This work represents the extension of a previous investigation of the 1783 lethal landslide–tsunami, provoking more than 1500 casualties in the coastal town of Scilla. The effects of the tsunami are assessed here over a wider domain and compared with the available observations from historical reports. The eastern end of Sicily (Capo Peloro) is the most affected site in the far-field. A tentative reconstruction of the coeval topography of the area is also realized, with interesting results.

Our ability to manage estuaries is currently limited by a poor understanding of how they will evolve into the future. This study explores flooding conditions at two US Pacific estuaries as controlled by changing climate. The hazard is characterized using a variety of models that track oceanic, atmospheric, and hydrologic forcing at decadal scales. It is found that flood surface height varies significantly across estuaries and can be expected to change in complex ways moving into the future.

This is the first scientific report of the flash flood of 20 August 2018 on “Raganello Gorge” (Southern Italy), an extreme event with rather specific features (very localized in space and time), which unfortunately caused 10 victims. The meteo-hydrological dynamics were reasonably reconstructed and the forecasting skills were evaluated using an innovative modelling approach, including fully coupled atmospheric-hydrological modelling and improved representation of Sea Surface Temperature.

In order to define a dry period, a threshold for wet days is usually considered to account for measurement errors and evaporation. In the present study, we compare the threshold of 1 mm d⁻¹, the most commonly used threshold, to a time-varying threshold describing evapotranspiration to compare how the risk of extreme dry spells is estimated with both thresholds. Results indicate that considering a fixed threshold can underestimate extreme dry spells during the extended summer.

Several authors have shown evidence of electromagnetic measurements prior to earthquakes. However, these investigations lack a physical mechanism to support them. That is why we developed a theory that could explain many of these phenomena. Specifically, we demonstrate that the generation of microcracks in the lithosphere due to stress changes can explain and describe these electromagnetic phenomena.

The article presents the Alsatian regional flood risk observatory ORRION, a participative online platform on which information is shared between individuals, stakeholders, engineers, and scientists. This maximizes knowledge capitalization and contributes to building a common knowledge base for flood risk. From this information, long chronicles of floods are derived for the Rhine, and most of its main Alsatian tributaries and their main characteristics are identified and analysed.

Vegetation indices based on satellite images, such as the normalized difference vegetation index (NDVI), have been used for damaged pasture insurance. The occurrence of damage is usually defined by NDVI thresholds mainly based on normal statistics. In this work a pasture area in Spain was delimited by MODIS images. A statistical analysis of NDVI was applied to search for alternative distributions. Results show that generalized extreme value distributions present a better fit than normal ones.

Large-scale risk assessments can be improved by a more direct relation between the type of exposed buildings and their flood impact. Compared to the common land-use-based approach, this model reflects heterogeneous structures and defines building-material-based vulnerability classes. This approach is particularly interesting for areas with large variations of building types, such as developing countries and large scales, and enables vulnerability comparison across different natural disasters.

Flood events are often complex in their origin and dynamics. The choice of computer model to simulate can hence determine which level of complexity can be represented. We here compare different models varying in complexity (hydrology with routing, 1-D routing, 1D/2D hydrodynamics) and assess how model choice influences the accuracy of results. This was achieved by using GLOFRIM, a model coupling framework. Results show that accuracy depends on the model choice and the output variable considered.

River flooding is a major safety concern. Sophisticated models are used to design ways to decrease flood risk, but until recently it was not feasible to calculate how uncertain these model predictions are. Using a new approach, we have now quantified the uncertainty of 12 interventions along the River Waal. Results show significant but not problematically high uncertainty. We demonstrate that the choice between interventions can be different when uncertainty is taken into account.

For 11 western US states we addressed gaps in existing wildfire risk assessments that do not explicitly consider cross-boundary fire transmission among major landowners and do not identify the sources of fire for exposed communities. Stochastic wildfire simulation outputs were post-processed in a geo-spatial framework. Our methods can be implemented across different regions of the world to inform fire management agency decisions on the locations of future fuel management projects.

In this paper we present an EM geophysics method designed for shallow purposes for determining the 3-D geometry of a sinkhole. In this case we are applying the technique for the Chac-Mool sinkhole in the Yucatán Peninsula. Thanks to scuba divers we have a broad path they followed along the subterranean rivers. Our 3-D model can be correlated with those scuba diver paths. We also found hidden rivers that pass very close to the surface (less than 10 m) making these areas susceptible to collapse.

The aim of this study is to investigate the differences in the mappable characteristics of earthquake-triggered and rainfall triggered landslides in terms of their frequency–area relationships, spatial distributions and relation with causal factors, as well as to evaluate whether separate susceptibility maps generated for specific landslide size and triggering mechanism are better than a generic landslide susceptibility assessment including all landslide sizes and triggers.

It is known that fragility functions reflect localities (building design standards and topography) and flow velocity is more important, as damage might occur before flow depth reaches its maximum value. This research demonstrates that it is possible to accurately predict building damage by considering related forces with high accuracy, including resistant force, based on building design standards. This method will be useful for damage assessment in areas that have no experience of tsunamis.

Water levels in bays are affected by open-ocean changes and wind. Tides are more dampened in the bays than storm surges and sea level rise. We compare observed and modeled levels with ocean conditions and combine them with analytical models. We consider the local setup, caused by wind along the bay. Expansion using the ADCIRC tidal database will allow coverage of other bay systems on the United States East Coast. Spatial estimates of water level can inform decisions about bay flooding hazards.

This study shows the possibility to improve the weather forecast at the very short range (0–3 h) using lightning and/or radar reflectivity observations. We consider two challenging events that occurred over Italy, named Serrano and Livorno, characterized by moderate and exceptional rainfall, respectively. The improvement given to the forecast by using the lightning and/or radar reflectivity observations is considerable. The best performance is obtained when using both data.

We quantified the impact of hurricanes Irma and Maria on the PTWC initial tsunami warning capability for the Caribbean region after accounting for hurricane-related seismic station outages. Within the eastern Caribbean region the hurricanes exacerbated outages to an astonishing 82 % of the available 76 stations. This resulted in up to 02:33 and 04:33 min added to the earthquake detection and response times, effectively knocking out PTWC's local tsunami warning capabilities in the region.

Koslanda in Sri Lanka is on people's minds due to frequently occurring landslides as the area is made vulnerable by both climatic and geomorphological settings. Integration of radar and optical remote sensing can be used to quite satisfactorily predict landslides with complementary information at a smaller scale. Inclusion of radar-derived factors in landslide susceptibility analysis of a bivariate and multivariate nature improves the prediction of high- and very-low-susceptibility areas.

Flood frequency curves are usually highly uncertain since they are based on short data sets of measured discharges or weather conditions. To decrease the confidence intervals, an efficient bootstrap method is developed. With this method, the data set of measured discharges of the Rhine river is extended by approximately 600 years. The study shows that historic flood events decrease the confidence interval of the flood frequency curve significantly, specifically in the range of large floods.

Despite California being a region highly prone to seismic activity, most homeowners are not covered against this risk. This study analyses the reasons for homeowners to purchase insurance to cover earthquake losses, with application in California. Results suggest that only two extreme situations would lead all owners to cover their home with insurance: (1) a widespread belief that a devastating earthquake is imminent; (2) a massive decrease in the average premium amount by a factor exceeding 6.