Bangladesh’s power grid is highly vulnerable to tropical cyclones. Using nearly a decade of daily data, we show landfalling storms cut national electricity supply by about 20 % on the day, with coastal regions hit hardest (up to 38 %). Damage comes from high winds, storm surge and heavy rain. Power imports from India often can’t help during big events because both areas are struck together. Building sturdier, climate-resilient infrastructure is essential.

Flooding causes not only physical damage but also significant psychological impacts that are often overlooked due to measurement challenges. This study evaluates how households and businesses value these impacts using willingness-to-pay methods in Peninsular Malaysia. Results show flood characteristics (duration) and social (family size) strongly influence psychological stress. Findings highlight the need for improved flood management and community resilience to reduce long-term social impacts.

Most studies of compound extremes assume stationary climate conditions. Here, we combine high-resolution regional climate modeling with non-stationary statistical methods to assess future changes in compound extremes over southeastern China. Our results demonstrate that non-stationary models more accurately capture shifts in the evolution of these events, suggesting that conventional approaches may systematically underestimate future risks.

Winter North Atlantic storms cause significant financial losses and damage in Europe. This study shows that modes of seasonal large-scale climate variability called the North Atlantic Oscillation and East Atlantic Pattern modulate the exposure to cyclone related extreme wind, precipitation and storm surge hazards across many parts of Europe. The results have the potential to be combined with skilful seasonal climate forecasts of climate modes to inform the insurance sector.

Coastal erosion is a global concern. Traditional approaches ignore the directional force of waves. The new approach here presented uses Longitudinal Wave Power (LWP) to better link wave conditions to coastal changes. Through numerical simulations and extreme analysis, LWP was concluded to be a superior variable for assessing coastal natural hazards, achieving 86 % prediction accuracy. This tool provides planners with a more reliable way to forecast shifting shorelines and manage coastal zones.

Obtaining probabilistic flood maps with numerical models is very time-consuming. Deep learning models can speed this up, but their predictions are difficult to verify without reference data, and they ignore structures like dikes or canals. This work introduces a mass-based validation measure to assess prediction plausibility and adapts a graph-based model to include hydraulic structures, enabling realistic, large-scale probabilistic flood mapping in the Netherlands.

Postfire debris flows are fast-moving landslides that pose a significant risk to downstream communities around the world. Accurately predicting how large postfire debris flows will be, before they occur, allows us to better understand the potential effects of future events. In this study, we develop a new method for predicting postfire debris-flow volume in the western United States. Results show that this new method outperforms existing volume models and can improve postfire hazard assessments.

Hurricanes are a destructive natural hazard. Historically, however, their Saffir–Simpson categories and losses are not well correlated. We combined hazard, exposure, and vulnerability data to predict losses from landfalling hurricanes for the United States. Our model significantly reduces errors between predicted and observed losses and is more skilful than hazard-only predictions. Additionally, we developed a novel loss-based hurricane classification scheme to aid risk management.

We propose a spatially weighted non-negative tensor factorization method for extracting pre-seismic geomagnetic anomalies from multi-station data. By integrating spatial constraints between observation stations and the seismic source into the decomposition framework, the method improves the physical interpretability of the extracted anomalies and offers a new methodological approach for seismic anomaly analysis.

We analyzed long-term groundwater, precipitation, and storm-surge records across coastal New York and Connecticut to estimate how often these hazards occur together. Overlap is highest along southwestern Long Island and western coastal Connecticut during the colder months, when groundwater is higher and mid-latitude cyclones are more common. Results from this study can support better preparation for coastal storms by taking into consideration the compounding effects of different flood drivers.

Reliable flood estimates are vital for effective flood risk management, yet short observation records often limit accuracy. We present a new approach that adds realistic "what if" flood scenarios from nearby similar catchments. Applied to thousands of small German basins, our method improves flood frequency estimates, reducing uncertainty and supporting more robust planning for extreme events.

This study investigates the main active faults located within the southwestern Valencia Trough, an offshore region east of the Spanish coast. Utilizing subsurface data, we identify and characterize the 3D geometry of several of these faults for the first time. Given that active faults pose a significant natural hazard owing to their potential to generate earthquakes, we also assess the seismic potential of the faults within the southwestern Valencia Trough.

In August 2024, a powerful storm hit Romania’s Black Sea coast, breaking rainfall records. We used a mix of ground and satellite sensors to track the storm’s development and impacts. The data revealed clear signs of intense rainfall, lightning, and ground vibrations likely linked to storm activity. Our study shows that combining different types of sensors can improve how we monitor extreme storms and may help in building better early-warning systems in coastal areas.

Mantle-derived gases can seep to the surface before a volcanic eruption, but detecting them is difficult when emissions are weak. We used seismic noise from individual stations at Cumbre Vieja Volcano, La Palma, to monitor shallow gas accumulation being modulated by atmospheric cycles. We find clear links between pressure variations and gas release, with a marked increase months before the 2021 eruption, highlighting potential for early warning and hazard assessment.

We applied a fast statistical model to estimate future extreme storm surges in Europe using data from 17 climate models – about twice as many as in past studies. Results show robust regional patterns – decreases in the Mediterranean and Moroccan coast, increases in the Irish Sea and Gulf of Finland – with high uncertainty in other areas. Out results increase our knowledge on future storm surge uncertainties, needed for informed coastal planning.

Atmospheric rivers, narrow and long corridors of intense horizontal moisture transport, may produce heavy precipitation where they are forced to rise over the orography. This has been recently shown to occur also in the Mediterranean basin, especially over the Alps and Apennines in Italy. The present study analyses the presence of atmospheric rivers in this area, in the period 1961–2024, and reveals a strong connection with extreme rainfall over northern-central Italy.

We introduce TerraTrack, an open-source tool for detecting and monitoring slow-moving landslides using Sentinel-2 data. It automates image acquisition, landslide identification, and time-series generation in an accessible and cloud-based workflow. TerraTrack supports early warning, complements interferometric synthetic aperture radar (InSAR), and offers a scalable solution for landslide hazard identification and monitoring.

In this study, we assess the impact of culvert blockages on flash flood dynamics. Applying the 2D Shallow Water Equations, we model multiple flash flood and blockage scenarios within a steep catchment in central Germany. The results emphasize the need for accurate culvert representation in hydrodynamic models and demonstrate how scenario-based blockage modeling can support the identification of critical infrastructure and the development of targeted mitigation measures.