With the growing integration of artificial intelligence (AI) into warning systems, an important question concerns how the public interprets and responds to warnings delivered by governmental authorities and AI systems. Existing studies have mainly examined single-source effects, while the dynamics of multi-source warnings under conditions of consistency and conflict remain insufficiently explored. Based on two online experiments with residents in China (N = 599), this study investigates the psychological processes of information trust and anticipated regret that shape how the public responds to different warning sources and consistency. The findings show that government-issued warnings generate higher trust than AI warnings, and consistent messages from both sources further enhance trust and protective intentions. In conflicting warning scenarios, anticipated regret becomes prominent, contributing to individuals’ tendency to follow the higher-level warning. The results support a dual-pathway conceptual framework of decision making in multi-source warning contexts, where cognitive trust grounded in institutional authority and technological support coexists with emotional motivation driven by anticipated regret. This study fills an empirical gap in multi-source warning research and offers theoretical and practical insights for building disaster warning systems that integrate institutional credibility with emerging AI technologies.
The rapid change of heat patterns resulting from human-induced climate change are affecting the heat thresholds used in early warning systems. In Indonesia, the current heat early warning guidelines employ a nationwide fixed threshold that, considering current heat trends, may fail to adequately capture localized heat stress and vulnerability. In such conditions, operational meteorologists (OMs) authorized to issue warnings may face challenges and dilemmas when implementing heat early warning systems (HEWS) using existing thresholds. This study explores the decision-making processes and challenges faced by OMs using a cross-sectional survey involving 140 OMs serving in urban coastal areas across 31 provinces in Indonesia using hierarchical ordinary least squares (OLS) regression. We identified terminology ambiguity and reliance on weather station data as the primary technical and institutional determinants of dissemination challenges. Furthermore, the decision to issue a warning is predicted by past experience with successful warnings and the adoption of a “liberal” warning philosophy. These findings also provide conceptual insights into how institutional constraints and personal judgment interact to shape HEWS operations and contribute to the broader theoretical understanding of OMs’ decision making under uncertainty.
This study reflects on the probability of observing an extreme event of interest within a finite dataset, whether derived from observations or model simulations, to inform risk assessment or climate adaptation efforts. To do so, we adopt the concept of engineering reliability, which is defined as the probability that a system remains in a satisfactory state, to assess the reliability of extreme events inferred from a dataset, whether this is from observations or model simulations. This assessment links the number of available observations or simulations to the low frequency of the event, providing a quantitative measure of confidence in our ability to observe or simulate such events over a given time horizon. This approach offers a fresh perspective on the interpretation of an extreme event, where the rarity of an event is considered not only in terms of its frequency but also relative to the length of the dataset used. Our reflections aim to guide preparedness for future extremes and highlight the scientific challenges inherent in their prediction and projection. We emphasize that while large ensembles are essential to overcome the limitations of historical observations, they should be used with caution to avoid overconfidence arising from underlying modeling assumptions. Finally, we stress that statistical extrapolation, whether it is parametric or non-parametric, is unavoidable, as the link between event frequency and the definition of extremes cannot be eliminated.
The early warning system (EWS) localization agenda has been ignored in the Early Warnings for All (EW4All) initiative. To date, no studies have analyzed the implementation of local EWS through the lens of institutional capacities. The scientific innovation of this study lies in proposing and applying a new framework called Implementation Capacity of Local Early Warning Systems (ICLEWS). Using an institutional capacity approach that focuses on the technical-administrative and political-relational aspects, ICLEWS examines local EWS implementation across the four warning subsystems (risk knowledge, monitoring, communication, and preparedness). The innovative framework comprises seven dimensions to monitor EWS implementation at the local level: (1) human resources of local disaster risk management (DRM) agency; (2) financial resources of local DRM agency; (3) risk mapping; (4) weather and subseasonal forecast, and rainfall monitoring; (5) risk communication strategies; (6) social participation; and (7) local and regional EWS governance. Based on a 2025 national survey completed by 41
Populations in sparsely populated areas are vulnerable to external shocks due to limited proximal infrastructure and low adaptive capacity. In Northern Australia, and for other sparsely populated areas where human habitation is at the “edge” and attracting and retaining people is a challenge, extreme weather events are commonplace. In this study we analyze historical population and hazard data for Northern Australia to identify and profile the scale of population loss across settlement types from extreme weather events. Understanding and charting how past events have impacted the population numbers of settlements in sparsely populated areas is vital for identifying future economic, demographic, and social impacts under climate change. We found that impacts differ according to the type of settlement and hazard, with larger and more economically diversified settlements impacted primarily by floods and cyclones. Findings in this study provide practical insights into the use of census and hazard data to inform more effective and locally tailored policy and planning strategies for disaster mitigation and climate change adaptation for settlements across Northern Australia.
Tropical cyclones pose increasing risks in Southern Africa, with impacts shaped not only by hazard intensity but also by underlying vulnerabilities and disaster risk management (DRM) capacity. This study examines preparedness, response, and recovery following Cyclone Ana (2022) in Nyanga District, Zimbabwe, drawing on qualitative data from focus group discussions, key informant interviews, and field observations. Findings reveal persistent gaps in community preparedness, early warning communication, and institutional coordination, with DRM systems remaining largely reactive, centralized, and unevenly implemented. Although warning information was widely disseminated, its technical and generalized nature limited local interpretation and did not consistently prompt protective action. Recovery efforts were similarly constrained by short-term funding cycles and uneven targeting, resulting in variable outcomes across communities. At the same time, strong local capacity for collective action was evident, with communities mobilizing indigenous knowledge, social networks, and mutual support to address immediate needs. Building on these findings, the article positions Cyclone Ana as a baseline for assessing subsequent reforms, including the introduction of impact-based forecasting, anticipatory action frameworks, and strengthened contingency planning. While these developments represent important progress, significant challenges remain in operationalizing them at the local level. The study identifies a critical implementation gap between policy commitments and practice, underscoring the need to better align legislative reform, financing mechanisms, and institutional capacity with community-centered approaches. By linking empirical insights to broader debates on disaster risk governance, the article contributes to understanding how DRM systems can transition from reactive response toward more adaptive, inclusive, and resilience-oriented pathways.
Publicly listed companies are increasingly disclosing climate-related financial risks to their businesses since the promulgation of the Task Force on Climate-related Financial Disclosures (TCFD), and more recently under the climate-related disclosures issued by the International Sustainability Standards Board (ISSB). Hence, financial risk exposures associated with geographically-distributed operations and business activities will need to be quantified and benchmarked. While extant research on individual companies or facilities has been available, no prior methodologies have explored these systemic risks based on industry sector classification in the context of listed indices. In this study, we analyzed the characteristics of corporate financial flood risks across regions and industry sectors for the constituent components in the Nikkei 225 Index, representing 225 companies and more than 18,000 facilities. The modeling approach integrates hydrological datasets with multiple climate models and financial records of corporations and facilities. Using estimated property damage (using plant, property, and equipment or PP E investment proxies) and revenue losses at the facility, corporate, and industry scales, the expected annual damage (EAD) exceeds USD 8.2 billion by 2030. Approximately 60
While international disaster risk reduction frameworks place household preparedness at the center of resilience strategy, substantial proportions of people living in hazard-prone areas remain underprepared. We develop an integrated framework combining protection motivation theory, the capabilities approach, and social contract theory to explain why some residents prepare while others do not. Drawing on survey data from 159 adults and 27 qualitative interviews, we analyze how sociodemographic characteristics, hazard experience, risk perception, residence status, neighborhood, and institutional trust relate to preparedness in Squamish. Preparedness levels were modest, with respondents completing a mean of 4.21 out of 10 preparedness actions. Lower formal education was the strongest negative predictor, while hazard experience and hazard concern were consistent positive predictors. Perceived hazard likelihood did not predict preparedness and, in some resident-only models, was negatively associated with participation in drills and training. We conclude that in Squamish, preparedness is shaped by an individual’s hazard concern, what they are materially able to do, and whether they believe institutions will uphold their side of the social contract.
Disaster and emergency management (DEM) systems are increasingly characterized by distributed authority, multi-agency interdependence, and complex hazard environments. While prior research has reframed DEM as a complex adaptive system, less attention has been given to how coherence, adaptive coordination, and legitimacy are sustained across decentralized networks where no single actor holds complete mandate or authority. This study examines system stewardship as a meta-governance function capable of addressing this gap. Drawing on constructivist networked grounded theory, the study analyzes 38 semistructured interviews with 40 practitioners across Aotearoa New Zealand’s DEM system. The analysis revealed four themes: governing a distributed system; enabling coordination through trust and inclusive participation; sharing information and sustaining learning; and sustaining capability over time. The study integrates these findings into a three-pillar framework of system stewardship comprising relational, informational, and institutional dimensions, each representing a distinct but interdependent set of enabling conditions for adaptive coordination. The study argues that sustaining DEM performance requires moving beyond episodic leadership during events toward the deliberate sustaining of these enabling conditions across the system as a whole. In doing so, it contributes a complexity-informed governance framework applicable to DEM systems operating under conditions of decentralization, uncertainty, and compound risk.
Coastal cities in southeastern China face increasing threats from typhoon-induced compound disasters (for example, torrential rainfall, urban waterlogging, and storm surges) that can cascade into interconnected disaster chains under climate change and rapid urbanization. However, dynamic multi-scale assessments of resilience to such compound disasters remain limited. This study develops an integrated framework that combines multi-scale geospatial analysis with explainable machine learning (XGBoost-SHAP). Using Fujian Province as a case study, we assess typhoon disaster chain urban resilience (TDCUR) in 2010, 2015, and 2020 across grid, administrative unit, and watershed scales, characterize spatiotemporal patterns, and apply XGBoost-SHAP as a post hoc diagnostic to summarize nonlinear indicator-TDCUR association patterns and their spatial concentration under the predefined TDCUR framework. The results indicate that: (1) Provincial TDCUR increased by 6.9
Increasingly frequent disasters and the heightened vulnerability of older adults highlight the need for equitable access to lifesaving resources such as evacuation shelters and emergency supply storage. This study evaluates whether the spatial accessibility of these facilities aligns with the distribution of older residents in core and non-core areas of Pudong New Area, Shanghai Municipality. We develop a spatial equity framework employing bivariate local Moran’s I, Lorenz curves, and Gini coefficients to quantify alignment and inequity, using data from the First National Survey on Natural Disaster Risks (2020–2022). Our analysis uncovers a pronounced spatial disconnect. Both shelters and supply storage cluster in the urban core, leaving older adults in peripheral areas with substantially lower access. Shelters correspond more closely to general population clusters than to older adult clusters, and this misalignment worsens as evacuation time thresholds increase. Emergency supply storage exhibit a more even distribution and a positive spatial correlation with ageing communities, yet Lorenz curve analysis reveals persistent inequities outside the core. These results demonstrate that current facility siting amplifies disaster inequities for older adults. We therefore recommend that emergency planning move beyond per capita targets to incorporate spatial accessibility metrics and age-sensitive siting strategies, ensuring that vulnerable populations have timely access to shelter and supplies when disasters occur.
The escalating threats of climate change and rapid urbanization to urban sustainability have intensified the urgency for effective flood recovery strategies, particularly regarding critical infrastructure such as road networks. Shenzhen, a megacity frequently hit by short-duration heavy rainfall and typhoon-induced storms, faces high flood risk that often causes severe road network disruption. This study proposes an integrated approach bridging flood simulation, loss assessment, and isochrone-based accessibility analysis to evaluate and mitigate flood impacts across the road network of Shenzhen City under various rainfall return periods. The flood simulation combines the Soil Conservation Service Curve Number (SCS-CN) model for runoff estimation with a DEM-based water accumulation algorithm. The results demonstrate that: (1) Increasing rainfall return periods lead to a progressive expansion of inundation areas, predominantly affecting commercial, educational, and industrial sectors, while road network loss based on the inundation depths escalates rapidly under a 10-year return period rainfall and stabilizes beyond the 20-year threshold; (2) Isochrone analysis reveals that accessibility to emergency centers undergoes accelerated decay for rainfall return periods shorter than 20 years, with the most pronounced degradation observed along the 4- to 6-min isochrones; (3) Post-disaster recovery strategies prioritizing isochrone decay directions outperform those based on road hierarchy, particularly in the 4-min critical zones. This research provides robust analytical tools and insights for identifying vulnerable road sections and nodes during flood events, facilitating the prioritization of road network recovery.
Climate-driven changes in precipitation, temperature, and runoff are intensifying nutrient pollution hazards in watersheds, particularly in urban-rural transitional zones characterized by fragmented land use and sparse monitoring. Predicting these climate-sensitive nonpoint source pollution (NPSP) dynamics is difficult because multisource inputs and rapidly evolving spatial patterns challenge conventional models. This study developed an artificial intelligence (AI) framework that integrates machine learning with remote sensing data assimilation to assess nutrient pollution in rapidly urbanizing basins. A random forest regression model, trained on 15 GIS-based predictors, serves as the core predictor for total nitrogen (TN) and total phosphorus (TP) concentrations. An ensemble Kalman filter (EnKF) assimilates remote sensing inversion (RSI) products into the model for dynamic updates. Applied to the Wenruitang River watershed in eastern China, the framework reduces TP prediction errors by up to 52.9
Community formation in socio-spatial human networks is an important mechanism through which populations cope with and mitigate the impacts of extreme weather hazards. However, limited research has examined the latent network characteristics that shape community formation in human mobility networks during natural hazard-related disasters. In this study, we analyzed human mobility networks in Harris County, Texas, during the managed power outages associated with Winter Storm Uri in 2021 to detect communities and evaluate their underlying characteristics. Specifically, we examined three dimensions of the detected communities: hazard exposure heterophily, sociodemographic homophily, and social connectedness strength. The results show that population movements were shaped by sociodemographic homophily, heterophilic hazard exposure, and social connectedness strength. We also found that communities containing a larger share of high-impact areas tend to drive population movements toward areas with weaker social connectedness. These findings highlight key characteristics that shape community formation in human mobility networks during hazard response. More broadly, the findings suggest that power utility operators should account for the characteristics of socio-spatial human networks when designing managed power outage strategies.
Flood early warning systems (FEWS) are established globally to support communities in building resilience to flooding and in preparing for and responding to disasters. However, amid climatic and technological change, it is unclear how FEWS might evolve to remain effective. We review resilience as the capacity to “bounce back” and introduce transilience as the capacity to “bounce forward” in the context of disaster risk reduction. We argue that FEWS can help flood-vulnerable communities innovate and find new opportunities to transform their flood risk management for the better (that is, transilience). Technological evolution, advancing flood monitoring and forecasting, creates opportunities for communities to improve the entire flood early warning value chain. We use case studies to show how communities have leveraged technology to address changing flood impacts and reduce vulnerabilities that often precipitate or exacerbate flood disasters. We also review how global multi-hazard early warning initiatives and priorities increasingly reflect transilience. This perspective contributes to conversations about the importance of making communities central to FEWS design and using technology to better understand and meet community needs in disaster risk management.
Debris flows pose significant threats to mountainous regions, necessitating accurate activity assessments for effective disaster mitigation and risk management. At a regional scale, debris flow studies have predominantly focused on susceptibility, without adequately addressing frequency and magnitude of these events. However, growing demands for hazard mitigation call for more detailed and comprehensive debris flow activity assessments. This study developed an integrated spatiotemporal debris flow activity assessment framework by combining spatial susceptibility modeling, temporal probability estimation, and potential event magnitude estimation. The assessment results for the Eastern Himalayan Syntaxis successfully identified historically active watersheds, including those impacted by catastrophic debris flows such as the 1953 Guxiang Glacier event. The study area was classified into five activity levels, with 37.8
Self-built buildings in rural areas exhibit varying levels of seismic performance, and despite the transformative development in the region over the past decades, such buildings continue to exist in Shanghai Municipality. This study conducted field research and theoretical calculations to gain an overall understanding of the seismic performance of these buildings. In this study, 819 self-built buildings were classified according to their age, field research was conducted to test mortar strength and connection reliability, and the seismic performance of these buildings was analyzed. The results confirm that a considerable stock of unreinforced masonry buildings constructed in the peri-urban townships of Shanghai, especially those erected prior to 1990, exhibit markedly deficient seismic performance and are highly vulnerable to collapse or severe damage from earthquakes. This poses a significant threat to catastrophic casualties and disproportionate economic losses. Consequently, a systematic, high-resolution survey coupled with targeted seismic retrofitting is urgently required to mitigate earthquake-induced disaster risks in building inventories.
Emergency evacuation signage, particularly exit signs installed along building corridors and evacuation routes, is critical for life safety during emergencies. Traditional signage design generally prioritizes maximizing visibility and spatial coverage. However, excessive sign installation may introduce visual clutter and interference among overlapping guidance cues, thereby limiting further improvements in evacuation performance. Therefore, a two-stage method is proposed to explore the optimal balance between luminous performance, installed quantity, and spatial distribution of exit signs for efficient evacuation guidance. First, a controlled experiment involving 30 participants was conducted to establish a U-shaped psychophysical relationship between achromatic contrast and the maximum recognition distance (MRD). The relationship is largely independent of both ambient illumination and observer gender. Second, the MRD data were integrated into Pathfinder to systematically assess how occupant density and sign quantity jointly influence evacuation efficiency. The results demonstrate that the optimal signage configuration shows significant density dependence. In the tested corridor scenario, three signs yield the shortest evacuation time at high occupant density, whereas fewer signs achieve comparable or better evacuation performance at medium and low densities. Moreover, a saturation effect is identified in the benefits of MRD improvement. Once the threshold is exceeded, further increases in MRD yield little or no additional improvement in evacuation efficiency. These findings are discussed in terms of hypothesized interference from overlapping guidance cues and congestion from overlapping signs. The study suggests that future signage design guidelines may benefit from incorporating density responsive and performance-based evaluation. Further experimental and field validations are needed to support broader standardization.
Coastal flooding poses a significant threat to China’s coastal zones, driven by the combined effects of global climate change and rapid urbanization. However, owing to notable discrepancies in data accuracy, model construction, and parameter settings, the applicability of findings from existing regional studies is limited when extrapolated to broader macroscale risk assessments. Therefore, this study quantitatively assessed the risk of casualties and economic losses in China’s coastal areas under different return period scenarios using multi-source spatial data based on the LISFLOOD-FP hydraulic model with a fine 30-m spatial resolution. Extreme water levels corresponding to return periods of 20-, 50-, 100-, 200-, and 500-years estimated by historical observations from 65 tide-gauge stations were employed as boundary conditions to simulate the flood inundation. Taking inundation depth-damage curves into consideration, we subsequently quantified the spatial distribution of the population casualty rates and economic losses under different inundation scenarios. Our assessment results reveal pronounced spatial characteristics in coastal flood risk, with the most severe impacts concentrated in low-lying urban areas, such as the Bohai Rim regions, the Yangtze River Delta, and the Pearl River Delta. The results indicate that under the 500-year return period inundation scenario, the total flooded coastal area across 11 provinces reaches 20,983 km2, with the number of casualties amounting to 176,000 and economic losses totaling CNY 303.8 billion yuan (about USD 42.2 billion). The high-resolution flood risk maps developed in this study provide spatial information and data support for national-scale coastal management, disaster risk reduction, and land-use planning in China’s coastal areas.
Effective disaster risk management requires bottom-up approaches that leverage citizen feedback to gain insights into disaster events on field. While much existing research in this domain focuses on social media platforms, the rich potential of government-operated digital feedback systems in capturing citizen experiences and concerns during disaster events remains underexplored. To address this gap, this study utilizes sentiment analysis to evaluate complaints submitted on the Pakistan Citizen Portal concerning flood and heat events. A hybrid methodology is adopted, combining VADER, SVM, LR, and BERT models to uncover sentiment trends and emotions expressed by citizens. Additionally, Latent Dirichlet Allocation is employed to identify key topics of concern, enhancing decision making and resource allocation. Our findings demonstrate the effectiveness of BERT in capturing sentiment trends while addressing the linguistic and cultural nuances inherent to the data. Emotions are detected based on Ekman’s six basic emotions, with sadness and anger emerging as the predominant ones. Key topics identified include property damage from floods and electricity problems related to heat events. The insights derived from this analysis provide a policy-relevant assessment of disaster risk management in Pakistan, illustrating how digital citizen feedback can inform resource allocation, preparedness planning, and communication strategies. By translating citizen sentiment into actionable insights, this study demonstrates the potential of digital platforms like the Pakistan Citizen Portal to enhance community resilience, strengthen preparedness, and support more responsive and inclusive disaster governance.
