Journal of Flood Risk Management最新文献

Flooding has become an increasing threat across the globe and risk is likely to rise in the coming decades. In this context, individuals and communities are expected to assume certain tasks and responsibilities, including following the rules and guidance of flood risk management (FRM) agencies. However, such collaboration does not always align seamlessly and may instead involve what we term deviant behaviour. Individuals or groups may act in ways that deviate from the expectations or aspirations of the state or the majority of society, with negative consequences. This paper suggests a typology of such deviant behaviours, with examples occurring before, during or after a flood event. The typology encompasses different forms of behaviours, ranging from the inappropriate and uncooperative to anti-social behaviour and illegal actions. We suggest that FRM agencies—often surprised by and unprepared for such behaviours—need to integrate a recognition of deviant behaviours into their operational logic rather than continue to ignore them. Furthermore, deviant behaviour is not fully addressed directly in the current academic FRM debates. Consequently, there remain significant gaps in understanding, such as the many reasons behind these phenomena, and further research should focus on the significance of deviant behaviours in FRM in relation to their drivers and the consequences.

Floods rank among the most frequent and widely managed hazards worldwide. While considerable research has examined public perceptions of flood risk, less attention has been paid to how people interpret institutional presence, action, and responsibility in flood-prone settings shaped by infrastructure and governance systems. This paper presents a systematic review of 35 empirical studies published between 2010 and 2024 that examine the relationship between institutional flood governance and public interpretation across structurally managed flood contexts. The review synthesizes findings across diverse geographies, governance scales, and infrastructure types, identifying common interpretive patterns in how people make sense of institutional roles in flood-risk management. Thematic analysis reveals six key dynamics: (1) infrastructure as a producer of risk, (2) governance fragmentation and public confusion, (3) preparedness as an interpretive practice, (4) trust and the levee effect, (5) fairness and timing in institutional response, and (6) recognition and communicative legibility. Across these themes, preparedness behavior emerges not solely as a response to risk knowledge or hazard experience, but as a reflection of how institutional action is seen, trusted, or contested. The review contributes to a growing body of research that reframes flood preparedness as a meaning-making process. It highlights the need for flood governance to move beyond technical protection toward more legible, responsive, and participatory engagement. Institutions are not only evaluated on their capacity to protect but also on their visibility, communication effectiveness, and perceived intent, factors that shape whether people prepare, trust, or withdraw in the face of managed risk.

Reducing flood damage is a key challenge for both administrators and communities, particularly in the face of increasing climate-related impacts. A fundamental prerequisite for effective damage reduction is a detailed understanding of the spatial distribution of territorial susceptibility to flood damage. In this context, we present an innovative, data-driven methodology for mapping susceptibility to flood damage to private assets, leveraging machine learning techniques to overcome some of the limitations and uncertainties associated with traditional approaches based on the superposition of hazard, exposure, and vulnerability layers. The proposed methodology is applied to a case study in the Tuscany region of Italy, using approximately 11,000 claims related to flood events that occurred between 2013 and 2023. The claims are analyzed in relation to 15 predisposing factors representing both territorial and socio-environmental characteristics of the study area. The introduction of negative samples, corresponding to locations where residential damage is assumed to be negligible or absent, allows the model to distinguish susceptible from non-susceptible areas. The susceptible areas are subsequently classified into damage magnitude classes based on citizens' self-reported assessments. The intrinsic variability and non-linearity of the process, together with the uncertainty associated with citizen-reported data and potential perception and reporting biases, are addressed through a second-stage ensemble of independently trained models. The final outputs include a majority-vote susceptibility map and a map of prediction stability. The model performs well in identifying the occurrence of reported damage, while the discrimination between damage magnitude classes remains more challenging, highlighting the intrinsic uncertainty associated with citizen-reported damage data. The resulting maps are intended as practical tools to support flood damage reduction efforts and should be viewed as complementary to existing flood risk maps used in land-use planning. The proposed framework provides a transferable approach for using observed damage records to characterize spatial patterns of potential damage, while its application to new geographical contexts requires site-specific calibration and validation.

Large earthquakes can disrupt river systems and produce abrupt, unforeseen changes in flood risk. Forward modelling of earthquake-related surface deformation and its influence on inundation extent, depth, and frequency enables pre-emptive assessment of exposure to this compound hazard. We present a novel paired fault-dislocation and hydrodynamic modelling framework that adapts the concept of Increased Flooding Vulnerability (IFV) to assess increases in flood depth associated with earthquake-induced ground deformation. We apply this framework to simulate rupture scenarios for the real-world case study of the Titri fault in New Zealand to evaluate changes in flood risk within the Taieri Basin, currently managed by the Lower Taieri Flood Protection Scheme (LTFPS). Results reveal zones of heightened exposure to coseismic and post-seismic flooding and demonstrate how deformation could undermine flood defenses. At 2.6 m of slip and a 45° dip on the modelled Titri fault, Dunedin Airport is flooded to depths of 0.3–1.5 m during a 2%–5% annual exceedance probability flood event. Fault uplift also obstructs the basin's drainage outlet, while footwall subsidence reduces infiltration and may facilitate coseismic lake formation. Early identification of areas facing increased flooding vulnerability provides a basis for long-term risk assessment and land-use planning in regions exposed to both seismic and flood hazards.

Extreme rainfall events frequently trigger flash floods, which cause widespread damage to ecosystems, infrastructure, and livelihoods. Accurate rainfall estimation is therefore essential for flood modeling and early warning systems. This study assessed the performance of the Tropical Rainfall Measuring Mission (TRMM) and the Global Precipitation Measurement Integrated Multi-Satellite Retrievals for GPM (GPM-IMERG) against weather station rainfall observations collected across KwaZulu-Natal, South Africa, between 2000 and 2022. Satellite-derived precipitation estimates were evaluated against field measured data using the correlation coefficient (CC), percentage bias (PBIAS), mean absolute error (MAE), and root mean square error (RMSE). The results showed that GPM-IMERG outperformed TRMM, exhibiting stronger agreement with weather station observations (CC = 0.97 compared with 0.86), a lower PBIAS (−3.76% vs. −5.94%), and smaller estimation errors (MAE = 5.48 mm and RMSE = 7.59 mm) than TRMM (MAE = 13.07 mm and RMSE = 15.33 mm). Furthermore, GPM-IMERG exhibited a statistically significant agreement with observed rainfall (p < 0.05), demonstrating improved precipitation estimation over its predecessor, TRMM. These findings demonstrate that GPM-IMERG (GPM_3IMERGM v07) provides reliable precipitation estimates for KwaZulu-Natal and has the potential to improve flood forecasting, strengthen early warning systems, and enhance disaster preparedness in regions vulnerable to extreme rainfall events.

Floods remain among the most frequent and damaging natural hazards globally, with effective flood early warning messages (FEWMs) playing a critical role in reducing disaster risk and enhancing community preparedness. Despite advances in early warning systems, many rural communities in developing countries continue to experience challenges in accessing, understanding, and responding to warning information. This study assessed the effectiveness of FEWMs in the Mokgareng community of the North West Province, South Africa. A quantitative cross-sectional survey design was adopted. Data were collected from 102 adult respondents using a structured questionnaire and analysed using descriptive statistics, a Generalised Linear Mixed Model (GLMM), and binary logistic regression in R Statistical Software. The findings revealed a substantial gap between awareness and exposure to FEWMs. Although most respondents reported understanding the meaning of FEWMs and expressed confidence in interpreting warning information, many indicated that they had not received or could not recall receiving warning messages. High levels of cellphone ownership suggest strong potential for mobile-based dissemination; however, significant concerns were identified regarding message reach, community consultation, and dissemination pathways. The logistic regression analysis showed that the time taken to receive warning messages was the only statistically significant predictor associated with FEWM receipt (β = 0.756, p = 0.012), highlighting the importance of timely dissemination. In contrast, perceptions regarding message effectiveness, adequacy of information, and satisfaction with municipal dissemination approaches were not statistically significant predictors. The study concludes that the effectiveness of FEWMs in the Mokgareng community is constrained more by dissemination failures than by comprehension challenges. Strengthening timely message delivery, improving community participation in warning system design, and enhancing trusted communication channels could significantly improve flood preparedness and response in vulnerable rural communities.

Urbanisation, climate uncertainty, and limited city budgets have heightened the need for efficient urban drainage management. While advanced hydrodynamic models assess the performance of drainage networks, they are computationally intensive, time-consuming, and expensive. This study introduces a spatial analysis framework that integrates the Topographic Wetness Index (TWI) and the Topographic Control Index (TCI), derived from high-resolution Digital Surface Models (DSMs) captured using Unmanned Aerial Systems (UAS), to evaluate storm drain positioning relative to surface-water flowpaths and microtopographic controls; however, it does not account for subsurface pipe capacity or sewer network performance. TWI was used to assess whether storm drains align with surface flow paths, while TCI identified drains located within topographic depressions. The optimal TWI threshold (90th percentile) was determined by comparing it with the 1-in-100-year Environment Agency flood map. Using this threshold, TWI-based classification achieved 53% accuracy and 75% precision in identifying drains along flow paths, while TCI-based analysis demonstrated 92% accuracy and 79% precision for detecting drains within depressions. The combined framework enables the identification of drainage inefficiencies at street level, informing maintenance prioritisation and the integration of nature-based solutions. By using UAS-derived geospatial data and topographic modelling, this approach provides a Tier 1 screening tool that is cost-effective, scalable, and proactive for adaptive stormwater planning.

This study presents a machine-learning framework for identifying areas of elevated flood risk using imbalanced, high-dimensional geospatial datasets. Using the Mandra River Basin in Greece as a case study, an extreme gradient boosting (XGBoost) classifier was trained on 11 flood-related features including hydrological, meteorological, soil and topographic data, whereas 454 geolocated citizens' emergency calls were used as target information. Class imbalance was addressed through a weighted-only strategy to preserve the integrity of the original dataset, while adjusting class contributions during the training process. Bayesian optimization and stratified 10-fold cross-validation were combined to tune and evaluate the model performance. Threshold-sweep analysis enables the identification of multiple operational settings based on precision–recall trade-offs and cost-sensitive criteria. The results revealed physically interpretable feature importance patterns, with terrain elevation (TE) (45%) and accumulated rainfall (25%) emerging as the dominant predictors according to feature importance analysis, followed by curve number (16%) and the Manning coefficient (9.2%). The proposed framework provides a reliable, data-driven tool for flood risk assessment, and supports a flexible, two-tier decision strategy that differentiates between emergency escalation and broader situational awareness.

Flooding is one of the most frequent and devastating natural hazards affecting urban areas globally, leading to considerable social, economic, and environmental impacts. In the present study, a novel framework for flood susceptibility mapping is proposed using the stochastic Hill-Climbing (SHC) method in the Analytical Hierarchy Process (AHP) integrated with Geographic Information System (GIS) technique. For illustrating the proposed methodology, a set of ten geo-environmental and human-induced factors are considered, which are elevation, land cover, slope, lithology, drainage density, rainfall intensity, proximity to rivers, topographic wetness index (TWI), soil classification, and groundwater level for one of the fast-growing districts, Raipur in India. These factors are weighted according to their relative importance in influencing flood occurrence, and the Flood Hazard Index (FHI) is evaluated across the study area using the rainfall data of various time periods. The results indicate that proximity to rivers, elevation, and slope are the major contributing factors to the flood susceptibility level in the study area. Depending upon the calculated FHI values, the study grouped the area into five levels of flood susceptibility: very high, high, moderate, low, and very low. The model's performance was checked with the Receiver Operating Characteristic (ROC) curve, where the predicted data was plotted between the data obtained from the District Disaster Management Authority, Raipur, and it reached a score of 0.828, showing good prediction ability. The final flood susceptibility map gives useful information for district authorities, planners, and disaster management teams to support better decisions by identifying high-risk zones which strengthen Raipur district's resilience to floods.

There is increasing interest among decision makers in retrofitting large numbers of sustainable drainage systems (SuDS) across urban areas to help deal with the growing issue of pluvial flooding. For the practitioner an important early stage in the development of a retrofit SuDS scenario is the optioneering process, when it is desirable to evaluate the flood risk benefit of the SuDS scenario for the first time. However, the modelling options currently available are unsuited to the task. Some require specific design information that is not available at the early stages of SuDS projects, while others ignore the complexity of the urban topography or the sub-surface drainage system. In this paper we used a dynamically coupled integrated 1D–2D hydrodynamic model to evaluate the flood risk benefit of a distributed retrofit SuDS scenario developed for an urban neighbourhood in the UK. The 675 SuDS that comprise the SuDS scenario were conceptualised through a multidisciplinary optioneering process and represented in the model by modifying the 2D surface and parameters. The model successfully represented pluvial flooding in the study area and provided a comprehensive means of comparing a distributed retrofit SuDS scenario to a Baseline scenario. The 2D surface was essential for representing the heterogeneous distribution of the simulated flooding, while the 1D sub-surface drainage model demonstrated the non-linearity of the drainage rate, which included a period of surcharge. The SuDS scenario delivered meaningful reductions in the depth of flooding, but the impact of the SuDS diminished as storm severity increased. Comparison of the model with the UK Environment Agency Risk of Flooding from Surface Water map suggests that it reproduced the spatial patterns of flood risk at the study site and highlighted the importance of the 1D representation of the drainage system. This study demonstrates the potential of a dynamically coupled integrated 1D–2D hydrodynamic model to produce spatially explicit evaluations of plausible retrofit SuDS scenarios developed by practitioners.