A GIS-enabled AHP approach for mapping urban flood susceptibility in Bhubaneswar city

Synergistic assessment of multi-scenario urban waterlogging through data-driven decoupling analysis in high-density urban areas: A case study in Shenzhen, China

Resilient landscape pattern for reducing coastal flood susceptibility

Due to the complex interaction of urban and mountainous floods, assessing flood susceptibility in mountainous urban areas presents a challenging task in environmental research and risk analysis. Data-driven machine learning methods can evaluate flood susceptibility in mountainous urban areas lacking essential hydrological data, utilizing remote sensing data and limited historical inundation records. In this study, two ensemble learning algorithms, Random Forest (RF) and XGBoost, were adopted to assess the flood susceptibility of Kunming, a typical mountainous urban area prone to severe flood disasters. A flood inventory was created using flood observations from 2018 to 2022. The spatial database included 10 explanatory factors, encompassing climatic, geomorphic, and anthropogenic factors. Artificial Neural Network (ANN) and Support Vector Machine (SVM) were selected for model comparison. To minimize the influence of expert opinions on model training, this study employed a strategy of uniformly random sampling in historically non-flooded areas for negative sample selection. The results demonstrated that (1) ensemble learning algorithms offer higher accuracy than other machine learning methods, with RF achieving the highest accuracy, evidenced by an area under the curve (AUC) of 0.87, followed by XGBoost at 0.84, surpassing both ANN (0.83) and SVM (0.82); (2) the interpretability of ensemble learning highlighted the differences in the potential distribution of the training data’s positive and negative samples. Feature importance in ensemble learning can be utilized to minimize human bias in the collection of flooded-site samples, more targeted flood susceptibility maps of the study area’s road network were obtained; and (3) ensemble learning algorithms exhibited greater stability and robustness in datasets with varied negative samples, as evidenced by their performance in F1-Score, Kappa, and AUC metrics. This paper further substantiates the superiority of ensemble learning in flood susceptibility assessment tasks from the perspectives of accuracy, interpretability, and robustness, enhances the understanding of the impact of negative samples on such assessments, and optimizes the specific process for urban flood susceptibility assessment using data-driven methods.

Extracting historical flood locations from news media data by the named entity recognition (NER) model to assess urban flood susceptibility

Flood susceptibility assessment for identifying flood-prone areas plays a significant role in flood hazard mitigation. Machine learning is an optional assessment method because of its high objectivity and computational efficiency, but how to get enough and accurate information of historical flood locations to train the machine learning models has been a key problem. In recent years, news media data from both news websites and social media authentication accounts has emerged as a promising source for natural science studies. However, the application of news media data in urban flood susceptibility assessment is still inadequate. This study proposed an approach of three tasks to use news media data on this topic. Firstly, flood locations were extracted from news media data based on a named entity recognition (NER) model. Then, a frequency or distance-based data quality control method was employed to improve the representativeness of the extracted flooded locations. Finally, flood conditioning factors with information of historical flood locations were input into a Support Vector Machine (SVM) model for flood susceptibility assessment. We took the central city of Dalian, China, as a case study. The results show that there was no significant difference of a T-test between the distributions of most flood conditioning factors at the flood locations from the news media data and the official planning report. In the obtained flood susceptibility map, the high flood susceptibility areas got a recall of 90% compared with the high flood hazard areas in the planning report. Performing data quality control in the frequency-based method can improve the precision of the flood susceptibility map by up to 5%, while the distance-based method is ineffective. This study provides an example and offers the value of applying new data sources and modern deep learning techniques for urban flood management. 

Urban flooding (often referred to as water logging) is defined as the submergence of normally dry city areas by a considerable volume of water caused by heavy precipitation or overflowing of water bodies. Flood susceptibility modelling, by combining the effects of natural and human factors, determines the sensitivity of the space to flood hazard. Urban flood modelling has gained attention recently and since the incidence of urban floods has increased rapidly, due attention needs to be given to the urban flood studies. In this case study, urban flood susceptibility modelling of Srinagar City, Jammu and Kashmir, India, using Fuzzy MLPNN, has been carried out in Geographic Information System (GIS) environment. Fuzzy MLPNN is a simple and straightforward approach that unifies the complexity of the phenomenon of urban flooding by integrating fuzzy mathematics and machine learning to build a predictive model for the analysis of urban flood susceptibility using geospatial data. Eight flood conditioning factors (elevation, slope, profile curvature, plan curvature, geology, distance from natural streams, MFI and LULC) were used as independent variables along with urban flood locations as the dependent variable. A precursory FSZ map of Srinagar City was created using the frequency ratio technique, and non-flooded locations were accordingly determined. The developed model reveals the susceptibility of each and every pixel (12.5 × 12.5 m area) in the study area. The FSI, illustrated by the FSZ Map of Srinagar, demonstrates considerable susceptibility of the city to urban flood hazard. The dominant influence of spatiality of precipitation and water bodies is indicated by the conclusion that highly susceptible regions of the city are those where MFI is high and proximity to natural drainage is low. The FSZ map was validated using Area under the ROC Curve (AUC) Analysis, which substantiates the efficiency of the Fuzzy MLPNN model. With 0.931 and 0.922 AUC values, the success rate and predictive performance of the FSZ map come out to be excellent, respectively.KeywordsFlood susceptibility modellingUrban floodFuzzy MLPNNSrinagar city

The application of integrating comprehensive evaluation and clustering algorithms weighted by maximal information coefficient for urban flood susceptibility

Assessing urban flooding risk in response to climate change and urbanization based on shared socio-economic pathways

Urban flooding threatens urban resilience and challenges SDGs 11 and 13. This study assesses urban building flood risk in Guangzhou by integrating flood susceptibility with building function vulnerability. Using a Random Forest (RF) model, it predicts flood susceptibility based on flood records, hydrological, topographical, and anthropogenic features. The Categorical Boosting (CatBoost) model identifies building functions using POI and AOI data. Results reveal significant spatial variations: central districts exhibit higher flood susceptibility, while peripheral areas remain less affected. Over half of the buildings are moderately vulnerable, with only a small fraction highly vulnerable. Based on flood susceptibility and functional vulnerability, Guangzhou is classified into three district types: central urban (Type I), intermediate urban (Type II), and suburban/rural (Type III). The study underscores the need for tailored flood risk management strategies to address these disparities and mitigate climate change-induced water hazards.

Urban flooding is a significant issue in coastal megacity Mumbai, where flood susceptibility is exacerbated by rapid urbanization and intense monsoon rainfall. This study develops a high-resolution flood susceptibility map for the Mumbai Metropolitan Region (MMR), using four machine learning algorithms: Random Forest, Artificial Neural Network, XGBoost, and Gradient Boosting Machine. The models were trained and validated using historical flood occurrence points, with nine conditioning factors: elevation, slope, rainfall, land use and land cover, building density, proximity to coastlines, road networks, and blue space. Models were performed with high accuracy, achieving 0.93 for GBM and XGBoost, 0.92 for RF, and 0.89 for ANN, respectively. The ensemble flood map, created based on the mean of four ML models, revealed that 25.3% of MMR is classified as high or very high flood susceptibility, while 34.3% falls into the low-susceptibility category. SHapley Additive exPlanations (SHAP) analysis showed that elevation, rainfall, and proximity to roads were the most influential predictors. Spatial validation revealed excellent overlap with historical flooding hotspots at Kurla, Chembur, and Sion. These findings provide critical policy insights for integrating flood susceptibility mapping into urban planning frameworks, supporting data-driven resilience strategies and sustainable infrastructure development in rapidly growing coastal megacities, like Mumbai.

Flood susceptibility prediction is complex due to the multifaceted interactions among hydrological, meteorological, and urbanisation factors, further exacerbated by climate change. This study addresses these complexities by investigating flood susceptibility in rapidly urbanising regions prone to extreme weather events, focusing on Gdańsk, Poland. Three popular ML techniques, Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Networks (ANN), were evaluated for handling complex, nonlinear data using a dataset of 265 urban flood episodes. An ensemble filter feature selection (EFFS) approach was introduced to overcome the single-method feature selection limitations, optimising the selection of factors contributing to flood susceptibility. Additionally, the study incorporates explainable artificial intelligence (XAI), namely, the Shapley Additive exPlanations (SHAP) model, to enhance the transparency and interpretability of the modelling results. The models’ performance was evaluated using various statistical measures on a testing dataset. The ANN model demonstrated a superior performance, outperforming the RF and the SVM. SHAP analysis identified rainwater collectors, land surface temperature (LST), digital elevation model (DEM), soil, river buffers, and normalized difference vegetation index (NDVI) as contributors to flood susceptibility, making them more understandable and actionable for stakeholders. The findings highlight the need for tailored flood management strategies, offering a novel approach to urban flood forecasting that emphasises predictive power and model explainability.

BackgroundUrban flood susceptibility evaluation (FSE) can utilize empirical and rational procedures to focus on the urban flood evaluation using physical coefficients and land-use change ratios. The main aim of the present paper was to evaluate a flood susceptibility model in the southern watersheds of Mashhad city, in Iran, for 2010, 2020, and 2030. The construction of the model depended on the utilization of some global datasets to estimate the runoff coefficients of the watersheds, peak flood discharges, and flood susceptibility evaluations.Results and conclusionsBased on the climatic precipitation and urban sprawl variation, our results revealed the mean values of the runoff coefficient (Cr) from 0.50 (2010) to 0.65 (2030), where the highest values of Cr (> 0.70) belonged to the watersheds with real estate cover, soil unit of the Mollisols, and the slope ranges over 5–15%. The averagely cumulative flood discharges were estimated from 2.04 m3/s (2010) to 5.76 m3/s (2030), revealing an increase of the flood susceptibility equal 3.2 times with at least requirement of an outlet cross-section by > 46 m2 in 2030. The ROC curves for the model validity explained AUC values averagely over 0.8, exposing the very good performance of the model and excellent sensitivity.

With global climate change and accelerating urbanization, urban flood is becoming more frequent worldwide. Understanding the urban vulnerability is crucial for making decisions on urban flood control. This study uses urban flood susceptibility (UFS) as an indicator, and comprehensively applies three machine learning models, XGBoost, CatBoost and LightGBM, in the Kangyi area of Ordos City. Combined with the Shapley Additive explanations method, the driving mechanism and spatial heterogeneity of flood susceptibility was explored in the study area. The results show: (1) Model performance comparison: All three models have high accuracy, with XGBoost performing well in overall classification (OA = 0.96) and CatBoost performing well in distinguishing flood/non-flood samples (AUC = 0.85). (2) Multi-model adaptability assessment: The proposed “model-factor-space” framework highlights the sensitivity of XGBoost to urbanization indicators, the ability of CatBoost to capture naturalgeographical elements, and the efficiency of LightGBM in analyzing terrain thresholds. (3) Dynamic thresholds and synergies: Impervious surface density (ISD) is the most critical factor, and when ISD > 0.2, the risk of flooding will continue to increase by 60%. Comprehensive analysis with spatial heterogeneity shows that high-risk areas are mainly affected by ISD, road density (> 10,000 m/km2) and low altitude (< 1300m) in urban built-up areas, while low-to-medium risk areas are sensitive to vegetation coverage (> 7,000) and Dis2Water bodies (> 1,500m). (4) Hierarchical governance strategy: A three-level spatial governance strategy is proposed: in the core area, priority is given to ISD control (< 0.2) and pipe network upgrades; in the transitional area, slope interception and ecological restoration are combined; and in the potential risk area, a multi-scale monitoring and early warning system is established for multi-scale monitoring.

The investigation identifies the geographic focus on the smart city of Bhubaneswar, Odisha, India, within the timeframe (2000–2024). The study identifies seasonal trends, extreme events, and their impacts on climate and human life. The maps are prepared using Python, like scatter plots of mean monthly and yearly rainfall (line, box, bar, and pie charts, etc.). Further, GIS-based maps prepared are the annual Rainfall distribution map of Odisha, the choropleth map for temperature distribution, slope map, relief and contour map of Bhubaneswar smart city, highlighting spatial impacts in Bhubaneswar zones (e.g., Bomi Khal, Naya Palli), etc. The slope, the contour and the population. Slum pockets maps have been prepared from the data of the Government of Odisha, NIC data and Digital Elevation Model. The analysis, supported by Python libraries (pandas, matplotlib, seaborn, geopandas, folium). The study identifies seasonal trends, extreme events and their impacts on climate and human life, focusing on urban flooding, infrastructure, health, agriculture and environmental degradation, with a geoinformatics approach. Recommendations include GIS-driven drainage mapping, rainwater harvesting, resource management, and climate-smart planning, with future research proposed for long-term trends and geospatial adaptation strategies. The extreme events of rainfall cause waterlogged areas, urban flooding and fatalities due to heat strokes. The study discusses the impacts on climate and human life and supports the town planners and developers in climate-resilient planning, allied with SDGs 11 and 13 for sustainable urban dwellings, developing a zero slums concept in the sustainable smart city, Bhubaneswar.

Urban flood susceptibility assessment based on convolutional neural networks