{"id":64364095,"identity":"c6a319d1-4347-45a1-b432-588353a60f96","added_by":"auto","created_at":"2024-09-12 07:41:00","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":87468,"visible":true,"origin":"","legend":"\u003cp\u003eThe entire methodology flow chart for Flood Susceptibility Mapping for Chennai Metropolitan Area, Tamil Nadu, India\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/cfc9e30af737be27418e402e.png"},{"id":64364092,"identity":"da0859cc-cfce-4ebd-a287-03bce4dae9eb","added_by":"auto","created_at":"2024-09-12 07:41:00","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":499036,"visible":true,"origin":"","legend":"\u003cp\u003eFactors for flood susceptibility models for Chennai metropolitan area like A) Elevation, B) Slope, C) Aspect and D) Rainfall\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/ea14d7362a348cfe62bc08a5.jpg"},{"id":64364086,"identity":"0c0350a4-98dc-4dd0-9160-6d9749bf6093","added_by":"auto","created_at":"2024-09-12 07:41:00","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":499810,"visible":true,"origin":"","legend":"\u003cp\u003eFactors for flood susceptibility models for Chennai metropolitan area like A) Topographic roughness index, B) Topographic wetness index, C) Stream power index and D) Sediment transport index\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/737669d787083e8ba8de1d74.jpg"},{"id":64364692,"identity":"0ec0fc49-0705-4403-9dff-7c1f69ce9f84","added_by":"auto","created_at":"2024-09-12 07:49:00","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":465050,"visible":true,"origin":"","legend":"\u003cp\u003eFactors for flood susceptibility models for Chennai metropolitan area like A) LULC, B) Distance from the road, C) lithology and D) Soil\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/7b5b58f2d69f16b5cb6ff089.jpg"},{"id":64364693,"identity":"53857db4-d2be-4f02-9276-45c020d39ad0","added_by":"auto","created_at":"2024-09-12 07:49:00","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":262367,"visible":true,"origin":"","legend":"\u003cp\u003eThe standard error bars demonstrate the accuracy of the findings and highlight the significance of flood conditioning elements using the information gain ratio.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/80fd80b6ceb84e5bf26234d9.png"},{"id":64364094,"identity":"8be7e235-a96b-4d68-90dc-78e884e931be","added_by":"auto","created_at":"2024-09-12 07:41:00","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":607987,"visible":true,"origin":"","legend":"\u003cp\u003eFlood susceptibility Map derived from two machine learning models (ANN \u0026amp; RF) for Chennai metropolitan Area\u003c/p\u003e","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/7aac9f08d18de7e4010c6d67.jpg"},{"id":64364088,"identity":"8e571b09-69e7-4748-9c46-9dc316140da4","added_by":"auto","created_at":"2024-09-12 07:41:00","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":35139,"visible":true,"origin":"","legend":"\u003cp\u003eANN and RF Model vulnerable to flooding and their accuracy evaluated using various error measures for training and testing data for Chennai metropolitan area.\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/2688530a96f451df45f6c565.png"},{"id":64364694,"identity":"4fa1092d-382d-449a-8510-bb0d4af93ee6","added_by":"auto","created_at":"2024-09-12 07:49:00","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":9446,"visible":true,"origin":"","legend":"\u003cp\u003eResults of the Friedman test for two models that are vulnerable to flooding for Chennai metropolitan area.\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/0f4c6ac5bd59454c569fbf5d.png"},{"id":66511166,"identity":"0609a44a-7218-4485-90be-32a8f73fa5aa","added_by":"auto","created_at":"2024-10-13 22:49:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3593941,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4194276/v1/39e36f74-8e71-4383-a18b-4e4f9d99bc8d.pdf"}],"financialInterests":"","formattedTitle":"Assessment of flood vulnerability in a coastal metropolitan city for sustainable environmental using Machine Learning methods","fulltext":[{"header":"Introduction","content":"\u003cp\u003eA flood is a type of natural disaster that happens when an area of typically dry land is submerged by a large volume of water (Douglas et al, 2000). Heavy rainfall, quick snowmelt, dam breaches, or overflowing rivers, lakes, and seas are a few of the possible causes of this phenomena (Hong et al, 2019). Floods may have a catastrophic impact on the environment and populated areas, frequently causing extensive damage and fatalities (Kuncheva et al, 2010). The earth may become saturated when heavy rain falls quickly, overburdening drainage systems and causing surface runoff (Luo et al, 2019). Water levels in rivers and streams rise quickly as a result of this extra water flowing into them (Kuriqi et al, 2018). This might cause floods in locations with insufficient infrastructure to handle the extra water (Masks et al, 2019). Similar to how snowmelt during the spring thaw may boost water flow into rivers and cause floods (Hosni et al, 2019). Depending on their severity, floods can be categorized into a variety of groups. Intense and rapid flash floods can happen minutes or hours after significant rainfall. Due to their sudden beginning and potential to catch individuals off guard, they constitute a serious hazard (Ma et al, 2020). On the other hand, when rivers gradually rise and overrun their banks, river floods occur over a longer period of time (Islam et al, 2021). Sustainability involves satisfying current requirements while preserving the capacity of future generations to fulfill their own. It encompasses the conscientious utilization of resources, the advocacy of environmental well-being, social fairness, and economic prosperity. Sustainable approaches strive to harmonize ecological, societal, and economic aspects for enduring welfare and resilience. Sustainable flood management practices can benefit agricultural land by replenishing soil nutrients, improving water availability, and enhancing crop growth. Proper flood control measures, such as the construction of sustainable drainage systems, can mitigate soil erosion, promote groundwater recharge, and foster ecosystem resilience, supporting long-term agricultural sustainability.\u003c/p\u003e \u003cp\u003eWhen ocean waters surge inland, frequently as a result of powerful storms or tsunamis, coastal floods happen (Nikolaos et al, 2019). A flood may have a significant negative effect on local residents and the environment. Floodwaters can cause economic losses by eroding the soil, destroying infrastructure, and destroying crops (Duque et al, 2020). Homes, businesses, and public buildings may sustain damage or destruction, forcing people to relocate and upsetting everyday life (Oeurng et al, 2011). Floods can result in substantial human suffering, including fatalities and serious injuries, in densely populated places (Hou et al, 2020). Flood impact reduction initiatives frequently include planning, response, and recovery techniques (Pradhan et al, 2010). Zoning for floodplains and building codes assist control where and how development happens in flood-prone regions (Akter et al, 2019). Communities can efficiently prepare for and respond to floods thanks to early warning systems and emergency plans (Pravalie et al, 2013). Levee construction and dam construction are examples of infrastructure upgrades that can assist manage and steer flooding (Falah et al, 2019). In certain areas, the frequency and severity of floods have increased in recent years due to the consequences of climate change (Rumelhart et al, 1986). Higher sea levels are a result of glaciers and polar ice melting due to rising global temperatures (Sahana et al, 2019). The susceptibility of many areas to flooding occurrences has risen as a result of this and increasingly intense weather patterns (Alexander et al, 2019).\u003c/p\u003e \u003cp\u003eBy replicating the dynamics and impacts of floods, flood modelling helps forecast and lessen their effects (Torcivia et al, 2020). It influences infrastructure development, disaster preparedness, and urban planning (Uthayakumar et al, 2020). Models calculate the possible flood extent, depth, and velocity by examining variables including rainfall, geography, and land use, enabling a well-informed risk assessment (Fang et al, 2020). These models support the development of floodplain management policies, resilient building design, and the optimization of evacuation plans (Bui et al, 2019). They make it easier to make decisions for catastrophe preparation and response, reducing property loss and protecting lives (Ruiz et al, 2008). For the development of models that are susceptible to flooding, many academics have used a variety of techniques (Kumar et al, 2019). Since there are many different types of flood-susceptible models, they can be employed in many different research, but in this research, we used an artificial neural network and random forest (Getahun et al, 2015). Hybrid machine learning techniques are being implemented and developed for several natural hazard models as a result of their great results (Kourgialas et al, 2011). On the optimum technique for simulating different types of natural hazards, such as landslides or flood susceptibility, there was, however, no general agreement. Researchers advise creating and experimenting with novel techniques for modelling natural disasters and mapping flood susceptibility (Cao et al, 2019).\u003c/p\u003e \u003cp\u003eIn order to simulate flood susceptibility in the Chennai metropolitan area, we also created ensemble machine learning methods including Artificial Neural Network and Random Forest models (Ghasemain et al, 2020). Few applications of the hybrid ensembles of the Random Forest (RF) and Artificial Neural Network (ANN) models for simulating natural disasters have been made (Chen et al, 2014). In this study, we selected the Chennai area, known for experiencing frequent flash floods, as their research case to develop a flash flood susceptibility map (Hong et al., 2019). The study's novelty can be attributed to two main aspects: (i) The production of a more comprehensive flood susceptibility model for the specific study area; and (ii) The introduction of a novel hybrid ensemble model called Random Forest (RF) combined with Artificial Neural Networks (ANN) to model flood susceptibility maps, a method never used before in in this study area. The contribution of this research involves utilizing both traditional models (RF and ANN) and ensemble models to accurately map flood susceptibility and identify flood hazards in the Chennai area (Deng et al., 2020). This innovative approach aims to improve flood prediction and mitigation strategies in the region. The study's findings will help regional and local authorities and policymakers reduce the risks associated with floods and establish effective mitigation strategies to prevent possible harm.\u003c/p\u003e"},{"header":"Study Area","content":"\u003cp\u003eThe city of Chennai and the surrounding urban and suburban areas are all included in the Chennai Metropolitan Area (CMA). It is a prominent commercial and cultural center in southern India, distinguished by a variety of businesses, institutions of higher learning, and historical sites. The CMA attracts both domestic and foreign investors and is essential to regional trade, technology, and services. However, the effects of increased urbanization include infrastructural needs, environmental degradation, and traffic congestion. For the Chennai Metropolitan Area to continue growing and prospering, it is imperative to solve these concerns through effective urban planning, transportation management, and sustainable development. Due to its geological history, the Chennai Metropolitan Area is primarily distinguished by a mix of soil types and lithology. A mixture of sedimentary, igneous, and metamorphic rocks makes up the region's geology. Alluvial soils, red soils, black soils, and laterite soils are examples of typical soil types. Red and black soils are common in the hinterlands, whereas alluvial soils are found in riverbanks and along the shore. Major river flows in study area which include kosasthaliayar river, Coocum river, Adyar river and Buckingham canal.\u003c/p\u003e \u003cp\u003eWaterlogging occurs when inadequately designed drainage systems are unable to manage significant monsoon rains. Water body encroachment and inappropriate trash disposal make the issue worse. The regular floods are also a result of insufficient upkeep of the current infrastructure and ineffective floodplain management. Comprehensive urban planning, better drainage systems, the restoration of water bodies, public awareness campaigns, and sustainable development practices are necessary to minimize and manage the recurrent flooding difficulties in the Chennai Metropolitan Area in order to meet these challenges. In this study, flood susceptibility models were used. These are the reasons listed above for taking severe measures for proper management plans in the Chennai Metropolitan area using the flood susceptibility analysis by using machine learning along with models like ANN and RF.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003eAccording to the methodological flow chart in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the current study used the flood inventory map, the development of flood conditioning factors, the evaluation of the flood conditioning factors using information gain ratio and multi-collinearity test, and flood susceptibility models using machine learning techniques (ANN and RF) (Nikolaos et al., 2019). The groundwork of a flood inventory map for the study region is the first stage in creating a map of flood vulnerability. The usual state at the flooded areas was achieved and investigated using the readily available GIS technology. The chance of a flood in the near future will rely on the likelihood of a flood in the recent past, making this a crucial stage in the flood prediction process. The development of the sensitive flood model is frequently extremely precise and difficult due to the requirement for numerous geographic topographical and hydrological parameters. Determining the causes of the flood is therefore crucial, and the systematically designated factors will confirm the correctness of the maps of flood susceptibility. The flood susceptibility literature that is currently available, twelve flood-influencing factors, including elevation, lithology, slope, aspect, topographic wetness index (TWI), topographic roughness index (TRI), sediment transport index (STI), stream power index (SPI), land use/land cover (Lu/Lc), distance to the river, soil type, and rainfall, were selected for the current study area.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe conversion of all affecting factors into a raster format with a spatial resolution of 30m was done. Topographic factors must be taken into account while modelling flood studies since they have a direct and indirect impact on the hydrological features of the study region. First, a Digital Elevation Model (DEM) for the research basin was produced in the ArcGIS 10.8 environment using the ASTER GDEM (Version 2). The production of Digital Elevation Models (DEMs) involves the collection of elevation data using various methods, such as airborne LiDAR or satellite-based radar. These data are processed to create a comprehensive representation of the Earth's surface, enabling the visualization and analysis of terrain features for applications in cartography, engineering, and environmental studies. From DEM topographic parameters in the ArcGIS environment, slope, lithology, aspect, TWI, SPI, STI, and TRI have all been produced.\u003c/p\u003e \u003cp\u003eElevation: Flooding and height are inversely associated; the higher the elevation, the lesser likely flooding will occur, and vice versa. It is possible to derive Chennai Metropolitan Areas elevation information using DEM data.\u003c/p\u003e \u003cp\u003eSlope: Another crucial factor that influences a flood is the slope, it controls the speed of the water's flow. The likelihood of water stagnation is reduced, infiltration is reduced, and flow velocity is increased with increasing slope angle. The slope for the research area is generated using the Arc toolbox viva spatial analyst tool, and the surface slope is generated using the Arc GIS software's Slope tool. Input data is DEM to generate slope as output.\u003c/p\u003e \u003cp\u003eAspect: Another component, aspect, determines the directions that flooded water moves in addition to maintaining soil humidity. The aspect thus has an indirect impact on flooding. Consider the section of a slope that is shaded, where the soil has a high relative humidity and there is significant runoff. Aspect tool in the surface tool subsection is used to generate aspects such as slope-wise aspects.\u003c/p\u003e \u003cp\u003eRainfall: One of the key elements that influence the likelihood of flooding has been identified as rainfall. because flooding may result after a brief period of heavy rain. We employed IDW interpolation and rainfall data from four Chennai meteorological stations to produce rainfall maps in the ArcGIS 10.8 environment. We interpolated using the IDW approach because we only had data for four locations, despite the fact that this method is strongly advised when there is a very little amount of data.\u003c/p\u003e \u003cp\u003eTRI: The TRI is the most important factors impelling flood events. It is based on the neighborhood's topography in the research area. The chance of a flood increasing with decreasing TRI values. By using Focal statistical tool to create Minimum, Maximum and Mean raster file with input data of DEM. To generate the TRI for the study area followed by the equation in the raster calculator tool,\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\text{T}\\text{R}\\text{I}=\\text{M}\\text{e}\\text{a}\\text{n}-\\text{M}\\text{i}\\text{n}\\text{i}\\text{m}\\text{u}\\text{m} / \\text{M}\\text{a}\\text{x}\\text{i}\\text{m}\\text{u}\\text{m}-\\text{M}\\text{i}\\text{n}\\text{i}\\text{m}\\text{u}\\text{m}\\)\u003c/span\u003e \u003c/span\u003e \u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;.1\u003c/p\u003e \u003cp\u003eTWI: It expresses the alteration in wetness of a basin spatially, is a significant determinant in the likelihood of flood. This index displays the amount of water present in each individual pixel in the area. Using the following equation, TWI are\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(TWI =\\frac{In \\left(As\\right)}{\\text{t}\\text{a}\\text{n}{\\beta }}\\)\u003c/span\u003e \u003c/span\u003e\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;.2\u003c/p\u003e \u003cp\u003eAs and, respectively, represent the explicit catchment area (m\u003csup\u003e2\u003c/sup\u003e m\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) and slope gradient (in degrees). In general, there is a direct correlation between floods and high TWI readings.\u003c/p\u003e \u003cp\u003eSPI: It has a substantial effect on the fluvial system. To determine the SPI, use the equation below,\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(SPI =As tan\\beta\\)\u003c/span\u003e \u003c/span\u003e\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;3\u003c/p\u003e \u003cp\u003eWhereas, As is the slope gradient stand in for the particular catchment region, and are denoted by (radians). Total SPI is the term used to describe both the bed's erodibility and its ability to transport sediment.\u003c/p\u003e \u003cp\u003eSTI: Another factor that might cause flooding is the STI, which can increase the frequency of flooding and cause damage to foundations. The following equation is used to derive the STI from the DEM are,\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(STI=\\left({\\frac{AS}{22.13})}^{2}* \\right({\\frac{sin\\beta }{0.0896})}^{2}\\)\u003c/span\u003e \u003c/span\u003e\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;.4\u003c/p\u003e \u003cp\u003ewhere the letter A stands in for the area upstream and the symbol designates each pixel of the slope. The hydro-climatic and geomorphologic parameters of the basin region are used to calculate the STI. As sediment is deposited, the channel's bed shifts, reducing the channel's capacity to hold water and leading to flooding.\u003c/p\u003e \u003cp\u003eLand use/Land cover: Flood frequency is directly impacted by LULC because it affects sediment transportation and surface runoff. Because the formation and infiltration of surface runoff are directly controlled by the LULC. As a result of these areas' inability to produce surface water and allow for water infiltration, flooding occurs more frequently there. The woodland region, on the other hand, provides for water infiltration that lessens flooding. When comparing hydrological responses at different temporal scales, the connection between flood episodes and vegetation density is inverse. In the present study, a Land Use Land Cover (Lu/Lc) map was created using Landsat and OLI (Operational Land Imager) satellite images. The Artificial Neural Network (ANN) technique was employed for this purpose, and the analysis was conducted using ENVI software (version 5.3) with a spatial resolution of 30 meters. This approach allowed for accurate classification and mapping of different land cover types in the study area, providing valuable information for various environmental and land management applications. The LULC map was divided into five categories: agricultural land, waste land, urban area, grassland, and aquatic body.\u003c/p\u003e \u003cp\u003eDistance to the river: The majority of flooded places are typically found close to rivers. Because the distance from the river affects the likelihood of flooding and the ratio of river flow to river, it is a crucial determining element in determining the research area's flood-prone areas. The likelihood of flood events decreases with increasing distance from the river. Flooding is connected to the loading of terrestrial water at the local level. In the current study, we created a map showing the distance to the river using Google Earth Pro, converted the KML file into a shapefile in the Arc GIS environment, and created a buffer for the main river to calculate the distance from the river.\u003c/p\u003e \u003cp\u003eSoil: One of the major determining factors that impacts how rainfall-runoff functions is soil. While other factors, like the local climate and the erosion process, also have an impact on how rainfall-runoff forms, the soil possessions sprightly control water penetration. The higher the rate of soil infiltration, the less frequently flooding happens. The National Bureau of Soil Survey provided the soil map that was utilised to digitise the soil for this investigation.\u003c/p\u003e \u003cp\u003eLithology: The study of rock properties known as lithology affects flood behaviour. Sand and other porous lithologies reduce flood risk by absorbing water. Clay-like impermeable rocks can cause surface runoff, which heightens floods. In order to better control floods, flood extent can be predicted and effective drainage systems can be planned by understanding lithology. Lithology map was prepared by using Geological Survey of India lithology map.\u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eMethod for flood influencing factors using Information gain ratio and multicollinearity test\u003c/h2\u003e \u003cp\u003eIt is essential to evaluate the importance of the flood affecting parameter or the probability for flooding before beginning the model's training sections. Based on each parameter's statistical traits and connection to the floods, its relative importance has been determined. The Information Gain Ratio (InGR) approach has been used to determine the influential factors for FSM prediction. An InGR value is assigned to each influencing element in order to quantify its significance. Higher InGR values are indicative of more pertinent influencing elements. The decision to use the InGR model in the current experiment was based on its simplicity and effectiveness. The InGR model is well-suited for the research objectives and provides valuable insights into the relevant influencing factors, making it an appropriate choice for the study. It is computed using the following equation are,\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(Gain Raito \\left(x, Z\\right)=\\frac{Entropy \\left(Z\\right)- {\\sum }_{1}^{n}{\\sum }_{i=1}^{n}\\frac{Zi}{z} Entropy \\left(zi\\right)}{-\\sum _{i=1}^{n}\\frac{Zi}{Z}\\text{log}\\frac{Zi}{Z}}\\)\u003c/span\u003e \u003c/span\u003e\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;5\u003c/p\u003e \u003cp\u003eIf the property x originates from training point Z with subsets Zi1\u0026thinsp;=\u0026thinsp;1, 2, 3, etc. Using a range of multi-collinearity tests, such as variance decomposition proportions, conditional index, VIF, and tolerances, influencing factors have been evaluated for all probability models. We utilised the Pearson's relationship coefficient and the VIF to determine the respective weights of the twelve flood training factors in this investigation. The VIF\u0026thinsp;\u0026gt;\u0026thinsp;9 and incredibly weak correlation serve as indicators of the problem of multicollinearity in the components. Therefore, if the conditioning factor's VIF value is more than 9, it is highly recommended to leave it out of the model.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eFood susceptibility modelling-ANN\u003c/h2\u003e \u003cp\u003eThe three-layer ANN model (input, hidden, and output layers) utilised in the current work using a Back-Propagation (BP) and error correction learning method has been effectively utilised in the flood susceptibility modelling. The input layers and ten hidden nodes in this investigation were configured with the same numbers as the critical parameters (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The output layer, on the other hand, uses a single node and is coded as 1 for flood occurrences and 0 for non-flood events. Although there are other techniques for training ANN models, BP is the most often used one. Thus, the BP based ANN model has been used to estimate the nonlinear connection between the essential parameters and the flood occurrences. First, BP selects the starting weights at random. There has been a comparison of calculated and observed values. Errors are defined as discrepancies between calculated values and observed values. Several error measurement approaches, including mean squared error (MSE) and root mean square error (RMSE), have been used to analyse it. The initial weights are adjusted based on the generalised delta rule to distribute the entire error across the network's neurons. This method is iterated when the degree of mistake is at a reduced degree.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eRandom Forest\u003c/h2\u003e \u003cp\u003eThe groundbreaking Random Forest (RF) approach combines classification and regression decision trees to make accurate predictions. It is a popular ensemble-learning method. Ho's concept of \"random selection features\" and Breiman's \"random subspace\" are essential components of the RF, which can be divided into two subgroups. The Random Subspace is an ensemble machine learning technique that generates multiple classifiers in order to boost the prediction accuracy of a subpar classifier right from the beginning. To predict the data classification, the RF performs numerous regression tree training stages and generates diverse sets of samples via sampling with replacement. The final classification chosen by the RF is based on the voting outcomes of several classifiers, ensuring a significant number of votes from each tree in the forest. During the regression tree's training phase, the observation datasets are categorized using rules based on response parameters until the prediction achieves the lowest possible node deviation.\u003c/p\u003e \u003cp\u003eOne of the major drawbacks of regression trees is their tendency to overfit the training data, resulting in poor performance when faced with an unknown dataset. However, Random Forest (RF) can help address this issue. In the RF algorithm, during the training of each regression tree, a random subset of input archives and predictor factors is chosen as input. Through various sampling approaches, different sets of regression trees are generated, each trained on a different randomly chosen subset. Using a total sample to train the decision trees is not recommended, as it disregards the importance of local samples. Therefore, RF employs a more robust approach by creating diverse subsets for training individual trees. In flood susceptibility analysis, the RF model serves as a benchmark for comparing outcomes with those of a new hybrid model, highlighting its usefulness in such applications.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe parameters of the machine learning algorithm used for flood susceptibility modelling\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS.No\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModel name\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDescription of parameters\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eANN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHidden layer-7, learning rate-0.38, momentum-0.25, seed-3, training time-750, validation threshold-15, Normal to binary filter-TRUE\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRandom forest\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBatch size-100, seed-4, number of iteration-175, max depth-2, calc out of bag-TRUE, Compute attribute importance-TRUE\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFactors for modelling flood susceptibility\u003c/h2\u003e \u003cp\u003eFor analysis of flood susceptibility in Chennai Metropolitan area by using machine learning models which include several factors which are elevation, slope, aspect, rainfall, TRI, TWI, SPI, STI, LULC, distance from the river, soil and lithology are prepared for the major input for training sets factors for flood susceptibility models (Vafakhah et al, 2020; Wu et al, 2020; Xie et al, 2019).\u003c/p\u003e \u003cp\u003eElevation: Elevation for Chennai metropolitan area are under the value of low from surface of the land ranges are \u0026minus;\u0026thinsp;26 to high rate of 163 (m), entire study are falls under the zone of low and medium due to coastal zone and western side part of study area cover most high range of elevation part. Due to the huge coverage of low elevation is favourable for flood susceptibility most occur near the coast surrounding areas and river bank side and high elevation is not so favourable. So he estimate of flood risk must take elevation into a major account, Low-lying places are at risk because floods can occur when water levels rise during periods of heavy rain. The Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA shows the elevation map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003eSlope: Slope for Chennai metropolitan area are under the slope ranges from 39.42 to 0, most of the region covers by low slope category because of Nearby coastal zone and some of the zone shows high slope value in the part of study area, 0 slope values represents water body. As like as elevation slope also plays a major role for flood analysis, because nearly and very low slope prefers flood vulnerable due to insufficient of water flow velocity is very low. This causes water to quickly build up, which inundates low-lying areas and causes floods. The Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB shows the slope map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003eAspect: Aspect for Chennai metropolitan area generate from flat \u0026minus;\u0026thinsp;1 followed by north 0\u0026ndash;22.5, northeast 22.5\u0026ndash;67.5, east 67.5\u0026ndash;112.5, southeast 112.5\u0026ndash;157.5, south 157.5\u0026ndash;202.5, southwest 202.5\u0026ndash;247.5, west 247.5\u0026ndash;292.5 and northwest shows 292.5\u0026ndash;360. Over all study area facing towards eastern direction due to flow river ends in ocean and some part of study area shows some other direction too. Riverbanks and drainage systems are overloaded by heavy rain, which causes water to quickly accumulate. Deforestation and urbanisation together worsen runoff, lowering natural absorption. Rainfall frequency and intensity increase due to climate change, which also intensifies weather patterns. In low-lying places, these elements combine to produce destructive flooding. These can be identity the flow direction of above mentioned by using aspect analysis. The Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC shows the aspect map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003eRainfall: Rainfall for Chennai metropolitan area are under the rainfall ranges from 249 to 544 mm, most of the region covers by high and very high rainfall pattern due to coastal processes, very low annual rainfall pattern receiving from north side and south west direction zone. Low-lying areas are flooded by excessive rain, which also overwhelms drainage systems and swells waterways. Flooding occurs when water builds up quickly, engulfing crops, buildings, and roadways. Rescue operations and evacuations that follow emphasise the devastation caused by natural flood. The Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD shows the Rainfall map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTRI: Chennai metropolitan area are under the roughness surface ranges from 0.1111 to 0.8888, most of the region covers by low and medium roughness terrain due to coastal zone, very low roughness surface shows from water body side. Due to the heavy rainfall receiving Chennai metropolitan area, as high topographic roughness index implies uneven terrain, which hinders water movement. Flooding is brought on by water accumulating in depressions and overtaxing drainage systems. Sharper water velocity and a higher risk of flooding are caused by steeper slopes, which intensify runoff. The Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA shows the TRI map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003eTWI: Chennai metropolitan area is under the wetness surface ranges from \u0026minus;\u0026thinsp;7.7246 to 11.5742, most of the region covers by low wetness terrain due to huge amount of urbanization around the study area, very low wetness surface shows from water body side. It evaluates landscape features that influence water accumulation and might cause flooding when high values indicate areas vulnerable to water accumulation. Due to inadequate drainage systems and elevated soil saturation, these areas are more susceptible to excess water during periods of heavy precipitation, raising the danger of flooding. The Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB shows the TWI map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003eSPI: Chennai metropolitan area are falls under very low stream power index zone, but in this study river part showing maximum amount of SPI at the rate of 14.5003. and other parts of study shows very low range up to 0. Increased water flow and sediment transport are indicated by high stream power index values, overwhelming natural channels. Flooding results from excessive rainfall or quick snowmelt. Deposition and erosion change the landscape and increase the chance of infrastructure damage. The Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC shows the SPI map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSTI: Chennai metropolitan area are cover very low amount of sediments transportations, along the side of water bodies shows high and very high sediments transportations zones. STI falls under the ranges of 0 to 339.016. When there has been a lot of rain, rivers that have a high Sediment Transport Index, which indicates increased sediment movement, may become blocked and have less channel capacity. By reducing water flow and increasing the possibility of overflow and flooding in downstream areas, this increases the risk of flooding. The Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD shows the STI map for Chennai metropolitan area.\u003c/p\u003e \u003cp\u003eLand use/Land cover: Over all study area are been analysis the Landsat 8 OLI/TIRS data consists of 30m resolution and though Artificial neural network and classified their output in the form NRSC level 1 classification and shows 5 major classes which includes like, Water Body, Agriculture Land, Barren land, Built-up land and grass land. On the study area covers mostly covered by built-up land. Rapid urbanization due to the modification of others lands into built up land resulting more vulnerable zones for flood susceptibility zonation. Urbanisation and deforestation, for example, can disturb natural drainage patterns, lowering water absorption and raising runoff. These Land Use and Land Cover (LULC) changes. By clogging up local rivers and creating fast water collection during periods of high rainfall, this change to natural landscapes can raise the danger of flooding. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA shows the land use and land cover map for the study area.\u003c/p\u003e \u003cp\u003eDistance to the river: In the study area shows major river flow continuously in the boundary, buffer zone is considered in order of 1000m interval from the major river flow, 5 major classes are classified buffer zones are distance from the road shows nearby river zone shows most vulnerable for flood susceptibility zone. Far distance from the river zone shows very less vulnerable zones. North and southwest zones are along falls under the above 5000m distance from the river side. Flooding is considerably exacerbated by river proximity. The river's water level rises during periods of intense precipitation, overflowing its banks and flooding neighbouring areas. Infrastructure and human populations nearby are more at risk. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB shows the distance from the river map for the study area.\u003c/p\u003e \u003cp\u003eLithology: Chennai metropolitan area falls under the lithology class are boulders beds, charnockite, fluvial, garnet, marine, pyroxene granulite, quartz \u0026ndash; conglomerate shingles and shale with limestone. Over all the study area cover by the lithology features are marine and fluvial features. Some of the zones covers by charnockite and shale with limestone. When there is insufficient lithology, such as impermeable rock strata, there may be surface runoff during periods of excessive rainfall. This runoff overwhelms drainage systems, limiting adequate groundwater penetration and resulting in floods. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC shows the lithology map for the study area.\u003c/p\u003e \u003cp\u003eSoil: Chennai metropolitan area falls under the soil class are calcareous clay soil, calcareous cracking clay soil, cracking clay soil, gravelly clay soil and sandy soil. Over all the study area cover by the soil features are cracking clay soil and gravelly clay soil. Some of the zones covers by calcareous clay soil. Because compacted soil is less capable of absorbing water, heavy rains increase surface runoff. Because the runoff overwhelms drainage systems, water builds up and travels overland rather than penetrating the impervious soil, resulting in floods. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD shows the soil map for the study area.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eFactors for flood susceptibility by InGR and multicollinearity tests\u003c/h2\u003e \u003cp\u003eUsing a 12-fold cross-validation procedure, the analyses' results are given in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, where each parameter's InGR values were calculated. The most significant impact factors, as shown by the InGR data, are the high InGR value LULC (0.48), DR (0.54), elevation (0.61), and slope (0.58). The total roughness indicator TRI (0.45), the silt transport index STI (0.42), and the stream power index SPI (0.44) all demonstrate significant importance. The total wetness index (TWI), rainfall (0.52), lithology (0.35), and soil (0.38), all show little significance. The fact that the aspect factor has a value of InGR\u0026thinsp;=\u0026thinsp;0.00, showing that it has minimal bearing on FSM prediction, is crucial. The results of the multi-collinearity test of the flood conditioning components and how it affects the capacity to foresee FSMs are shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The Pearson's correlation coefficient for the association between floods and their potential sources is shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003epresents the results of a multi-collinearity test (VIF) examination of the flood conditioning components and influences the ability to anticipate FSMs.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"13\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameters\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eElevation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSlope\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAspect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eDR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRainfall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eLULC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eTRI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eTWI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSPI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eSTI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003eSoil\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003eLithology\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMulticollinearity test (VIF)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.948\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.328\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.930\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.186\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.046\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.043\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e2.414\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e2.470\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e2.165\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e2.149\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e3.520\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e1.950\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eshows the Pearson's correlation coefficient for the relationship between floods and its possible causes.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"13\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameters\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eElevation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSlope\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAspect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eD R\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRainfall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eLULC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eTRI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eTWI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSPI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eSTI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c12\"\u003e \u003cp\u003eSoil\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c13\"\u003e \u003cp\u003eLithology\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCorrelation (Pearson)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.985\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.875\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.692\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.524\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.425\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.324\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.082\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c12\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c13\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eThe analysis of flood susceptible models\u003c/h2\u003e \u003cp\u003eFor each pixel in the Chennai metropolitan region, maps of flood vulnerability were computed using the two machine learning models, such as ANN and RF. Based on the aforementioned experiment findings, it has been determined that the random forest model is the highest-performing benchmark model for the geographical datasets. For reclassifying the flood sensitive models, ArcGIS 10.8 provides a variety of techniques. Quantile and natural break approaches are two that have been extensively described in the literature that is currently accessible on flood susceptibility investigations. The quantile is a well-known reclassification method that has mostly been used to reclassify flood-prone maps since it can produce better results than other reclassification methods. This is exactly the quantile reclassifying strategy was used for the current investigation. Five categories, including extremely low, low, moderate, high, and very high, were created from the reclassification of the flood susceptibility models. Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e shows the flood susceptibility maps from 2 different derived models (ANN \u0026amp; RF). Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e shows Two models that are vulnerable to flooding had their accuracy evaluated using various error measures for training and testing data. Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e shows the results of the Friedman test for models that are prone to flooding in ANN and RF with a p value\u0026thinsp;=\u0026thinsp;0.05.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe results of flood susceptibility model show derived from 2 different models like ANN and RF. In results of ANN represents entire coastal zones are falls under the very high flood susceptibility zone, nearby covers by high and moderate flood susceptibility zones. North side and southwest zones facing very low and low category flood susceptibility. In the model of RF shows half of the coastal zones are covered by very high and nearby surrounding areas are high flood susceptibility. Over all high area covered by moderate zones and little bit coverage near the southwest and northeast zones are falls under low and very low flood susceptibility zones. On the comparability analysis of both models ANN and RF models shows most of the zones are falls under moderate, high and very high flood susceptibility zones, especially Chennai, T. Nagar, Adyar, Lakshmi Nagar, Thoraipakkam, Sholinganallur and Tiruvottiyur zones are falls under very high flood susceptibility zones. Padianallur, Ennore, Avadi and Porur are falls under the zones of high flood susceptibility but in ANN method shows very high flood susceptibility. Karanodai, New vellanur, Thirumudivakkam, chromepet, Tambaram and Perungalathur are zones falls under low flood susceptibility zones but in ANN model shows very low flood susceptibility zones. Over all comparability shows more moderate zones occurs in RF model and less amount of area occupancy in ANN model. Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows the area classification of ANN model and RF model flood susceptibility zones, along the Chennai metropolitan area.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTwo models that are vulnerable to flooding had their accuracy evaluated using various error measures for training and testing data.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eS.NO\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eANN Model\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRF Model\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRMSE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.256\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.298\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMAE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.175\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.196\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.856\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.924\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe results of the Friedman test for models that are prone to flooding in ANN and RF with a p value\u0026thinsp;=\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS.NO\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMean Ranks\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eChi-Square\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSignificance\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eANN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e198.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e368.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFlood susceptibility zones area classification by using ANN and RF\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eS.NO\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFlood Susceptibility Zones\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eANN (Area Sq.km)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRF (Area Sq.km)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery Low\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModerate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery High\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e113\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe Chennai Metropolitan Area (CMA) has already had various flooding incidents, with significant examples in recent years. Old Mahabalipuram Road (OMR), Anna Nagar, Tambaram, Guindy, Velachery, Mogappair, and other places in and around Chennai have historically been prone to flooding. These flooding episodes are primarily the result of a confluence of circumstances, including high rainfall, insufficient stormwater drainage infrastructure, encroachment on water bodies, and urban growth without effective land-use planning. No data values factors are taken in to model that\u0026rsquo;s particular zones represent very low and low vulnerability zones. The overall comparatively ANN model \u0026amp; RF model shows more number area occupancy in the ANN model than the RF model, So more approximately accurate flood model susceptibility for Chennai metropolitan area. Major location like Chennai, T. Nagar, Adyar, Thoraipakkam and Lakshmi nagar falls under very vulnerability condition zone and some of the zones like Karanodai, chromepet, Tambaram and Perungalathur falls under low vulnerability zones.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAn accurate assessment of flood susceptibility is a key step for the protection of people and the creation of practical and effective mitigation solutions (Sahana et al., 2019). As a result, the main necessity is to control flooding to avoid it and lessen the damage it causes (Youssef et al, 2011; Zaharia et al, 2020). As a result, the FSM has evolved into the global management pillar (Sahana et al., 2019). In order to get findings with a very high degree of precision and accuracy, researchers always strive to employ innovative and reliable procedures (Youssef et al., 2011). For this reason, we also tried to employ several cutting-edge models, such ANN and RF algorithms, to create maps of Chennai's flood vulnerability. Modelling the region's flood vulnerability is essential since nationwide flooding has been a common occurrence for a very long time.\u003c/p\u003e \u003cp\u003eAmong the applied approaches, ANN and RF have been widely used for landslide susceptibility modelling (Xu, 2013; Wu et al, 2020; Luo et al, 2019; Hong et al, 2019; Ghasemain et al, 2020;Fang et al, 2020; Chen et al, 2014; ), flood susceptibility modelling (Islam et al, 2021; Bui et al, 2019; Dano et al, 2019; Falah et al, 2019; Kourgialas et al, 2011; Nikolaos et al, 2019; Pradhan, 2010), and forest fire modelling (Vafakhah et al, 2020; Luo et al, 2019), but scarce studies applied random subspace and Dagging model for different natural hazards prediction including flood susceptibility. Additionally, the information gain ratio was used to evaluate how the flood conditioning settings affected the situation. Since ANN and RF performed better in the current study for both training and testing datasets, it can be concluded that these models are also very competent of modelling flood vulnerability. From the model results validation taken place based on the 2 model\u0026rsquo;s outcomes similar vulnerability shows the validation process of low and high vulnerability zones. Therefore, in order to anticipate such natural disasters with great precision, it is strongly advised to use the RF model.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis research introduces two novel hybrid ensemble models, ANN and RF, integrated with machine learning models for flood risk mapping in the Chennai metropolitan area. Twelve variables, including elevation, slope, lithology, aspect, SPI, TWI, STI, land use and land cover (LULC), rainfall, distance from river, TWI, and soil types, were utilized to identify 280 flooding spots. Feasibility of flood conditioning parameters was assessed using VIF, IGR, and Pearson's correlation matrix, with feature selection techniques such as information gain ratio employed to gauge their impact. The RF and ANN models exhibited superior flexibility and predictive capacity, warranting their use in flood susceptibility mapping. However, a combination of both models was utilized for its reliability. The study concluded that RF is one of the most effective methods for predicting flood susceptibility.\u003c/p\u003e \u003cp\u003eThe Chennai Metropolitan area spans 113 sq.km, with 35 to 20 sq.km considered particularly vulnerable to floods. Limitations in modeling temporal changes, such as SPI and land use, necessitate future studies with temporal datasets. Sensitivity analysis on significant variables can further enhance model performance. The RF model offers advantages over other models due to its reduced parameter count, optimization power, and rapid convergence, facilitating efficient flash flood vulnerability mapping.\u003c/p\u003e \u003cp\u003eAddressing flood threats in Chennai requires upgraded drainage systems, strict building standards in flood-prone areas, technology-driven early warning systems involving the community, promotion of green spaces, and construction of flood-resistant infrastructure. Community education on flood preparedness and disaster response is crucial. Effective flood control necessitates collaboration between governmental and non-governmental organizations and local communities. Resilience can be enhanced through drills, resource allocation, and ongoing monitoring of flood-prone regions. Sustainable flood control demands a comprehensive strategy integrating infrastructure development, community involvement, and emergency readiness.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eV. Nivedita - Manuscript writing, analysis, data processing, review and editing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eS. Sabarunisha Begum - Data processing, writing, review and editing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eU. Sakthi - Analysis, editing and corrections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eV. Sellam - Data processing, analysis and editing.\u003c/p\u003e\n\u003cp\u003eC. Navaneethan - Analysis, editing and corrections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMohamed Yacin Sikkandar - Analysis, editing and corrections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eT. Jayasankar - Analysis, editing and corrections.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eS. Vivek - Data processing, analysis and editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e: The authors would like to express their gratitude to the anonymous referees and the editor/associate editor for their constructive comments and valuable suggestions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e This work does not have any funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e All data generated or analysed during this study are included in the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e The authors declare no conflict of interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e Not applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbu Reza Md Towfiqul Islam, Swapan Talukdar, Susanta Mahato, Sonali Kundu, Kutub Uddin Eibek, Quoc Bao Pham, Alban Kuriqi, Nguyen Thi Thuy Linh (2021), Flood susceptibility modelling using advanced ensemble machine learning models https://doi.org/10.1016/j.gsf.2020.09.006.\u003c/li\u003e\n\u003cli\u003eAkter, J., Das, L., Meher, J.K., Deb, A., (2019). Evaluation of different large-scale predictor-based statistical downscaling models in simulating zone-wise monsoon precipitation over India. Int. J. Climatol. 39, 465\u0026ndash;482.\u003c/li\u003e\n\u003cli\u003eAlexander, K., Hettiarachchi, S., Ou, Y., Sharma, A., (2019). Can integrated green spaces and storage facilities absorb the increased risk of flooding due to climate change in developed urban environments? J. Hydrol. 579, 12420\u003c/li\u003e\n\u003cli\u003eBui, D.T., Tsangaratos, P., Ngo, P.T.T., Pham, T.D., Pham, B.T., (2019) b. Flash flood susceptibility modeling using an optimized fuzzy rule-based feature selection technique and tree-based ensemble methods. Sci. Total Environ. 668, 1038\u0026ndash;1054.\u003c/li\u003e\n\u003cli\u003eCao, Q., Mehran, A., Ralph, F.M., Lettenmaier, D.P, (2019). The role of hydrological initial conditions on Atmospheric River floods in the Russian River basin. J. Hydrometeorol.\u003c/li\u003e\n\u003cli\u003eChen, W., Li, W., Hou, E., Zhao, Z., Deng, N., Bai, H., Wang, D., (2014). Landslide susceptibility mapping based on GIS and information value model for the Chencang District of Baoji, China. Arab. J. Geosci. 7 (11), 4499\u0026ndash;4511\u003c/li\u003e\n\u003cli\u003eDano, U.L., Balogun, A.L., Matori, A.N., Wan Yusouf, K., Abubakar, I.R., Said Mohamed, M.A,Aina, Y.A., Pradhan, B.,(2019). Flood susceptibility mapping using GIS-based analytic network process: A case study of Perlis, Malaysia. Water 11 (3), 615.\u003c/li\u003e\n\u003cli\u003eDeng, S., Liu, S., Mo, X., (2020). Assessment of Three Common Methods for Estimating Terrestrial Water Storage Change with Three Reanalysis Datasets. J. Clim. 33 (2),511\u0026ndash;525\u003c/li\u003e\n\u003cli\u003eDouglas, E.M., Vogel, R.M., Kroll, C.N., (2000). Trends in floods and low flows in the United States: impact of spatial correlation. J. Hydrol. 240 (1\u0026ndash;2), 90\u0026ndash;105.\u003c/li\u003e\n\u003cli\u003eDuque, E.L., Aquino, P.T, (2020). Anthropometric Analysis in Automotive Manual Transmission Gearshift Quality Perception. CTI SYMPOSIUM 2018. Springer Vieweg, Berlin, Heidelberg, pp. 97\u0026ndash;109.\u003c/li\u003e\n\u003cli\u003eFalah, F., Rahmati, O., Rostami, M., Ahmadi sharaf, E., Dimakopoulos, I.N., Pourghasemi, H.R., (2019). Artificial neural networks for flood susceptibility mapping in data-scarce urban areas. Spatial modeling in GIS and R for Earth and Environmental Sciences. Elsevier, pp. 323\u0026ndash;336 139.\u003c/li\u003e\n\u003cli\u003eFang, Z., Wang, Y., Peng, L., Hong, H., (2020). Integration of convolutional neural network and conventional machine learning classifiers for landslide susceptibility mapping. Comput. Geosci., 104470\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-Ruiz, J.M., Reg\u0026uuml;\u0026eacute;s, D., Alvera, B., Lana-Renault, N., Serrano-Muela, P., Nadal Romero, E., ... Arn\u0026aacute;ez, J., (2008). Flood generation and sediment transport in experimental catchments affected by land use changes in the central Pyrenees. J. Hydrol.356 (1-2), 245\u0026ndash;260.\u003c/li\u003e\n\u003cli\u003eGetahun, Y.S., Gebre, S.L., (2015). Flood hazard assessment and mapping of flood inundation area of the Awash River Basin in Ethiopia using GIS and HEC-GeoRAS/HEC-RAS model. J. Civil Environ. Eng. 5 (4), 1.\u003c/li\u003e\n\u003cli\u003eGhasemain, B., Asl, D.T., Pham, B.T., Avand, M., Nguyen, H.D., Janizadeh, S., (2020). Shallow landslide susceptibility mapping: A comparison between classification and regression tree and reduced error pruning tree algorithms. Vietnam J. Earth Sci\u003c/li\u003e\n\u003cli\u003eHong, H., Miao, Y., Liu, J., Zhu, A.X., (2019). Exploring the effects of the design and quantity of absence data on the performance of random forest-based landslide susceptibility mapping. Catena 176, 45\u0026ndash;64.\u003c/li\u003e\n\u003cli\u003eHong, H., Shahabi, H., Shirzadi, A., Chen, W., Chapi, K., Ahmad, B.B., ... Bui, D.T., (2019). Landslide susceptibility assessment at the Wuning area, China: A comparison between multi-criteria decision making, bivariate statistical and machine learning methods. Nat. Hazards 96 (1), 173\u0026ndash;212.\u003c/li\u003e\n\u003cli\u003eHosni, M., Abnane, I., Idri, A., de Gea, J. M. C., \u0026amp; Alem\u0026aacute;n, J. L. F. (2019). Reviewing Ensemble Classification Methods in Breast Cancer. Comput. Methods Prog. Biomed.\u003c/li\u003e\n\u003cli\u003eHou, G., Shao, Q., Zou, B., Dai, L., Zhang, Z., Mu, Z., Zhai, J., 2020. A Novel Underwater Si-multaneous Localization and Mapping Online Algorithm Based on Neural Network. ISPRS Int. J. Geo Inf. 9 (1), 5.\u003c/li\u003e\n\u003cli\u003eKourgialas, N.N., Karatzas, G.P., (2011). Flood management and a GIS modelling method to assess flood-hazard areas\u0026mdash;a case study. Hydrol. Sci. J. 56 (2), 212\u0026ndash;225.\u003c/li\u003e\n\u003cli\u003eKumar, R., Nayak, P.C., Patra, J.P., Mani, P., (2019). Chapter-10-Impact of climate change on floods. National Institute of Hydrology.\u003c/li\u003e\n\u003cli\u003eKuncheva, L.I., Rodr\u0026iacute;guez, J.J., Plumpton, C.O., Linden, D.E., Johnston, S.J., (2010). Random subspace ensembles for fMRI classification. IEEE Trans. Med. Imaging 29 (2), 531\u0026ndash;542.\u003c/li\u003e\n\u003cli\u003eKuriqi, A., Ardi\u0026ccedil;lioǧlu, M.,(2018). Investigation of hydraulic regime at middle part of the Loire River in context of floods and low flow events. Pollack Periodica 13 (1),145\u0026ndash;156.\u003c/li\u003e\n\u003cli\u003eLuo, X., Lin, F., Chen, Y., Zhu, S., Xu, Z., Huo, Z., ... Peng, J., (2019). Coupling logistic model tree and random subspace to predict the landslide susceptibility areas with considering the uncertainty of environmental features. Sci. Rep. 9 (1), 1\u0026ndash;13.\u003c/li\u003e\n\u003cli\u003eMa, J., Ding, Y., Cheng, J.C., Jiang, F., Tan, Y., Gan, V.J., Wan, Z.,(2020). Identification of high impact factors of air quality on a national scale using big data and machine learning techniques. J. Clean. Prod. 244, 118955.\u003c/li\u003e\n\u003cli\u003eMasks, D.L.G., Starr, N.B., Brady, M.A., Gaylord, N.M., Driessnack, M., Duderstadt, K., (2019). Burns' Pediatric Primary Care (7th Edition). Elsevier\u003c/li\u003e\n\u003cli\u003eNikolaos, S., Kleomenis, K., Elias, D., Panagiotis, S., Panagiota, L., Vagelis, P., Christos, C., (2019). A Robust Remote Sensing\u0026ndash;Spatial Modeling\u0026ndash;Remote Sensing (RMR) Approach for Flood Hazard Assessment. Spatial Modeling in GIS and R for Earth and Environmental Sciences. Elsevier, pp. 391\u0026ndash;410.\u003c/li\u003e\n\u003cli\u003eOeurng, C., Sauvage, S., S\u0026aacute;nchez-P\u0026eacute;rez, J.M., (2011). Assessment of hydrology, sediment and particulate organic carbon yield in a large agricultural catchment using the SWAT model. J. Hydrol. 401 (3-4), 145\u0026ndash;153.\u003c/li\u003e\n\u003cli\u003ePradhan, B., (2010). Flood susceptible mapping and risk area delineation using logistic regression, GIS and remote sensing. J. Spat. Hydrol. 9, 2.\u003c/li\u003e\n\u003cli\u003ePrăvălie, R., Costache, R., (2013). The vulnerability of the territorial-administrative units to the hydrological phenomena of risk (flash-floods). Case study: the subcarpathian sector of Buzău catchment. Analele Universității din Oradea\u0026ndash;Seria Geografie 23 (1),91\u0026ndash;98.\u003c/li\u003e\n\u003cli\u003eRumelhart, D.E., Hinton, G.E., McClelland, J.L., (1986). A general framework for parallel distributed processing. Parallel distributed processing: Explorations in the microstructure of cognition. 1, p. 26 45-76.\u003c/li\u003e\n\u003cli\u003eSahana, M., Patel, P.P., (2019). A comparison of frequency ratio and fuzzy logic models for flood susceptibility assessment of the lower Kosi River Basin in India. Environ.Earth Sci. 78 (10), 289\u003c/li\u003e\n\u003cli\u003eTorcivia, C.E.G., L\u0026oacute;pez, N.N.R., (2020). Preliminary Morphometric Analysis: R\u0026iacute;o Talacasto Basin, Central Precordillera of San Juan, Argentina. Advances in Geomorphology and Quaternary Studies in Argentina. Springer, Cham, pp. 158\u0026ndash;168.\u003c/li\u003e\n\u003cli\u003eUthayakumar, J., Metawa, N., Shankar, K., Lakshmanaprabu, S.K., (2020). Financial crisis pre-diction model using ant colony optimization. Int. J. Inf. Manag. 50, 538\u0026ndash;556.\u003c/li\u003e\n\u003cli\u003eVafakhah, M., Loor, S.M.H., Pourghasemi, H., Katebikord, A., (2020). Comparing performance of random forest and adaptive neuro-fuzzy inference system data mining models for flood susceptibility mapping. Arab. J. Geosci. 13, 417.\u003c/li\u003e\n\u003cli\u003eWu, Y., Ke, Y., Chen, Z., Liang, S., Zhao, H., Hong, H., (2020) b. Application of alternating decision tree with AdaBoost and random subspace ensembles for landslide susceptibility mapping. Catena 187, 104396.\u003c/li\u003e\n\u003cli\u003eXie, H., Dong, J., Shen, Z., Chen, L., Lai, X., Qiu, J., ... Chen, X., (2019). Intra-and inter-event characteristics and controlling factors of agricultural nonpoint source pollution under different types of rainfall-runoff events. Catena 182, 104105.\u003c/li\u003e\n\u003cli\u003eXu, C., (2013). Assessment of earthquake-triggered landslide susceptibility based on expert knowledge and information value methods: a case study of the 20 April 2013 Lushan, China Mw6. 6 earthquakes. Disaster Advances 6 (13), 119\u0026ndash;130\u003c/li\u003e\n\u003cli\u003eYoussef, A.M., Pradhan, B., Hassan, A.M., (2011). Flash flood risk estimation along the St. Katherine Road, southern Sinai, Egypt using GIS based morphometry and satellite imagery. Environ. Earth Sci. 62 (3), 611\u0026ndash;623.\u003c/li\u003e\n\u003cli\u003eZaharia, L., Ioana-Toroimac, G., Perju, E.R.,(2020). Hydrological Impacts of Climate Changes in Romania. Water Resources Management in Romania. Springer, Cham, pp. 309\u0026ndash;351\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Flood, Environment, ANN Model, RF Model, Algorithms, Machine Learning, Vulnerability, Risk Assessment, Mitigation Plans","lastPublishedDoi":"10.21203/rs.3.rs-4194276/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4194276/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eFloods are very destructive natural catastrophes that significantly harm people, the environment, and structures. Due to their dynamic nature, flash flood-prone locations are difficult to predict. We used cutting-edge machine learning algorithms for early flash flood detection to overcome this. To evaluate flood vulnerability in Chennai, Tamil Nadu, India, we employed and evaluated two combined ensemble models: the Artificial Neural Network (ANN) and Random Forest (RF). To put these models into practice, we built a GIS framework out of 280 historical flooding sites and twelve flood-related parameters. Using information gain ratio and multicollinearity diagnostic tests, we evaluated the association between flood occurrences and pertinent factors. Statistical criteria like \"Freidman\" were used to compare the prediction abilities of ANN and RF models. Both ANN and RF models performed better than expected when simulating flood susceptibility. In order to reduce flood-related hazards and create efficient mitigation plans, state and local authorities, as well as policymakers, will benefit from the study's findings and methodology. Flood susceptibility percentages were 18% very low, 16% low, 13% moderate, 22% high, and 31% extremely very high, according to the research of ANN. While this was going on, RF revealed that the likelihood of flooding was 7% very low, 11% low, 34% moderate, 31% high, and 18% extremely very high. Maintaining a strong drainage system, using regulated building techniques in sensitive regions, setting up early warning systems, creating resilient infrastructure, and educating people are all necessary to reduce floods in Chennai. For resource allocation, disaster response, and readiness exercises, effective coordination between government agencies, non-governmental organizations (NGOs), and local populations is essential. Chennai's resistance to floods depends on a multi-pronged approach that includes infrastructure improvement, pro-active planning, and public awareness.\u003c/p\u003e","manuscriptTitle":"Assessment of flood vulnerability in a coastal metropolitan city for sustainable environmental using Machine Learning methods","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-12 07:40:55","doi":"10.21203/rs.3.rs-4194276/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2df3d3b8-7793-48b0-b681-8940970c1b51","owner":[],"postedDate":"September 12th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-10-13T22:41:27+00:00","versionOfRecord":[],"versionCreatedAt":"2024-09-12 07:40:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4194276","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4194276","identity":"rs-4194276","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}