Votre recherche

Floods stand as one of the most harmful natural disasters, which have become more dangerous because of climate change effects on urban structures and agricultural fields. This research presents a comprehensive flood mapping approach that combines multi-sensor satellite data with a machine learning method to evaluate the July 2021 flood in the Netherlands. The research developed 25 different feature scenarios through the combination of Sentinel-1, Landsat-8, and Radarsat-2 imagery data by using backscattering coefficients together with optical Normalized Difference Water Index (NDWI) and Hue, Saturation, and Value (HSV) images and Synthetic Aperture Radar (SAR)-derived Grey Level Co-occurrence Matrix (GLCM) texture features. The Random Forest (RF) classifier was optimized before its application based on two different flood-prone regions, which included Zutphen’s urban area and Heijen’s agricultural land. Results demonstrated that the multi-sensor fusion scenarios (S18, S20, and S25) achieved the highest classification performance, with overall accuracy reaching 96.4% (Kappa = 0.906–0.949) in Zutphen and 87.5% (Kappa = 0.754–0.833) in Heijen. For the flood class F1 scores of all scenarios, they varied from 0.742 to 0.969 in Zutphen and from 0.626 to 0.969 in Heijen. Eventually, the addition of SAR texture metrics enhanced flood boundary identification throughout both urban and agricultural settings. Radarsat-2 provided limited benefits to the overall results, since Sentinel-1 and Landsat-8 data proved more effective despite being freely available. This study demonstrates that using SAR and optical features together with texture information creates a powerful and expandable flood mapping system, and RF classification performs well in diverse landscape settings. © 2025 by the authors.

Accurate prediction of wave overtopping rates is essential for flood risk assessment along coral reef coastlines. This study quantifies the uncertainty sources affecting overtopping rates for vertical seawalls on reef flats, using ensemble simulations with a validated non-hydrostatic SWASH model. By generating extensive random wave sequences, we identify spectral resolution, wave spectral width, and wave groupiness as the dominant controls on the uncertainty. Statistical metrics, including the Coefficient of Variation ((Formula presented.)) and Range Uncertainty Level ((Formula presented.)), demonstrate that overtopping rates exhibit substantial variability under randomized wave conditions, with (Formula presented.) exceeding 40% for low spectral resolutions (50–100 bins), while achieving statistical convergence ((Formula presented.) around 20%) requires at least 700 frequency bins, far surpassing conventional standards. The (Formula presented.), which describes the ratio of extreme to minimal overtopping rates, also decreases markedly as the number of frequency bins increases from 50 to 700. It is found that the overtopping rate follows a normal distribution with 700 frequency bins in wave generation. Simulations further demonstrate that overtopping rates increase by a factor of 2–4 as the JONSWAP spectrum peak enhancement factor ((Formula presented.)) increases from 1 to 7. The wave groupiness factor ((Formula presented.)) emerges as a predictor of overtopping variability, enabling a more efficient experimental design through reduction in groupiness-guided replication. These findings establish practical thresholds for experimental design and highlight the critical role of spectral parameters in hazard assessment. © 2025 by the authors.

Flood intensity has significantly increased globally, which poses significant challenges for the environment and urban areas. This research aimed to evaluate the performance of different flood hazard assessment approaches to identify flood risk areas in Erbil city, Kurdistan region of Iraq. The analytical hierarchy process and the new ArcGIS pro flood simulation tool were compared to identify flood-risk areas and assess their performance based on historical flood events. Multiple factors such as rainfall, aspect, topographic wetness index, elevation, flow accumulation, lithology, soil data, normalized difference vegetation index, normalized difference built-up index, land use land cover, slope, stream power index, drainage density, evapotranspiration, infiltration rates, and distance to roads were considered to identify flood risk areas. Using the Analytical Hierarchy Process, areas spanning over 35 km² (3.9%) and 74 km² (27%) of Erbil city were identified as very high and high flood susceptible, respectively. However, the results of AFS indicated that an area of 66.3 km² (7.3%) of Erbil city will be inundated during rainfall intensity of 60 mm/day. The receiver operating characteristic area under the curve assessments showed the accuracy of AFS to be 95.3% and that of the Analytical Hierarchy Process to be 92.2%. The comparison analysis emphasized the effectiveness of ArcGIS Pro flood simulation in terms of accurate flood inundation assessments. This research provides significant insights into suitable approaches to flood hazard assessment by considering different scales and data availability, helping policymakers and urban planners understand floods better and implement appropriate mitigation strategies accordingly. © 2025, Union of Iraqi Geologists. All rights reserved.

African hydrological systems are incredibly complex and highly sensitive to climate variability. This review synthesizes observational data, remote sensing, and climate modeling to understand the interactions between fluvial processes, water cycle dynamics, and anthropogenic pressures. Currently, these systems are experiencing accelerating warming (+0.3 °C/decade), leading to more intense hydrological extremes and regionally varied responses. For example, East Africa has shown reversed temperature–moisture correlations since the Holocene onset, while West African rivers demonstrate nonlinear runoff sensitivity (a threefold reduction per unit decline in rainfall). Land-use and land-cover changes (LULCC) are as impactful as climate change, with analysis from 1959–2014 revealing extensive conversion of primary non-forest land and a more than sixfold increase in the intensity of pastureland expansion by the early 21st century. Future projections, exemplified by studies in basins like Ethiopia’s Gilgel Gibe and Ghana’s Vea, indicate escalating aridity with significant reductions in surface runoff and groundwater recharge, increasing aquifer stress. These findings underscore the need for integrated adaptation strategies that leverage remote sensing, nature-based solutions, and transboundary governance to build resilient water futures across Africa’s diverse basins.

Cotton (Gossypium spp.) is an important industrial crop, but it is vulnerable to waterlogging stress. The relationship between cotton yields and waterlogging indicators (CY-WI) is fundamental for waterlogging disaster reduction. This review systematically summarized and analyzed literature containing CY-WI relations across 1970s–2020s. China conducted the most CY-WI experiments (67%), followed by Australia (17%). Recent decades (2010s, 2000s) contributed the highest proportion of CY-WI works (49%, 15%). Surface waterlogging form is mostly employed (74%) much more than sub-surface waterlogging. The flowering and boll-forming stage, followed by the budding stage, performed the most CY-WI experiments (55%), and they showed stronger negative relations of CY-WI than other stages. Some compound stresses enhance negative relations of CY-WI, such as accompanying high temperatures, low temperatures, and shade conditions, whereas some others weaken the negative CY-WI relations, such as prior/post drought and waterlogging. Anti-waterlogging applications significantly weaken negative CY-WI relations. Regional-scale CY-WI research is increasing now, and they verified the influence of compound stresses. In future CI-WI works, we should emphasize the influence of compound stresses, establish regional CY-WI relations regarding cotton growth features, examine more updated cotton cultivars, focus on initial and late cotton stages, and explore the consequence of high-deep submergence.

Nature-based solutions (NbS) are increasingly recognized as strategic alternatives and complements to grey infrastructure for addressing water-related challenges in the context of climate change, urbanization, and biodiversity decline. This article presents a critical, theory-informed review of the state of NbS implementation in European water management, drawing on a structured synthesis of empirical evidence from regional case studies and policy frameworks. The analysis found that while NbS are effective in reducing surface runoff, mitigating floods, and improving water quality under low- to moderate-intensity events, their performance remains uncertain under extreme climate scenarios. Key gaps identified include the lack of long-term monitoring data, limited assessment of NbS under future climate conditions, and weak integration into mainstream planning and financing systems. Existing evaluation frameworks are critiqued for treating NbS as static interventions, overlooking their ecological dynamics and temporal variability. In response, a dynamic, climate-resilient assessment model is proposed—grounded in systems thinking, backcasting, and participatory scenario planning—to evaluate NbS adaptively. Emerging innovations, such as hybrid green–grey infrastructure, adaptive governance models, and novel financing mechanisms, are highlighted as key enablers for scaling NbS. The article contributes to the scientific literature by bridging theoretical and empirical insights, offering region-specific findings and recommendations based on a comparative analysis across diverse European contexts. These findings provide conceptual and methodological tools to better design, evaluate, and scale NbS for transformative, equitable, and climate-resilient water governance.

Floods are natural hazards that have the greatest socioeconomic impact worldwide, given that 23% of the global population live in urban areas at risk of flooding. In this field of research, the analysis of flood risk has traditionally been studied based mainly on approaches specific to civil engineering such as hydraulics and hydrology. However, these patterns of approaching the problem in research seem to be changing in recent years. During the last few years, a growing trend has been observed towards the use of methodology-based approaches oriented towards urban planning and land use management. In this context, this study analyzes the evolution of these research patterns in the field by developing a bibliometric meta-analysis of 2694 scientific publications on this topic published in recent decades. Evaluating keyword co-occurrence using VOSviewer software version 1.6.20, we analyzed how phenomena such as climate change have modified the way of addressing the study of this problem, giving growing weight to the use of integrated approaches improving territorial planning or implementing adaptive strategies, as opposed to the more traditional vision of previous decades, which only focused on the construction of hydraulic infrastructures for flood control.

ABSTRACT Waterlogging is a critical abiotic stress factor severely affecting maize, one of the World's most widely cultivated cereal crops. Globally, maize is a crucial food crop, grown in diverse agro‐climatic zones, from subtropical to temperate climates. Waterlogging, resulting from flooding, intense rainfall and inefficient drainage systems, continues to be a major abiotic stress factor influencing crop productivity globally. Prolonged exposure to excess soil moisture leads to oxygen depletion in the root zone, resulting in restricted aerobic respiration, impaired nutrient uptake and disruption of physiological processes. This review aims to provide a comprehensive overview of the morphological, physiological and biochemical changes maize undergoes in response to waterlogging stress. Key aspects such as root system adaptation, reduction in photosynthetic efficiency, accumulation of reactive oxygen species (ROS) and hormonal imbalances are systematically examined. Furthermore, we delve into the metabolic shifts that enable maize to survive under anaerobic conditions, including alterations in energy metabolism, carbohydrate partitioning, and activating antioxidant defence mechanisms. The role of key signalling molecules such as ethylene is explored, highlighting their involvement in regulating stress responses. Additionally, the review discusses agronomic and genetic approaches for improving waterlogging tolerance in maize, including the development of stress‐resilient cultivars through breeding and biotechnological interventions. By synthesising recent advances in understanding maize's response to waterlogging, this paper identifies gaps and proposes future research directions, focusing on the integration of molecular and field‐based strategies. The insights from this review are crucial for developing sustainable agricultural practices aimed at mitigating the adverse impacts of waterlogging on maize productivity, particularly in flood‐prone areas. Breeding for waterlogging resilience integrates the creation of robust varieties, plant morphology optimisation, and utilisation of tolerant secondary traits through combined conventional and biotechnological breeding strategies.

ABSTRACT Natural flood management strategies (NFMs) encompass a variety of measures implemented across catchments to mitigate flood risks while providing multiple benefits. In recent years, NFMs have gained increasing attention from researchers and policymakers. However, despite the growing body of research, there remains a lack of a critical review that quantitatively synthesises the reported performance of different NFMs by analysing their effects on key hydrological parameters. To address this gap, we conducted a systematic review of NFMs based on 145 peer‐reviewed papers covering 216 case studies across 37 countries, following the preferred reporting items for systematic reviews and meta‐analyses (PRISMA) guidelines. Our analysis moves from a descriptive overview of the evidence base to a novel, quantitative investigation of three critical themes: the characteristics of studied NFM schemes, the methodologies used for their assessment, and their quantitative hydrological performance and its influencing factors. Results indicate that 31% of the studies identified flood peak reduction as the most commonly targeted hydrological objective. A significant positive correlation was found between intervention diversity and intensity (Spearman's ρ  = 0.53). Furthermore, our methodological analysis reveals a critical trade‐off in the literature, with empirical monitoring typically used in small catchments over shorter durations, while modelling is used to assess a greater diversity of interventions at larger scales, with truly combined approaches being notably rare (11%). Notably, river and floodplain management (RFM) demonstrated higher effectiveness, achieving an average flood peak reduction of 30%, particularly in larger catchments. Bearing the often multi‐faceted aims of NFMs in mind, this paper provides key suggestions for future research.

The flood disasters are prevalent in the Ganga–Brahmaputra (GB) basin with recurrent occurrences and severe impacts across the major watersheds. The present study analyses the vulnerability of 44 watersheds to flood inundation and its impact on cropland, urban areas, and population. The Sentinel-1 dataset was utilised to analyse flood extent and frequency from 2015 to 2022, enabling the identification of flood-prone watersheds in the Ganga–Brahmaputra basin. The analysis revealed that 7 watersheds in the Ganga basin and 12 watersheds in the Brahmaputra basin are particularly vulnerable to flooding. The flood hazard analysis was performed using fuzzy-AHP (Analytic Hierarchy Process), focusing on six parameters, including topographic wetness index (TWI), elevation, precipitation, drainage density, distance from river, and NDVI for the selected 19 watersheds. The inundation analysis from 2015 to 2022 revealed that the maximum flood extent was observed in 2020, with an affected area of 33,537.6 km2 and 34,937.9 km2 in the Ganga–Brahmaputra basin, respectively. The flood hazard analysis identified Upper Ganga (8877.52 km2), Ghaghara (18573.9 km2) and Teesta (1543.06 km2) as having the highest proportion of their geographical area under very high-hazard zone and the highest percentage in the very low hazard zones were observed in Jamuneshwary (1093.55 km2), Atreyee (4410.42 km2), and Kulsi (1273.89 km2). By first mapping these watersheds with precision and then using various parameters for flood hazard analysis, it ensures accurate identification of flood-prone areas, offering valuable insights for flood management and mitigation in a critical region. © Indian Academy of Sciences 2025.

Urbanization and climate change keep intensifying extreme rainfall events. Previous studies have explored urban flood susceptibility, yet a comprehensive approach that unifies these perspectives has remained underdeveloped. This study established a holistic framework using the SD-PLUS-LightGBM model with multiple variables under three SSP-RCP scenarios to predict spatial-temporal dynamics of flood susceptibility in the Greater Bay Area between 2030 and 2050. Compared with traditional models, LightGBM established superior predictive accuracy and operational reliability for urban flood susceptibility mapping. The results indicated a non-linear expansion of high-susceptibility zones, with SSP5–8.5 projections showing a two-fold increase in vulnerable areas by 2050 relative to 2020 baselines. Regions experiencing pronounced susceptibility transitions were expected to grow significantly (0.23 % of the total area), concentrated in historic urban cores and peri‑urban interfaces. This study offered an in-depth approach to stormwater management along with targeted recommendations for sustainable urban planning and design. © 2025

This study proposes a hybrid urban flood damage prediction framework that integrates a Deep Feed-Forward Neural Network (DFNN) with a Rainfall-Runoff (R-R) model and the Korean Flood Risk Assessment Model (K-FRM). The model predicts 10 types of flood risk indicators (FRIs), including damage to residential and non-residential buildings, using only simplified rainfall variables (SRVs), eliminating the need for complex hydrodynamic simulations. Synthetic rainfall scenarios were generated for training and fed into the R-R model, whose outputs were processed through K-FRM to produce training data for the DFNN model. The optimized DFNN model was validated by comparing its predictions with flood damage estimates from K-FRM, demonstrating a Nash-Sutcliffe Efficiency (NSE) of up to 0.87 and an R2 of up to 0.88, indicating strong predictive performance across flood risk indicators. These results highlight the effectiveness of the DFNN-based hybrid approach in capturing flood damage patterns and providing rapid predictions using forecasted rainfall data. The proposed method offers a practical and computationally efficient tool for urban flood risk management and disaster mitigation planning. © 2025 The Authors

Spring floods occurring in the Midwest US are often aggravated by meteorological and hydrological conditions. In this study, the seasonal influences of precipitation (PRCP) and river stage (RS) on groundwater level (GWL) fluctuations were analyzed for the Middle and Lower Platte watersheds along the Platte River in a Midwestern catchment that is vulnerable to spring floods. Statistical analysis was conducted to simulate GWL with a moving average-based time lag consideration approach by using multiple hydrological data sets for 25 study sites. The results showed that the time lag consideration approach appropriately provided the regional information of water infiltration characteristics among GWL, PRCP, and RS for each study site. Also, the correlations of GWL with PRCP and RS were found to vary depending on the season. Especially in the early spring season, the correlation of GWL with PRCP is very weak (correlation coefficient=0.001 to 0.198). This may be due to entirely or partially frozen ground, which prevents rainwater from penetrating into the aquifer, causing large amounts of runoff and contributing to frequent flooding in early spring. In addition, statistical analysis showed that accounting for the time lag of PRCP and RS improved GWL simulation performance, and their influence varied by season. © 2025 American Society of Civil Engineers.

During and after a disaster, selected services and systems are needed to recover and maintain important functions of society. These are deemed critical infrastructure (CI). When these services are disrupted due to the impacts of a disaster, response and recovery may be slowed or halted. As flooding events are occurring more often across larger geographic extents, advancing methods for assessing risks of flooding to CI is vital. We use Utah, USA as a case study to demonstrate a novel, transferable approach for assessing fine-scale flood risks to CI across large geographic areas. Specifically, our assessment approach integrates high-resolution building footprints of schools, first responder facilities, and hospitals, and flood risk maps from a state-of-the-art big data flood model and the U.S. Federal Emergency Management Agency (FEMA). We show that 94 CI facilities across Utah are at risk of severe flooding, and that those risks to CI are almost entirely overlooked by FEMA flood risk maps. Though nearly every CI building is located outside of FEMA flood zones, FEMA maps inaccurately and incompletely represent flood risks, indicating that future flood risk assessment approaches should use flood risk maps from other sources. The approach we introduce can be used to assess flood risks to CI elsewhere, and case study results can be applied to inform flood risk reduction efforts in Utah. © 2025 Elsevier Ltd

Flooding, caused by the excessive accumulation of water on land, disrupts activities in floodplain regions, particularly during the rainy season. The main objective is to map Flood vulnerability areas and identify regions most vulnerable to flooding to inform effective flood management strategies using an integrated approach that combines remote sensing, geographic information systems (GIS), and the analytical hierarchy process (AHP) to assess Flood vulnerability in the Wuseta Watershed. The research was conducted in three phases: pre-fieldwork, fieldwork, and post-fieldwork. Key factors influencing Flood vulnerability such as drainage density, elevation, land use/land cover, and slope were hierarchically weighted to produce a Flood vulnerability map. Rainfall distribution was not considered as a contributing factor the Ethiopian Meteorological Agency has installed only one weather station in the study area, located in Wuseta watershed. As a result, the rainfall distribution is considered uniform throughout the watershed, making it unsuitable for flood susceptibility assessment. The Flood vulnerability map categorizes the watershed into five zones: very high (0.07 km2), high (4.65 km2), moderate (7.86 km2), slight (4.41 km2), and very slight (0.001 km2). The results show that the upstream, northern, northwestern, and northeastern areas of the watershed face slight to very slight Flood vulnerability, while the southern region is highly vulnerable to flooding. These findings provide valuable insights for policymakers and local communities, aiding in the development of targeted mitigation strategies and raising awareness of flood-prone areas. This study underscores the value of integrating geospatial technologies and multi-criteria decision analysis in flood risk assessment, particularly in data-scarce regions, to enhance disaster preparedness and climate resilience. © The Author(s) 2025.

Flooding is a persistent hazard in tropical regions of India, primarily driven by intense precipitation and further aggravated by anthropogenic activities. Despite ongoing efforts, a gap persists in the development of comprehensive risk models that integrate hazard, vulnerability, and exposure components at a watershed level. This research seeks to bridge that gap by implementing a multi-criteria decision analysis (MCDA) technique, specifically the Analytical Hierarchy Process (AHP), to generate a risk map for the tropical Meenachil River Basin (MRB), originating in the Western Ghats, southwest India. Nine conditioning factors (CFs) were evaluated to assess hazard, and the resulting hazard layer was integrated with vulnerability data and different exposure factors (EFs), such as built-up height, built-up surface, built-up volume, population, and total exposure, to produce a risk map. Validation of the hazard model utilizing the Receiver Operating Characteristic (ROC) curve achieved an excellent Area Under Curve (AUC) of 0.825, along with high accuracy (0.818), F1-score (0.802), precision (0.812), and recall (0.793). Approximately 11% of the MRB lies in a very high hazard zone and 1.51% in a very high risk zone. These results advocate for sustainable flood management by identifying key risk zones, thereby facilitating the implementation of focused site-specific mitigation strategies. © The Author(s) 2025.