Have coastal embankments reduced flooding in Bangladesh?

Reclassifying historical disasters: From single to multi-hazards

Multi-hazard events, characterized by the simultaneous, cascading, or cumulative occurrence of multiple natural hazards, pose a significant threat to human lives and assets. This is primarily due to the cumulative and cascading effects arising from the interplay of various natural hazards across space and time. However, their identification is challenging, which is attributable to the complex nature of natural hazard interactions and the limited availability of multi-hazard observations. This study presents an approach for identifying multi-hazard events during the past 123 years (1900-2023) using the EM-DAT global disaster database. Leveraging the 'associated hazard' information in EM-DAT, multi-hazard events are detected and assessed in relation to their frequency, impact on human lives and assets, and reporting trends. The interactions between various combinations of natural hazard pairs are explored, reclassifying them into four categories: preconditioned/triggering, multivariate, temporally compounding, and spatially compounding multi-hazard events. The results show, globally, approximately 19 % of the 16,535 disasters recorded in EM-DAT can be classified as multi-hazard events. However, the multi-hazard events recorded in EM-DAT are disproportionately responsible for nearly 59 % of the estimated global economic losses. Conversely, single hazard events resulted in higher fatalities compared to multi-hazard events. The largest proportion of multi-hazard events are associated with floods, storms, and earthquakes. Landslides emerge as the predominant secondary hazards within multi-hazard pairs, primarily triggered by floods, storms, and earthquakes, with the majority of multi-hazard events exhibiting preconditioned/triggering and multivariate characteristics. There is a higher prevalence of multi-hazard events in Asia and North America, whilst temporal overlaps of multiple hazards predominate in Europe. These results can be used to increase the integration of multi-hazard thinking in risk assessments, emergency management response plans and mitigation policies at both national and international levels.

A novel framework for addressing uncertainties in machine learning-based geospatial approaches for flood prediction

Globally, many studies on machine learning (ML)-based flood susceptibility modeling have been carried out in recent years. While majority of those models produce reasonably accurate flood predictions, the outcomes are subject to uncertainty since flood susceptibility models (FSMs) may produce varying spatial predictions. However, there have not been many attempts to address these uncertainties because identifying spatial agreement in flood projections is a complex process. This study presents a framework for reducing spatial disagreement among four standalone and hybridized ML-based FSMs: random forest (RF), k-nearest neighbor (KNN), multilayer perceptron (MLP), and hybridized genetic algorithm-gaussian radial basis function-support vector regression (GA-RBF-SVR). Besides, an optimized model was developed combining the outcomes of those four models. The southwest coastal region of Bangladesh was selected as the case area. A comparable percentage of flood potential area (approximately 60% of the total land areas) was produced by all ML-based models. Despite achieving high prediction accuracy, spatial discrepancy in the model outcomes was observed, with pixel-wise correlation coefficients across different models ranging from 0.62 to 0.91. The optimized model exhibited high prediction accuracy and improved spatial agreement by reducing the number of classification errors. The framework presented in this study might aid in the formulation of risk-based development plans and enhancement of current early warning systems.

Impact of cropping system diversification on productivity and resource use efficiencies of smallholder farmers in south-central Bangladesh: a multi-criteria analysis

Diversification of smallholder rice-based cropping systems has the potential to increase cropping system intensity and boost food security. However, impacts on resource use efficiencies (e.g., nutrients, energy, and labor) remain poorly understood, highlighting the need to quantify synergies and trade-offs among different sustainability indicators under on-farm conditions. In southern coastal Bangladesh, aman season rice is characterized by low inputs and low productivity. We evaluated the farm-level impacts of cropping system intensification (adding irrigated boro season rice) and diversification (adding chili, groundnut, mungbean, or lathyrus) on seven performance indicators (rice equivalent yield, energy efficiency, partial nitrogen productivity, partial potassium productivity, partial greenhouse gas footprint, benefit-cost ratio, and hired labor energy productivity) based on a comprehensive survey of 501 households. Indicators were combined into a multi-criteria performance index, and their scope for improvement was calculated by comparing an individual farmer's performance to top-performing farmers (highest 20%). Results indicate that the baseline system (single-crop aman season rice) was the least productive, while double cropped systems increased rice equivalent yield 72-217%. Despite gains in productivity, higher cropping intensity reduced resource use efficiencies due to higher inputs of fertilizer and energy, which also increased production costs, particularly for boro season rice. However, trade-offs were smaller for diversified systems including legumes, largely owing to lower N fertilizer inputs. Aman season rice had the highest multi-criteria performance index, followed by systems with mungbean and lathyrus, indicating the latter are promising options to boost food production and profitability without compromising sustainability. Large gaps between individual and top-performing farmers existed for each indicator, suggesting significant scope for improvement. By targeting indicators contributing most to the multi-criteria performance index (partial nitrogen productivity, energy efficiency, hired labor energy productivity), results suggest further sustainability gains can be achieved through future field research studies focused on optimizing management within diversified systems.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s13593-022-00795-3.

Integrated modeling of 2D urban surface and 1D sewer hydrodynamic processes and flood risk assessment of people and vehicles

In order to accurately simulate the whole urban flooding processes and assess the flood risks to people and vehicles in floodwaters, a 2D-surface and a 1D-sewer integrated hydrodynamic model was proposed in this study, with the module of flood risk assessment of people and vehicles being included. The proposed model was firstly validated by a dual-drainage laboratory experiment on the flood inundation process over a typical urban street, and the relative importance of model parameters and model uncertainties were evaluated using the GSA-GLUE method. Then the model was applied to simulate an actual urban flooding process that occurred in Glasgow, UK, with the influence of the sewer drainage system on flood inundation processes and hazard degree distributions of people and vehicles being comprehensively discussed. The following conclusions are drawn from this study: (i) The proposed model has a high degree of accuracy with the NSE values of key hydraulic variables greater than 0.8 and the GSA indicates that Manning roughness coefficients for surface and sewer flows, inlet weir and orifice discharge coefficients, are the most relevant parameters to influence the simulated results; (ii) vehicles are vulnerable to larger water depths while human stability is significantly influenced by higher flow velocities, with the overall flood risk of people being less than that of vehicles; and (iii) about 88.7% of the total inflow volume was drained to the sewer network, and the sewer drainage system greatly reduced the flood risks to people and vehicles except the local areas with large inundation water depths, where the sewer drainage increased the local flow velocity leading to higher flood risks especially for people.

Flood damage assessment with Sentinel-1 and Sentinel-2 data after Sardoba dam break with GLCM features and Random Forest method

Accurate mapping and monitoring of flooded areas are immensely required for disaster management purposes, such as for damage assessment and mitigation. In this study, the flood damage mapping performances of two satellite Earth Observation sensors, i.e., European Space Agency's Sentinel-1 (S1) synthetic aperture radar (SAR) and Sentinel-2 (S2) multispectral optical instruments, were evaluated using the Random Forest (RF) supervised classification method and various feature types. The study area was Sardoba Reservoir (Uzbekistan) and its surroundings, in which a disastrous dam failure occurred on May 1, 2020. After the failure of a part of the earthfill dam, a large region with settlements and agricultural areas in Uzbekistan and Kazakhstan was flooded. S1 and S2 cloudless data with a short temporal interval acquired soon after the event were available for the area. Four different data availability scenarios, such as (i) only S1 pre- and post-flood data; (ii) only S2 pre- and post-flood data; (iii) S1 pre- and post-flood and S2 pre-flood data; and (iv) S1 and S2 pre- and post-flood data were evaluated in terms of classification accuracy. In addition to the polarization information of S1 and the intensity values of S2 bands, feature maps produced from these datasets, such as vegetation and water indices, textural information obtained from gray level co-occurrence matrix (GLCM), and the principal component analysis (PCA) bands were employed in the RF method. The results show that the fusion of S1 and S2 data exhibit very high classification accuracy for the flooded areas and can separate the inundated vegetation as well. The use of S2 pre-event data together with the S1 pre- and post-event data is recommended for obtaining high accuracy even when post-event optical data is not available.

Water Level Forecasting Using Spatiotemporal Attention-Based Long Short-Term Memory Network

Bangladesh is in the floodplains of the Ganges, Brahmaputra, and Meghna River delta, crisscrossed by an intricate web of rivers. Although the country is highly prone to flooding, the use of state-of-the-art deep learning models in predicting river water levels to aid flood forecasting is underexplored. Deep learning and attention-based models have shown high potential for accurately forecasting floods over space and time. The present study aims to develop a long short-term memory (LSTM) network and its attention-based architectures to predict flood water levels in the rivers of Bangladesh. The models developed in this study incorporated gauge-based water level data over 7 days for flood prediction at Dhaka and Sylhet stations. This study developed five models: artificial neural network (ANN), LSTM, spatial attention LSTM (SALSTM), temporal attention LSTM (TALSTM), and spatiotemporal attention LSTM (STALSTM). The multiple imputation by chained equations (MICE) method was applied to address missing data in the time series analysis. The results showed that the use of both spatial and temporal attention together increases the predictive performance of the LSTM model, which outperforms other attention-based LSTM models. The STALSTM-based flood forecasting system, developed in this study, could inform flood management plans to accurately predict floods in Bangladesh and elsewhere.

Implications of Flood Risk Reduction Interventions on Community Resilience: An Assessment of Community Perception in Bangladesh

Bangladesh, a flat densely populated country in a dynamic delta, is vulnerable to recurring flood disasters. Various types of structural and non-structural flood risk reduction interventions have been implemented over the years to safeguard the people and assets. In that context, the present study assesses the community perception about the implications of such diverse interventions on community resilience, in three reasonably proximate settlements, with varying characteristics: the Type 1 settlement has a flood protection embankment; the Type 2 settlement has no flood risk reduction intervention, and the Type 3 settlement has non-structural interventions. Through a mixed-method assessment in selected settlements, the study results reveal both positive and negative implications of these interventions on local communities. While the embankment has contributed towards enhancing infrastructural resilience in the Type 1 settlement, it still reportedly does not provide complete flood safety. On the other hand, the non-structural measures are reported to have increased community competencies in the Type 3 settlement, but the long-term sustainability of these traits is uncertain. Furthermore, the study results uncover “connectedness among local communities” as an inherent characteristic in all three locations, whereas flood risk reduction interventions are stated to be partly associated with social tension and the marginalization of certain socio-economic groups.

A dynamic, convenient and accurate method for assessing the flood risk of people and vehicle

With the increase of extreme rainstorm caused by climate change, and the development of urbanization and the improvement of people's living standard, there is an urgent need to draw a dynamic, convenient and accurate flood risk map for different disaster bearing bodies, so as to protect people's lives and properties, as well as improve people's risk awareness and facilitate people's lives. This study mainly researched a method of drawing a dynamic, convenient and accurate flood risk map for people and vehicle. In this study, the surface runoff is simulated by GPU (Graphics Processing Unit) Accelerated Surface Water Flow and Transport model (GAST model), meanwhile, the flood risk of people and vehicle is graded base on the incipient velocity formulas, the most unfavorable principle and grading method, finally, the method is applied in two application cases. The following results are obtained: (1) this method could assess the flood risk of people and vehicle dynamically, conveniently and accurately; (2) the flood risk of people is less than that of vehicle, for the same flood hazard, time and place; (3) the adverse effect of water depth on flood risk for vehicle is greater than that for people. This method of assessing the flood risk map of people and vehicle is of great significance, for flood risk management, land use plan and emergency management department to reduce flood disaster risk.

Unraveling the complexity of human behavior and urbanization on community vulnerability to floods

Floods are among the costliest natural hazards and their consequences are expected to increase further in the future due to urbanization in flood-prone areas. It is essential that policymakers understand the factors governing the dynamics of urbanization to adopt proper disaster risk reduction techniques. Peoples' relocation preferences and their perception of flood risk (collectively called human behavior) are among the most important factors that influence urbanization in flood-prone areas. Current studies focusing on flood risk assessment do not consider the effect of human behavior on urbanization and how it may change the nature of the risk. Moreover, flood mitigation policies are implemented without considering the role of human behavior and how the community will cope with measures such as buyout, land acquisition, and relocation that are often adopted to minimize development in flood-prone regions. Therefore, such policies may either be resisted by the community or result in severe socioeconomic consequences. In this study, we present a new Agent-Based Model (ABM) to investigate the complex interaction between human behavior and urbanization and its role in creating future communities vulnerable to flood events. We identify critical factors in the decisions of households to locate or relocate and adopt policies compatible with human behavior. The results show that when people are informed about the flood risk and proper incentives are provided, the demand for housing within 500-year floodplain may be reduced as much as 15% by 2040 for the case study considered. On the contrary, if people are not informed of the risk, 29% of the housing choices will reside in floodplains. The analyses also demonstrate that neighborhood quality-influenced by accessibility to highways, education facilities, the city center, water bodies, and green spaces, respectively-is the most influential factor in peoples' decisions on where to locate. These results provide new insights that may be used to assist city planners and stakeholders in examining tradeoffs between costs and benefits of future land development in achieving sustainable and resilient cities.

Factors contributing to farm-level productivity and household income generation in coastal Bangladesh's rice-based farming systems

Large changes have taken place in smallholder farming systems in South Asia's coastal areas in recent decades, particularly related to cropping intensity, input availability, climate risks, and off-farm activities. However, few studies have investigated the extent to which these changes have impacted farm-level crop productivity, which is a key driver of food security and poverty in rainfed, low-input, rice-based systems. The objective of this study was to conduct an integrated assessment of variables related to socioeconomic status, farm characteristics, and crop management practices to understand the major factors influencing crop productivity and identify promising leverage points for sustainable development in coastal Bangladesh. Using a panel survey dataset of 32 variables from 502 farm households located within polder (coastal embankment) and outside polder systems during 2005-2015, we employed statistical factor analysis to characterize five independent latent factors named here as Farming Challenges, Economic Status, Crop Management Practices, Asset Endowment, and Farm Characteristics. The factor Farming Challenges explained the most variation among households (31%), with decreases observed over time, specifically households located outside polders. Individual variables contributing to this factor included perceived cyclone severity, household distance to main roads and input-output markets, cropping intensity, and access to extension services. The most important factors for increasing crop productivity on a household and per unit area basis were Asset Endowment and Crop Management Practices, respectively. The former highlights the need for increasing gross cropped area, which can be achieved through greater cropping intensity, while the latter was associated with increased fertilizer, labor, and pesticide input use. Despite the importance of these factors, household poverty trajectory maps showed that changes in off-farm income had played the strongest role in improving livelihoods in this coastal area. This study can help inform development efforts and policies for boosting farm-level crop productivity, specifically through agricultural intensification (higher cropping intensity combined with appropriate and efficient use of inputs) and expanding opportunities for off-farm income as key pathways to bring smallholder households out of poverty.

Does climate change stimulate household vulnerability and income diversity? Evidence from southern coastal region of Bangladesh

Bangladesh is one of the most climate-vulnerable countries globally, where the livelihood of agro-based dependent people became vulnerable due to different natural hazards, especially in the southern coastal part. This study investigates the influence of climate change on household vulnerability and income diversity, data collected from the climate-vulnerable coastal areas of Bangladesh. Both panel data regression and structural equation model were employed to examine the vulnerability status, whereas income diversity was measured through diversity index and "Type-66" livelihood strategy. Results reveal that sources of income have diversified over time. However, the study also reveals that climate change-especially the increase in salinity has affected crop production, resulting in increased income vulnerability of small and marginal farmers who are highly reliant on farm income. Moreover, findings reveal that climate change has influenced households to diversify into low-income sources that do not help to overcome their income vulnerability. Therefore, a cooperative land management system, establishment of embankment, training, and skill development programs are needed to generate feasible alternative income sources to improve the livelihood of coastal people.

Mapping the spatial and temporal variability of flood susceptibility using remotely sensed normalized difference vegetation index and the forecasted changes in the future

Accurate runoff forecasting plays a considerable role in the appropriate water resource planning and management. The spatial and temporal evaluation of the flood susceptibility was explored in the Quebec basin, Canada. This study provides a new strategy for runoff modelling as one of the complicated variables by developing new machine learning techniques along with remote sensing. A novel scheme of the Group Method of Data Handling (GMDH) known as the generalized structure of GMDH (GSGGMDH) is developed to overcome this classical approach's limitation. A simple time series based scenario with exogenous variables including precipitation and Normalized Difference Vegetation Index (NDVI) was introduced for runoff forecasting. MODIS data included MOD13Q1 product was employed and a JavaScript code was developed to preprocess collected data in the Google Earth Engine (GEE) environment. Using different seasonal and non-seasonal lags of all input variables, the developed GSGMDH found the most optimum input combination for each station in terms of simplicity and accuracy, simultaneously (average values; SI = 0.554, RMSRE = 1.55, MAE = 5.076). The precipitation values are modelled with the CanEsm2 climate change model. To apply NDVI for runoff forecasting, a simple spatial-temporal GSGMDH based model was developed (average values; SI = 0.27; RMSRE = 8.27, MAE = 0.08). The forecasting results indicated that the months in which the maximum runoff occurred have changed, and these months have increased compared to the historic period. In the historical period, the frequency of maximum runoff was in April and March. Still, for the two forecasting periods (i.e. 2020-2039 and 2040-2059), the months in which the maximum runoff has occurred have changed, and their amount has been reduced and added to other months, especially February and August.

Storm impacts on a coupled human-natural coastal system: Resilience of developed coasts

Human occupation of and alteration of the world's coast has transformed large stretches of it into Coupled Human-Natural Systems (CHANS) in which humans both influence and are influenced by coastal evolution. In such systems, human activity is as critical on natural resilience as processes and sediment supply derived from the natural setting. Pre- and post-storm observations of these interactions on the intensively developed Atlantic coast of the Gulf of Cádiz, (Spain and Portugal) are examined to determine natural and engineering resilience. Three case studies are used in three CHANS, showing that human interventions interact in complex ways with the natural system influencing post-storm recovery. In natural coasts, storm impact is assessed in terms of geomorphological response; on developed coasts, it is quantified as damage to infrastructure or loss of amenity. Preparedness, availability of resources, choice of response and the speed at which human agencies respond affect resilience for post-storm beach behaviour. Results show in some sites natural resilience adjusting by post-storm sediment transfers and an equilibrium morphology that may differ from pre-storm morphology; engineering resilience ensured that CHANS regained their pre-storm human infrastructure and amenity. Their management requires a fundamentally different approach to that of natural coastlines. The current immature stage of understanding of CHANS (especially the human preparedness and response components) is illustrated by the case studies presented where short-term political decisions and reactions to storms play a strong role in post-storm response. The nature and extent of many developed coasts as CHANS is slowly becoming more widely acknowledged, but to increase natural resilience and decrease vulnerability in CHANS better planning is required so that future storms are anticipated and when they happen, pre-planned human response actions are activated. Storms are an integral and inevitable element in the behaviour of coastal CHANS, not a disaster or emergency.

Coping with crisis on the coast: The effect of community-developed coping-strategies on vulnerability in crisis-prone regions of the Ganges delta

Coastal communities are prone to crises. Repeated exposure to crises constrains the ability of residents to access basic needs such as health, water and food, and may increase their vulnerability levels. In response, communities develop coping strategies such as depoldering (temporary breaching of embankments for TRM: tidal rivers management) and anti-aquaculture movements. However, existing research has not adequately explored the relationship between coping strategies and vulnerability. Theoretical literature is characterized by ambiguity on how various geocentric and anthropocentric factors affect vulnerability in the presence of community-developed coping strategies. Therefore, to advance theoretical knowledge in this field, this article first conceptualizes an integrated framework on the association between vulnerability and coping strategies by merging anthropocentric and geocentric approaches. It then uses mixed methods drawn from social science (surveys, semi-structured interviews), geography (spatial tools) and statistics (multiple regression) on data collected from the coastal belt of Bangladesh to demonstrate that coping strategies may have an effect on vulnerability in crisis-prone coastal regions. The significance of this study is that it demonstrates how the association between vulnerability and coping strategies is likely to be nuanced: depending on a) the type of vulnerability (food/water/health), and b) the coping strategy (TRM vs. anti-aquaculture movements). Different coping strategies are associated with different kinds of vulnerability and these relationships depend on local context (other anthropocentric and geocentric variables). Community movements against aquaculture could reduce food vulnerability, whereas TRM may reduce water vulnerability. Reduction in health vulnerability may instead be associated with urbanization and infrastructure development.

Simulating flash flood hydrographs and behavior metrics across China: Implications for flash flood management

China frequently suffers from considerable and disastrous flash floods with wide areal coverage and high frequency. Obtaining useful information to support flash flood management and decision-making is challenging for massive flash flood events that vary greatly in spatio-temporal characteristics. In this study, hydrological modelling approach (CNFF) and cluster analysis were integrated to assess simulation reliability of entire flash flood processes including both hydrographs and behavior characteristics in a manner of similarity classification, rather than at event scale. A total of 207 hourly events from 13 mountainous catchments with diverse physiographic and meteorological characteristics across China were selected for study. Representative flash flood types were classified using normalized hydrographs with diverse spatio-temporal patterns by k-means clustering algorithm. For individual flash flood types, simulation reliability of CNFF was assessed in capturing corresponding hydrographs, seven behavior metrics measuring flash flood magnitude, intensity, occurrence time, flood timescale, rates of change and variability, and their uncertainties. Results showed that three (fast, intermediate and slow) flash flood types were identified from all the flash flood events with overall average silhouette index of 0.45. Hourly hydrographs of three flash flood types were well reproduced by CNFF, with absolute average relative error of runoff within 15% and Nash-Sutcliffe Efficiency above 0.55. All the behavior metrics were the most accurately reproduced for slow flash flood type with the least average relative root-mean-square error (0.30), followed by intermediate (0.52) and fast (0.58) types. Moreover, the slow flash flood type had the most reliable but greatest uncertainty interval of both hydrograph and behavior metrics, with average relative interval length being 1.24 and 71.96%, and 93.10% and 100% of observations contained in 95% confidence interval, respectively. This study provided efficient and detailed information for flash flood management, and extended application scope of hydrological models to encompass flash flood types and behavior metrics.

Impact of climate change and management strategies on water and salt balance of the polders and islands in the Ganges delta

Enhancing crop production, particularly by growing a crop in the typically-fallow dry season is a key strategy for alleviating poverty in the Ganges delta region. We used a polder water and salt balance model to examine the impact of several crop management, salt management and climate change scenarios on salinity and crop evapotranspiration at Dacope and Amtali in Bangladesh and Gosaba in India. A key (and unsurprising) finding is that salt management is very important, particularly at the two drier sites, Dacope and Gosaba. Good salt management lowers salinity in the shallow groundwater, soil and water storage ponds, and leads to more irrigation. Climate change is projected to alter rainfall, and this in turn leads to modelled increases or decreases in runoff from the polders, and thence affect salt concentrations in the soil and ponds and canals. Thus, the main impacts of climate change are through the indirect impacts on salt concentrations, rather than the direct impacts of the amount of water supplied as rainfall. Management practices to remove salt from polders are therefore likely to be effective in combatting the impacts of projected climate change particularly at Dacope and Gosaba.

The potential of Tidal River Management for flood alleviation in South Western Bangladesh

Reduced sediment deposition, land subsidence, channel siltation, and salinity intrusion has been an unintended consequence of the construction of polders in the south western delta of Bangladesh in the 1960s. Tidal River Management (TRM) is a process that is intended to temporarily reverse these processes and restore sediment deposition and land elevation at the low-lying sites, known as 'beels', where TRM is carried out. However, there is limited evidence to prioritise sites for TRM on the basis of its potential effectiveness at alleviating flooding. In this study, the south western delta of Bangladesh was classified according to different flood susceptible zones. In south western Bangladesh, the major portion of agricultural and aquaculture land is located within flood susceptible zones (65% and 81%, respectively). 44.5% of the total population in embanked regions live in areas classified as being flood susceptible. This study identified 106 'beels' suitable for TRM. Modelling of potential sediment deposition predicted that the consequent increase in land elevation could be up to 1.4 m in five years, which would alleviate land subsidence and modify several geomorphological factors such as aspect, slope, curvature, and Stream Power Index (SPI). Implementation of TRM at these sites could potentially reduce the probability of annual flooding from 0.86 (on average) to 0.57 (on average). Therefore, TRM could lower the flood susceptible area by 35% in suitable 'beels'. Whilst during the implementation of TRM agriculture has to cease for a few years, a systematic programme of TRM could result in a long-term increase in agricultural production by reducing flood susceptibility of agricultural lands in delta regions.

Development of novel hybridized models for urban flood susceptibility mapping

Floods in urban environments often result in loss of life and destruction of property, with many negative socio-economic effects. However, the application of most flood prediction models still remains challenging due to data scarcity. This creates a need to develop novel hybridized models based on historical urban flood events, using, e.g., metaheuristic optimization algorithms and wavelet analysis. The hybridized models examined in this study (Wavelet-SVR-Bat and Wavelet-SVR-GWO), designed as intelligent systems, consist of a support vector regression (SVR), integrated with a combination of wavelet transform and metaheuristic optimization algorithms, including the grey wolf optimizer (GWO), and the bat optimizer (Bat). The efficiency of the novel hybridized and standalone SVR models for spatial modeling of urban flood inundation was evaluated using different cutoff-dependent and cutoff-independent evaluation criteria, including area under the receiver operating characteristic curve (AUC), Accuracy (A), Matthews Correlation Coefficient (MCC), Misclassification Rate (MR), and F-score. The results demonstrated that both hybridized models had very high performance (Wavelet-SVR-GWO: AUC = 0.981, A = 0.92, MCC = 0.86, MR = 0.07; Wavelet-SVR-Bat: AUC = 0.972, A = 0.88, MCC = 0.76, MR = 0.11) compared with the standalone SVR (AUC = 0.917, A = 0.85, MCC = 0.7, MR = 0.15). Therefore, these hybridized models are a promising, cost-effective method for spatial modeling of urban flood susceptibility and for providing in-depth insights to guide flood preparedness and emergency response services.

Capturing transformation of flood hazard over a large River Basin under changing climate using a top-down approach

Existing flood modeling studies over coastal catchments involving different combinations of model chain setup imparting complex information fails to entail the needs of policy or decision-makers. Thus, a comprehensive framework that pertains to the requirements of practitioners and provides more perspicuous flood hazard information is required. In this paper, a novel approach translating complex flood hazard information in the form of decision priority maps derived using a rational combination of models (physical and statistical) is elucidated at the finest administrative scale. The proposed methodology is illustrated over a highly flood-prone deltaic region in Mahanadi River Basin, India, to characterize impacts of climate change for a 1:100 years return period flood event under future conditions (2026-2055). The modeled flood events are further analyzed to capture the transformation dynamics of flood hazard classes (FHCs) in near-future, for prioritizing areas with greater hazard potential. Interestingly, the results capture a high transformation characteristic from low to high FHCs in agriculture-dominated areas, which are significantly greater than the areas experiencing flood hazard reduction. The results show a significant increase of 12.5% and 27.35% in areas with high FHCs under RCP4.5 and RCP8.5 scenarios, respectively. Moreover, a notable climate change response is indicated under both climate change scenarios, with approximately 22% (RCP4.5) and 25% (RCP8.5) in villages showing a drastic increment in flood hazard magnitude. The results thus highlight the importance of identifying and prioritizing the areas for flood adaptation where a relative change in flood hazard potential is higher due to climate change. Therefore, we conclude that this study can provide an insight into the implication of new approaches for effective communication of flood information by bridging the gaps between scientific communities and decision-makers in appraisal for better flood adaptation measures.

Machine Learning to Evaluate Impacts of Flood Protection in Bangladesh, 1983–2014

Impacts of climate change adaptation strategies need to be evaluated using principled methods spanning sectors and longer time frames. We propose machine-learning approaches to study the long-term impacts of flood protection in Bangladesh. Available data include socio-economic survey and events data (death, migration, etc.) from 1983–2014. These multidecadal data, rare in their extent and quality, provide a basis for using machine-learning approaches even though the data were not collected or designed to assess the impact of the flood control investments. We test whether the embankment has affected the welfare of people over time, benefiting those living inside more than those living outside. Machine-learning approaches enable learning patterns in data to help discriminate between two groups: here households living inside vs. outside. They also help identify the most informative indicators of discrimination and provide robust metrics to evaluate the quality of the model. Overall, we find no significant difference between inside/outside populations based on welfare, migration, or mortality indicators. However, we note a significant difference in inward/outward movement with respect to the embankment. While certain data gaps and spatial heterogeneity in sampled populations suggest caution in any conclusive interpretation of the flood protection infrastructure, we do not see higher benefits accruing to those living with higher levels of protection. This has implications for Bangladesh’s planning for future and more extreme climate futures, including the national Delta Plan, and global investments in climate resilient infrastructure to create positive social impacts.