What’s the worst that could happen? Recent floods are often claimed to be beyond our imagination. Imagination is the picturing of a situation in our mind and the emotions that we connect with this situation. But why is this important for disasters? This survey found that when we cannot imagine a devastating flood, we are not preparing in advance. Severe-weather forecasts and warnings need to advance in order to trigger our imagination of what might happen and enable us to start preparing.

This study utilizes the global copula Bayesian model averaging technique for accurate and reliable flood modeling, especially in coastal regions. By integrating multiple precipitation datasets within this framework, we can effectively address sources of error in each dataset, leading to the generation of probabilistic flood maps. The creation of these probabilistic maps is essential for disaster preparedness and mitigation in densely populated areas susceptible to extreme weather events.

Seismic risk management uses reference earthquake scenarios, but the criteria for selecting them do not always consider consequences for exposed assets. Hence, we adopt a definition of representative scenarios associated with a return period and loss level to select such scenarios among a large set of possible earthquakes. We identify the scenarios for the residential-building stock and power supply in Valparaíso and Viña del Mar, Chile. The selected scenarios depend on the exposed assets.

Landslides are complex phenomena causing important impacts in vulnerable areas, and they are often triggered by rainfall. Here, we develop a new approach that uses information on the temporal clustering of rainfall, i.e. multiple events close in time, to detect landslide events and compare it with the use of classical empirical rainfall thresholds, considering as a case study the region of Lisbon, Portugal. The results could help to improve the prediction of rainfall-triggered landslides.

Iranian insurers have been offering earthquake coverage since the 1990s. However, despite international best practices, they still do not use modern methods for risk pricing and management. As such, they seem to be accumulating seismic risk over time. This paper examines the viability of this market in Iran by comparing the local market practices with international best practices in earthquake risk pricing (catastrophe modeling) and insurance risk management (European Solvency II regime).

Snowpack simulations are increasingly employed by avalanche warning services to inform about critical avalanche layers buried in the snowpack. However, validity concerns limit their operational value. We present methods that enable meaningful comparisons between snowpack simulations and regional assessments of avalanche forecasters to quantify the performance of the Canadian weather and snowpack model chain to represent thin critical avalanche layers on a large scale and in real time.

This study investigates inconsistencies in impact force as part of extended column tests (ECTs). We measured force-time curves from 286 practitioners in Scandinavia, Central Europe, and North America. The results show a large variability in peak forces and loading rates across wrist, elbow, and shoulder taps, challenging the ECT's reliability. 

Modelling tsunami generation due to a rapid submarine earthquake is a complex problem. Under a variety of realistic conditions in a subduction zone, we propose and test an efficient solution to this problem: a tool that can compute the generation of any potential tsunami in any ocean in the world. In the future, we will explore solutions that would also allow us to model tsunami generation by slower (time-dependent) seafloor displacement.

Weather prediction uncertainties can be described as sets of possible scenarios – a technique called ensemble prediction. Our machine learning technique translates them into more easily interpretable scenarios for various users, balancing the detection of high precipitation with false alarms. Key parameters are precipitation intensity and space and time scales of interest. We show that the approach can be used to facilitate warnings of extreme precipitation.

The study proposes a new framework, named FLEXTH, to estimate flood water depth and improve satellite-based flood monitoring using topographical data. FLEXTH is readily available as a computer code, offering a practical and scalable solution for estimating flood depth quickly and systematically over large areas. The methodology can reduce the impacts of floods and enhance emergency response efforts, particularly where resources are limited.

Despite the visible flood damage, mental health is a growing concern. Yet, there is limited data in Germany on mental health impacts after floods. A survey in a heavily affected region revealed that 28 % of respondents showed signs of post-traumatic stress disorder 1 year later. Risk factors include gender, serious injury or illness due to flooding, and feeling left alone to cope with impacts. The study highlights the need for tailored mental health support for flood-affected populations.

Early warning is essential to minimise the impact of flash floods. We explore the use of highly detailed flood models to simulate the 2021 flood event in the lower Ahr valley (Germany). Using very high-resolution models resolving individual streets and buildings, we produce detailed, quantitative, and actionable information for early flood warning systems. Using state-of-the-art computational technology, these models can guarantee very fast forecasts which allow for sufficient time to respond.

Associating extreme weather events with changes in the climate remains difficult. We have explored two ways these relationships can be investigated: one using a more common method and one relying solely on long-running records of meteorological observations.

Our results show that while both methods lead to similar conclusions for two recent weather events in Sweden, the commonly used method risks underestimating the strength of the connection between the event and changes to the climate.

This study delves into the dynamics of vulnerability within a multi-hazard context, proposing an enhanced impact-chain-based framework that analyses the augmentation of vulnerability. The case study refers to the flood events and the COVID-19 pandemic that affected Romania (2020–2021). The impact chain shows that (1) the unforeseen implications of impacts, (2) the wrongful action of adaptation options, and (3) inaction can form the basis for increased vulnerability.

Our study explored how seasonal flood forecasts could enhance insurance premium accuracy. Insurers traditionally rely on historical data, yet climate fluctuations influence flood risk. We employed a method that predicts seasonal floods to adjust premiums accordingly. Our findings showed significant year-to-year variations in flood risk and premiums, underscoring the importance of adaptability. Despite limitations, this research aids insurers in preparing for evolving risks.