In this work, we present new methods for calculating the mean, standard deviation, median, and modal locations of the boundaries of volcanic hazards. These calculations are based on a new, mathematically rigorous definition of a probabilistic hazard map – a way to map the probabilities of inundation by a given hazard. We apply this analysis to several models of volcanic flows: simple models of viscous flows, complex models of a tabletop granular flow, and a complex model of a volcanic mud flow.

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 twelve 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.

On 17 July 1998, a tsunami caused serious damage on the northern coast of Papua New Guinea about 20 min after an earthquake of magnitude 7. The tsunami has been attributed to a submarine landslide that occurred about 13 min after the earthquake. We investigated whether evidence of the landslide could be identified in the seismic records for tsunami warning purpose, and concluded that the 1998 landslide was too small to be evident in the seismic records.

Dynamical downscaling of an NWP model is closely related to the parameterization of convection. This is an unique study on a very severe cyclonic storm OCKHI over the Arabian Sea in which we explore the impact of downscaling of grid resolution in COSMO regional NWP model and investigate the effect of explicit treatment of convection on the performance of model. Switching off the convection parameterization yield good values of CAPE but lead to poor performance in terms of rainfall simulations.

Mitigating landslide risk requires information on landslide hazard at a suitable scale to inform decisions. We develop simple rules to identify landslide hazard – the probability of being hit by a landslide – then test their performance using six existing landslide inventories from recent earthquakes. We find that the best rules are: minimize your maximum look angle to the skyline and avoid steep (> 10°) channels with many steep (> 39°) areas that are upslope.

The potential of W-band radar reflectivity to improve the quality of analyses and forecasts of heavy precipitation events in the Mediterranean area is investigated. The 1D + 3DVar assimilation method has been adapted to assimilate the W-band reflectivity in the Météo-France kilometer-scale NWP model AROME. The results suggest that the joint assimilation of W-band reflectivity and horizontal wind profiles lead to a slight improvement of moisture analyses and rainfall precipitation forecasts.

Extra-tropical windstorms represent one of the biggest natural hazards for Europe. In this paper we use a method to categorise the tracks (based on the shape of the track) of these storms into three different groups. The characteristics of the storms in each of the three groups is then analysed with regards to size, duration or intensity. By using extreme-value statistics we can further estimate what would be a maximum intensity storm for each of the three groups.

We introduce a new prediction-oriented method for hazard assessment of volcaniclastic debris flows, based on multiple models. We apply our procedure to a case study of the 1955 Atenquique flow, using three widely used depth averaged models. Depending on how it is looked at, the exercise provides useful information in either model selection or data inversion. Connecting inverse problems and model uncertainty represents a fundamental challenge in the future development of multi-model solvers.

We propose in this study a fundamentally new approach for the definition of minimum rainfall required for the initiation of landslides based on satellite-derived antecedent rainfall estimates directly coupled with data on surface susceptibility for landslides. We apply our approach in the western branch of the East African Rift and provide first regional rainfall thresholds for landsliding in tropical Africa.

Cellular Automata are useful tools to simulate wildfire propagation. We design a Cellular Automata to simulate a severe wildfire that took place in Portugal in 2012 and resulted in almost 25 thousand hectares burned. The explosive stage is adequately modeled when refining the role played by the wind in fire spreading. Results show a probability of ignition out of the limits of the observed scar, an information that the model may help choosing locations to allocate resources for fire combat.

Before the opening of the railway, the deformation of Qinghai-Tibet Railway was very small and considered stable. After opening, the overall stability of the railway section was good. The main deformation areas are concentrated in the areas where railway lines turn and geological disasters are concentrated. In order to ensure the safety of railway operation, it is necessary to carry out long-term time series observation along the Qinghai-Tibet Railway.

We mapped – for the first time – the probability of occurrence of drought over Spain, with the overriding aim of improving current drought assessment, management and mitigation measures and strategies across the region. Spatially, our estimations suggest a higher probability of extreme drought events in southern and central areas of Spain, compared to northern and eastern regions. Nevertheless, there are strong differences between drought indices, as well as among drought timescales.

In Kenli district in the Yellow River Delta (YRD) region of China, using Landsat data, the optimal inversion model of soil salt content (SSC) for different seasons was determined, the SSC exhibited a gradually increasing trend from the south-west to the north-east, and its seasonal dynamics were as that soil salinity accumulates in spring, decreases in summer, rises in autumn, and peaks in winter. The results can provide data support for the treatment and utilization of saline alkali soil.

Due to the soil holding capacity is very limited in karst area, it is necessary to consider the allowable loss of soil. Here we took thermodynamic dissolution model of carbonate rocks and the lithological characteristics to estimate soil loss tolerance, then corrected and quantitatively evaluated the soil erosion. We identified the spatial pattern of soil loss tolerance in karst area, found RUSLE overestimated soil erosion in karst area and proposed an idea for evaluating soil erosion in karst.

We developed statistical models simulating the probability of large wildfires in France from the climate forcing. The models were able to reproduce both spatial and temporal variability in large wildfires across different environmental regions. The models have wide applications, including improving our understanding of the drivers of large wildfires over the historical period and providing a basis for simulations of future large wildfires from climate projections.

This research aimed to assess the tsunami flow velocity and height reduction produced by a planned elevated road spanned parallel to the coast of Banda Aceh called Banda Aceh Outer Ring Road (BORR). Cornell Multi-Grid Coupled Tsunami Model (COMCOT) was used to simulate eight scenarios of the tsunami. One of them was based on the 2004 Indian Ocean tsunami. Two magnitudes of earthquake were used, that is, 8.5 and 9.15 Mw. The elevated road is potential to mitigate impacts of future tsunamis.

Flood defences have the potential to be integrated with many other functions. While assessments of multifunctional dikes are usually conservative, using a probabilistic framework allows for synergies to be integrated and the risks to be made explicit. This change leads to a better perspective on the protection level of a flood defence and the effectiveness of reinforcements, allowing for better implementation of multifunctional elements on flood defences.

A methodology to optimize the design of an offshore tsunami network array is presented, allowing placement of sensors to be used in a Early Tsunami Warning System framework. The method includes multiple tsunami parameters as a measure of the predictive accuracy through a single cost function. The resulting network performance was tested against an historical event, suggesting that having such a network in place could have provided meaningful information for the hazard assessment.

The research has been done to assess the performance of relative positioning over long baselines in determining the accuracy of site velocities from repeated short session GPS measurements. GPS repeated measurements were generated from the IGS data, and the results were compared with PPP derived findings. Major outcome of this study is that relative positioning over long baselines produces similar accuracies to PPP. A newly proposed refinement method also improves the available PPP accuracy.

Rainfall Intensity–Duration–Frequency (IDF) curves are used as the basis for the design, management, and operation of water resources infrastructure at local and regional scales. Given anticipated intensification of the hydrological cycle with anthropogenic global warming, quantitative information on the future extreme rainfall hazard is needed by practitioners. This paper analyses projected changes in IDF curves based on high-resolution regional climate model simulations over North America.

Assessing the risk of complex systems to natural hazards is an important and challenging problem. In today's socio-technological world, the connections and interdependencies between exposed elements are crucial. These complex relations call for a paradigm shift in collective risk assessment. This two-part paper proposes a new holistic approach to assess the risk of complex systems based on Graph Theory. Part II presents an application to a pilot study in Mexico City.

The study integrates tsunami hazards that create risk into: hazard-product risk and hazard-affected risk. Consistent with different factors evaluated for these two risk dimensions, different spatial distributions between them can be formed. The study recommends tsunami shelter accessibility to be enhanced where hazard-product risk is large and where hazard-affected risk is high in a different and targeted way. The tsunami risk situation’s complexity in coastal port cities makes this interesting.

Assessing the risk of complex systems to natural hazards is an important and challenging problem. In today’s socio-technological world, the connections and interdependencies between exposed elements are crucial. These complex relations call for a paradigm shift in collective risk assessment. This two-part paper proposes a new holistic approach to assess the risk of complex systems based on Graph Theory. Part I presents the approach and discusses how it can be used to more thoroughly assess risk.

Results indicate that 2071–2099 will face the most severe droughts, with the highest frequency and longest duration. In terms of spatial distribution, drought hazards in north China will be greater than in south China, especially the non-monsoon region. Comparing the two scenarios, drought hazards from RCP8.5 are higher than that from RCP4.5, i.e., higher frequency, longer duration of drought and more significant drying trends. These results provide a reference for adaption and reduction.

Flood maps can help stakeholders and the public understand their flood risk. We evaluated the quality of publicly-accessible flood maps in Canadian communities designated as flood risk areas. We found that most (62 %) are low-quality (meeting less than half of the criteria) and the highest score was 78 % (7 of 9 criteria met). Canada must make a more concerted effort to produce high-quality flood maps to support its international commitment to disaster risk reduction.

Field surveys after major disastrous earthquakes have shown that poor performance of buildings in earthquake affected areas is the leading cause of human fatalities and economic losses. The evaluation of seismic fragility for existing building stocks has become a crucial issue due to the frequent occurrence of earthquakes in the last decades. They are required for the estimation of fatalities and monetary losses due to structural damage in destructive natural disasters like earthquakes.

A methodology is proposed for the quantitative risk assessment for roadways subjected to rockfalls, retaining wall failures, slow moving landslides, and coastal erosion induced failures. It includes the calculation of the probability of occurrence of each hazard with a given level, based on an extensive collection of field data, and its association with the consequences. The latter were assessed considering the road damage repair cost for each level, in terms of a fixed Unit Cost.

We study rockfall hazard to a town below a cliff in an earthquake prone area. Large trailer-truck size boulders are scattered down slope above town. Mapping boulder locations and sizes (in the field and in past aerial photos) and calculating their predicted trajectories down slope using computer simulation yielded a hazard map for rockfall impact. hazard is reduced where slope angle is below 10°. Dating rockfalls coincides with past earthquakes and predicts probability for future rockfalls.

In recent years, several valuable nature reserves in Belgium have been severely damaged by wildfires. In order to optimize the risk management and get prepared for a possibly increasing number of such events, a first wildfire risk map is developed for Belgium, based on data that was obtained from the government and newspaper articles.

Long-term geo-referenced time series showed spatio-temporal variations of large fires (LF ≥ 100 ha) throughout the French Mediterranean with 21 % of the total LF burned area occurring on a surface previously burned. The region was impacted up to 5–6 times by recurrent LF, the East experiencing fewer but larger LF despite fire weather conditions decreasing eastwards. The efficiency of fire management emerged in early 1990s but LF outbreak during extreme weather conditions remains of major concern.

It is necessary to analyze the weight of multi factors on the risk consequence of dam break. When the number of influencing factors exceeds 10 or more, the analysis of its weight will become very difficult. In this manuscript, the cloud model, an artificial intelligence calculation method, is used to transform the subjective factors into a large number of data for the improved entropy weight method. The result is objective and reasonable, providing a new way of analyzing multi factor weights.

Climate adaptation of deltas is a hot topic given their high population density. We found that measures within the embanked floodplains of the Rhine delta cannot compensate for plausible increases in river discharge and sea level rise, except large-scale floodplain lowering. Measures located in state-owned areas are relatively effective, which would simplify decision making. Natural management of the land greatly improves biodiversity, while flood safety effects were limited.