Landslides represent a major natural hazard and are often triggered by typhoons. We present a new 2D model computing the respective role of rainfall infiltration, atmospheric depression and groundwater in slope stability during typhoons. The results show rainfall is the strongest factor of destabilisation. However, if the slope is fully saturated, near the toe of the slope or during the wet season, rainfall infiltration is limited and atmospheric pressure change can become the dominant factor.

One of the most severe and impactful urban wildfire events in Croatian history has been reconstructed and analyzed. The study identified some important meteorological influences related to the event: the synoptic conditions of the Azores anticyclone, cold front, and upper-level shortwave trough all led to the highest fire weather index in 2017. A low-level jet, locally known as bura wind that can be explained by hydraulic jump theory, was the dynamic trigger of the event.

We quantify the influence of wave–wind characteristics, offshore water level and sea level rise (projected up to 2200) on the occurrence of flooding events at Gâvres, French Atlantic coast. Our results outline the overwhelming influence of sea level rise over time compared to the others. By showing the robustness of our conclusions to the errors in the estimation procedure, our approach proves to be valuable for exploring and characterizing uncertainties in assessments of future flooding.

Natural hazards such as earthquakes, landslides, and flooding do not always occur as stand-alone events. After the 2008 Wenchuan earthquake, a co-seismic landslide blocked a stream in Hongchun. Two years later, a debris flow breached the material, blocked the Min River, and resulted in flooding of a small town. We developed a multi-process model that captures the full cascade. Despite input and process uncertainties, probability of flooding was high due to topography and trigger intensities.

This is a comprehensive approach to risk assessment that considers the dynamic relationship between loss and damage. The study combines physical risk with social science to mitigate the disaster caused by earthquakes in Nepal, taking socioeconomical parameters into account such that the risk estimates can be monitored over time. The main objective is to recognize the cause of and solutions to seismic hazard, building the interrelationship between individual, natural, and built-in environments.

We develop probabilistic earthquake forecasts that include different spatial information (e.g. fault locations, strain rate) using a point process method. The performance of these models is tested over three different periods and compared with existing forecasts. We find that our models perform well, with those using simulated catalogues that make use of uncertainty in model parameters performing better, demonstrating potential to improve earthquake forecasting using Bayesian approaches.

Automated snow avalanche terrain mapping provides an efficient method for large-scale assessment of avalanche hazards, which informs risk management decisions for transportation and recreation. This research reduces the cost of developing avalanche terrain maps by using satellite imagery and open-source software as well as improving performance in forested terrain. The research relies on local expertise to evaluate accuracy, so the methods are broadly applicable in mountainous regions worldwide.

We propose a model named variable fuzzy set and information diffusion (VFS–IEM–IDM) to assess the dynamic risk of compound hazards, which takes into account the interrelations between the hazard drivers, deals with the problem of data sparsity, and considers the temporal dynamics of the occurrences of the compound hazards. To examine the efficacy of the proposed VFS–IEM–IDM model, a case study of typhoon–rainstorm risks in Shenzhen, China, is presented.

We showed the effects of altering the choice of cumulus schemes, surface flux options, and spectral nudging with a high level of sensitivity to cumulus schemes in simulating an intense typhoon. We highlight the advantage of using an ensemble of cumulus parameterizations to take into account the uncertainty in simulating typhoons such as Haiyan in 2013. This study is useful in addressing the growing need to plan and prepare for as well as reduce the impacts of intense typhoons in the Philippines.

On 7 February 2021, a large rock–ice avalanche occurred in Chamoli, Indian Himalaya. The resulting debris flow swept down the nearby valley, leaving over 200 people dead or missing. We use a range of satellite datasets to investigate how the collapse area changed prior to collapse. We show that signs of instability were visible as early 5 years prior to collapse. However, it would likely not have been possible to predict the timing of the event from current satellite datasets.

We evaluate through first-order kinetic energy models, the minimum volume and mass of a pyroclastic density current generated at the Aso caldera that might affect any of five distal infrastructure sites. These target sites are all located 115–145 km from the caldera, but in well-separated directions. Our constraints of volume and mass are then compared with the scale of Aso-4, the largest caldera-forming eruption of Aso.

We promote a new method to detect waterlogging disasters. Residents are directly affected by waterlogging, and we can collect their comments on social networks. Compared to official-authentication and personal-certification users, the microblogs posted by general users can better show the intensity and timing of waterlogging. Through text and sentiment features, we can separate microblogs with waterlogging information from other ones and mark high-risk regions on maps.

We present a new landslide inventory for the 2021, M 7.2, Haiti, earthquake. We compare characteristics of this inventory with those of the 2010 seismically induced landslides, highlighting the much larger total area of 2021 landslides. This fact could be related to the larger earthquake magnitude in 2021, to the more central location of the fault segment ruptured in 2021 with respect to coastal zones, and/or to possible climatic preconditioning of slope failures in the 2021 affected area.

It is impossible to protect all issues against flood risk. To prioritise protection, economic analyses are conducted. The French Ministry of the Environment wanted to make available damage functions that we have developed for several sectors. For this, we propose a methodological framework and apply it to the model we have developed to assess damage to agriculture. This improves the description, validation, transferability and updatability of models based on expert knowledge.

The study presents the application of high-resolution coastal modelling for wave hindcasting on the Emilia-Romagna coastal belt. The generated coastal databases which provide an understanding of the prevailing wind-wave characteristics can aid in predicting coastal impacts.

This study advances previous models through machine learning and multi-sensor-verified results. Using spatial and meteorological data from the study area (Suncheon–Wanju Highway in Gurye-gun), the amount and location of black ice were modelled based on system dynamics to predict black ice and then simulated with the geographic information system (m²). Based on the model results, multiple sensors were buried at four selected points in the study area, and the model was compared with sensor data.

Droughts represent a particularly complex natural hazard and require explorations of their multiple causes. Part of the complexity has roots in the interaction between the continuous changes in and deviation from normal conditions of the atmosphere and the land surface. The exchange between the atmospheric and surface conditions defines feedback towards dry or wet conditions. In semi-arid environments, energy seems to exceed water in its impact over the evolution of conditions, favoring drought.

Wildfire spatial patterns are determined by fire ignition sources and vegetation fuel moisture. Fire ignitions can be mediated by humans (owing to proximity to human infrastructure) or caused by lightning (owing to fuel moisture, average annual rainfall and local weather). When moisture in dead vegetation is below 20 % the probability of a wildfire increases. The results of this research enable accurate spatial mapping of ignition probability to aid fire suppression efforts and future research.

Avalanches are natural hazards that threaten people and infrastructure. With climate change, avalanche activity is changing. We analysed the change in frequency and size of avalanches in the Krkonoše Mountains, Czechia, and detected important variables with machine learning tools from 1979–2020. Wet avalanches in February and March have increased, and slab avalanches have decreased and become smaller. The identified variables and their threshold levels may help in avalanche decision-making.

I mapped over 5000 landslides triggered by a moment magnitude 6.0 earthquake that occurred in 2015 in the Sabah region (Malaysia). I analyzed their number, dimension and spatial distribution by dividing the territory into 1 km² cells. I applied the Environmental Seismic Intensity (ESI-07) scale, which allows the categorization of earthquake damage due to environmental effects. The presented approach promotes the collaboration among the experts in landslide mapping and in ESI-07 assignment.