GeoHazards

In the mapping and assessment of mountain hazard susceptibility using machine learning models, the selection of model parameters plays a critical role in the accuracy of predicting models. In this study, we present a novel approach for developing a prediction model based on random forest (RF) by incorporating ensembles of hyperparameter optimization. The performance of the RF model is enhanced by employing a Bayesian optimization (Bayes) method and a genetic algorithm (GA) and verified in the Wudu section of the Bailong River basin, China, which is a typical hazard-prone, mountainous area. We identified fourteen influential factors based on field measurements to describe the “avalanche–landslide–debris flow” hazard chains in the study area. We constructed training (80%) and validation (20%) datasets for 378 hazard sites. The performance of the models was assessed using standard statistical metrics, including recall, confusion matrix, accuracy, F1, precision, and area under the operating characteristic curve (AUC), based on a multicollinearity analysis and Relief-F two-step evaluation. The results indicate that all three models, i.e., RF, GA-RF, and Bayes-RF, achieved good performance (AUC: 0.89~0.92). The Bayes-RF model outperformed the other two models (AUC = 0.92). Therefore, this model is highly accurate and robust for mountain hazard susceptibility assessment and is useful for the study area as well as other regions. Additionally, stakeholders can use the susceptibility map produced to guide mountain hazard prevention and control measures in the region. Full article

The computational simulation of rivers is a useful tool that can be applied in a wide range of situations from providing real time alerts to the design of future mitigation plans. However, for all the applications, there are two important requirements when modeling river behavior: accuracy and reasonable computational times. This target has led to recent developments in numerical models based on the full two-dimensional (2D) shallow water equations (SWE). This work presents a GPU accelerated 2D SW model for the simulation of flood events in real time. It is based on a well-balanced explicit first-order finite volume scheme able to run over dry beds without the numerical instabilities that are likely to occur when used in complex topography. The model is applied to reproduce a real event in the reach of the Ebro River (Spain) with a downstream reservoir, in which a study of the most appropriate boundary condition (BC) for modeling of the dam is assessed (time-dependent level condition and weir condition). The whole creation of the model is detailed in terms of mesh optimization and validation. The simulation results are compared with field data over the flood duration (up to 20 days), allowing an analysis of the performance and time saved by different GPU devices and with the different BCs. The high values of fit between observed and simulated results, as well as the computational times achieved, are encouraging to propose the use of the model as a forecasting system. Full article

Offshore basalts, most commonly found as oceanic crust formed at mid-ocean ridges, are estimated to offer an almost unlimited reservoir for CO₂ sequestration and are regarded as one of the most durable locations for carbon sequestration since injected CO₂ will mineralize, forming carbonate rock. As part of the Solid Carbon project, the potential of the Cascadia Basin, about 200 km off the west coast of Vancouver Island, Canada, is investigated as a site for geological CO₂ sequestration. In anticipation of a demonstration proposed to take place, it is essential to assess the tendency of geologic faults in the area to slip in the presence of CO₂ injection, potentially causing seismic events. To understand the viability of the reservoir, a quantitative risk assessment of the proposed site area was conducted. This involved a detailed characterization of the proposed injection site to understand baseline stress and pressure conditions and identify individual faults or fault zones with the potential to slip and thereby generate seismicity. The results indicate that fault slip potential is minimal (less than 1%) for a constant injection of up to ~2.5 MT/yr. This is in part due to the thickness of the basalt aquifer and its permeability. The results provide a reference for assessing the potential earthquake risk from CO₂ injection in similar ocean basalt basins. Full article

The final goal of this paper is to contribute to the difficult task of understanding and forecasting submarine volcanic eruption activity by proposing a method to quantify discolored water. To achieve this purpose, we quantitatively analyzed the discolored seawater seen before and after the eruption of the marine environment around the Indonesian submarine volcano “Anak Krakatau”, which erupted at the end of December 2018, from the viewpoint of the “dominant wavelength”. The atmospherically corrected COM-C SGLI data for 17 periods from the eruption from October 2018 to March 2019 were used. As a result, the following three main items were found. First, the average ± standard deviation of the entire dominant wavelength was 497 nm ± 2 nm before the eruption and 515 nm ± 35 nm after the eruption. Second, the discolored water area around the island derived from SGLI was detected from the contour line with dominant wavelengths of 500 nm and 560 nm. Third, the size of a dominant wavelength of 500 nm or more in the discolored water areas changed in a complicated manner within the range of almost 0 to 35 km². The area of the dominant wavelength of 500 nm or more slightly increased just before the eruption. Finally, it was proven that the “dominant wavelength” from the SGLI proposed in this paper can be a very effective tool in understanding or predicting submarine volcanic activity. Full article

Coastal environments are highly recognized for their spectacular morphological features and economic activities, such as agriculture, maritime traffic, fishing, and tourism. In the context of climate change and the evolution of physical processes, the occurrence of intense natural phenomena adjacent to populated coastal areas may result in natural hazards, causing human and/or structural losses. As an outcome, scientific interest in researching and assessing multi-hazard susceptibility techniques has increased rapidly in an effort to better understand spatial patterns that are threatening coastal exposed elements, with or without temporal coincidence. The islands of Milos and Thira (Santorini Island) in Greece are prone to natural hazards due to their unique volcano-tectonic setting, the high number of tourist visits annually, and the unplanned expansion of urban fabric within the boundaries of the low-lying coastal zone. The main goal of this research is to analyze the onshore coastal terrain’s susceptibility to natural hazards, identifying regions that are vulnerable to soil erosion, torrential flooding, landslides and tsunamis. Therefore, the objective of this work is the development of a multi-hazard approach to the South Aegean Volcanic Arc (SAVA) islands, integrating them into a superimposed susceptibility map utilizing Multi-Criteria Decision-Making (MCDM) analysis. The illustrated geospatial workflow introduces a promising multi-hazard tool that can be implemented in low-lying coastal regions globally, regardless of their morphometric and manmade characteristics. Consequently, findings indicated that more than 30% of built-up areas, 20% of the transportation network, and 50% of seaports are within the high and very high susceptible zones, in terms of the Extended Low Elevation Coastal Zone (ELECZ). Coastal managers and decision-makers must develop a strategic plan in order to minimize potential economic and natural losses, private property damage, and tourism infrastructure degradation from potential inundation and erosion occurrences, which are likely to increase in the foreseeable future. Full article

Geyser geothermal fields are scenic volcanic landforms that often contain tens to hundreds of thermal spot vents that erupt boiling water or contain bubbling mud pools. The fields are potentially hazardous sites due to boiling water temperatures and changes in vent locations and eruption dynamics, which are poorly understood. Here we report on the rapid and profound changes that can affect such a geyser field and ultimately lead to a dangerous, unanticipated eruption. We studied the Geyser Valley, Kamchatka Peninsula, which is a field of geysers and other thermal features and boiling pools. Using high-resolution tri-stereo satellite data and unmanned aerial systems (UAS) with optical and thermal infrared cameras in 2018 and 2019, we were able to identify a newly emerging explosion site. Structure-from-motion analysis of data acquired before and after the explosion reveals morphological and thermal details of the new vent. The explosion site produced an aureole zone of more than 150 m³ of explosively redeposited gravel and clay, a slightly elliptical crater with a diameter of 7.5 m and a crater rim 0.30 m high. However, comparison with archives of photogrammetric data suggests that this site was thermally active years earlier and contained a crater that was obscured and covered by landslides and river sediments. The results allow us to develop a conceptual model and highlight the hazard potential of thermal features buried by landslides and clastic deposits. Sudden explosions may occur at similar sites elsewhere, highlighting the need for careful assessment and monitoring of geomorphological and hydrological changes at geyser sites in other regions. Full article

We present a simple nonlinear model of a shear-beam building that experiences large nonlinear deformations and collapse when excited by large pulses of strong earthquake ground motion. In this paper, we introduce the model and show that its properties can be selected to be consistent with the damage observed in a seven-story hotel in San Fernando Valley of the Los Angeles metropolitan area during the 1994 Northridge earthquake. We also show an example of excitation that leads to the collapse of the model. We illustrate the response only for a sequence of horizontal pulses. We will describe the response of the same model to horizontal, vertical, and rocking motions at its base, as well as for more general excitation by strong earthquake ground motion, in future papers. Full article

The Himalayan state of Sikkim is prone to some of the world’s largest landslides, which have caused catastrophic damage to lives, properties, and infrastructures in the region. The settlements along the steep valley sides are particularly subject to frequent rainfall-triggered landslide events during the monsoon season. The region has also experienced smaller rock slope failures (RSF) after the 2011 Sikkim earthquake. The surface displacement field is a critical observable for determining landslide depth and constraining failure mechanisms to develop effective mitigation techniques that minimise landslide damage. In the present study, the persistent scatterers InSAR (PSI) method is employed to process the series of Sentinel 1-A/B synthetic aperture radar (SAR) images acquired between 2015 and 2021 along ascending and descending orbits for the selected areas in Gangtok, Sikkim, to detect potentially active, landslide-prone areas. InSAR-derived ground surface displacements and their spatio-temporal evolutions are combined with field investigations to better understand the state of activity and landslide risk assessment. Field investigations confirm the ongoing ground surface displacements revealed by the InSAR results. Some urban areas have been completely abandoned due to the structural damage to residential housing, schools, and office buildings caused by displacement. This paper relates the geotechnical investigations carried out on the ground to the data obtained through interferometric synthetic aperture radar (InSAR), focusing on the triggering mechanisms. A strong correlation between seasonal rainfall and landslide acceleration, as well as predisposing geological-structural setting, suggest a causative mechanism of the landslides. Full article

Greece, like the rest of the Mediterranean countries, faces wildland fires every year. Besides their short-term socioeconomic impacts, ecological destruction, and loss of human lives, forest fires also increase the burnt areas’ risk of flash flood phenomena, as the vegetation, which acted in a protective way against runoff and soil erosion, is massively removed. Among the most severe wildland fire events in Greece were those of summer 2021, which were synchronous to the very severe heat waves that hit the broader area of the Balkan Peninsula. More than 3600 km² of land was burnt and a significant amount of natural vegetation removed. Three of the burnt areas are examined in this work, namely, Attica, Northern Euboea, and the Peloponnese, in order to assess their risk of future flash flood events. The burnt areas were mapped, and their geological and geomorphological features studied. Flash flood hazard assessment was accomplished through a Boolean logic-based model applied through Geographic Information Systems (GIS) software, which allowed the prioritization of the requirement for protection by identifying which locations were most prone to flooding. The largest part of our study areas is characterized by geomorphological and geological conditions that facilitate flash flood events. According to our findings, in almost all study areas, the regions downstream of the burnt areas present high to very high flash flood hazard, due to their geomorphological and geological features (slope, drainage density, and hydrolithology). The only areas that were found to be less prone to flood events were Vilia and Varimpompi (Attica), due to their gentler slope inclinations and overall geomorphological characteristics. It is known that vegetation cover acts protectively against flash floods. However, in this case, large areas were severely burnt and vegetation is absent, resulting in the appearance of flash floods. Moreover, imminent flooding events are expected to be even more intense in the areas downstream of the burnt regions, possibly bearing even worse impacts on the local population, infrastructure, etc. Full article

The eruption of Cumbre Vieja (also known as Tajogaite volcano, 19 September–13 December 2021, Spain) is an example of successful emergency management. The lessons learnt are yet to be fully disclosed as is whether the response can be further improved. The latter may include tools to predict lava flow inundation rheological characteristics, amongst other issues related to volcanic eruptions (i.e., ash fall and gas emission). The aim of this study was to explore if a scientific open-source, readily available, lava-flow-modelling code (VolcFlow) would suffice for lava emplacement forecasting, focusing on the first seven days of the eruption. We only the open data that were released during the crisis and previously available data sets. The rheology of the lava, as well as the emission rate, are of utmost relevance when modelling lava flow, and these data were not readily available. Satellite lava extent analysis allowed us to preliminarily estimate its velocity, the average flow emitted, and flow viscosity. These estimates were numerically adjusted by maximising the Jaccard morphometric index and comparing the area flooded by the lava for a simulated seven-day advance with the real advance of the lava in the same timescale. The manual search for the solution to this optimization problem achieved morphometric matches of 85% and 60%. We obtained an estimated discharge rate of about 140 m³/s of lava flow during the first 24 h of the eruption. We found the emission rate then asymptotically decreased to 60 m³/s. Viscosity varied from 8 × 10⁶ Pa s, or a yield strength of 42 × 10³ Pa, in the first hours, to 4 × 10⁷ Pa s and 35 × 10³ Pa, respectively, during the remainder of the seven days. The simulations of the lava emplacement up to 27 September showed an acceptable distribution of lava thickness compared with the observations and an excellent geometrical fit. The calculations of the calibrated model required less time than the simulated time span; hence, flow modelling can be used for emergency management. However, both speed and accuracy can be improved with some extra developments and guidance on the data to be collected. Moreover, the available time for management, once the model is ready, quasi-linearly increases as the forecasting time is extended. This suggests that a predictive response during an emergency with similar characteristics is achievable, provided that an adequate rheological description of the lava is available. Full article

This paper presents the results of landslide prediction modelling for young glacial areas performed using statistical methods. The area in question is urbanized and therefore mass wasting activity is a matter of interest to both the local community and the authorities. The analysis was based on the 2011 ‘Register of landslides and areas prone to mass movements with a scale of 1:10,000 for the city of Gdansk’ and the 2012 incomplete ‘Register of landslides and areas prone to mass movements with a scale of 1:10,000 for the city of Gdynia’. The research took into account geological, geomorphological, hydrological, hydrogeological, and anthropogenic conditions. The landslide susceptibility map was created using the statistical landslide index. The calculated indices were used to create a map of Gdansk’s landslide susceptibility. In Gdansk, 84.50% of the total diagnosed landslide area belongs to the high susceptibility class, 14.25% to the moderate susceptibility class, and only 1.25% to the low or very low susceptibility class. After extrapolation, the data was also used to create a susceptibility map for the remaining parts of the Tri-City area, Sopot and Gdynia. The difficulty of extrapolating landslide data for neighboring urban areas was indicated. In Gdansk, which had been covered by geological mapping, the best modelling results were obtained with a large number of causal factors. In Gdynia and Sopot, for which the statistical landslide index value was extrapolated from Gdansk, the best results were obtained when selected causal factors were considered. In Sopot and Gdynia, 81.6% of the landslide area belongs to the high susceptibility class, 15.1% to the moderate class, and 3.3% to the low susceptibility class. These results emphasize a different role of some causal factor classes in the occurrence of landslides in neighboring urban areas. The resultant maps show the areas in which mass wasting is the most probable in the future. Full article

The role of the Manning roughness coefficient in modifying a tsunami time series of flow depth inundation was studied in Iquique, Chile, using a single synthetic earthquake scenario. A high-resolution digital surface model was used as a reference configuration, and several bare land models using constant roughness were tested with different grid resolutions. As previously reported, increasing the Manning n value beyond the standard values is essential to reproduce mean statistics such as the inundated area extent and maximum flow depth. The arrival time showed to be less sensitive to changes in the Manning n value, at least in terms of the magnitude of the error. However, increasing the Manning n value too much leads to a critical change in the characteristics of the flow, which departs from its bore-like structure to a more gradual and persistent inundation. It was found that it is possible to find a Manning n value that resembles most features of the reference flow using less resolution in the numerical grids. This allows us to speed up inundation tsunami modeling, which could be useful when multiple inundation simulations are required. Full article

Climate change is expected to raise river discharge and sea level in the future, and these near-term changes could alter the river flow regime and sedimentation pattern of future floods. Present hazard assessment studies have limitations in considering such morpho-dynamic responses in evaluating flood hazards or risks. Here, we present a multi-model-based approach to quantify such potential hazard parameters influenced by climate change for the most vulnerable communities living on river bars and islands of the Brahmaputra–Jamuna River. River flood-flow and flood wave propagation characteristics are predicted to be affected by changing temporal distribution patterns of precipitation as a result of enhanced global warming. Increased incidences of large multi-peak floods or uncommon floods resulting in long-duration floods driven by sea-level rise may happen as a result of this. To assess it, we have set up a hydromorphic model, Delft3D, for the Brahmaputra–Jamuna River forced by upstream flow, generated from a hydrological model SWAT, over the Brahmaputra basin. The simulations cover moderate, wettest, and driest conditions of the RCP8.5 scenario, and the results reflect the flooding consequences of the near-future, mid-century, and end-century. Floods in the Brahmaputra–Jamuna River are becoming more severe, frequent, and long-lasting, as a result of climate change, and are expected to last until the end of November rather than the current September timeline. While assessing the hazard, we found that the pattern and timing of the flood are as equally important as the peak of the flood, as the river continuously adjusts its cross-sectional area with the flow. The study also demonstrates that, depending on their location/position, climate-induced hazards can affect sand bars/islands disproportionally. The high flood depth, duration, and sedimentation have a significant impact on the sand bars downstream of the river, making them more vulnerable. Full article

As an earthquake is capable of causing significant losses, a strain buildup and release model following an earthquake is of importance for mitigation purposes. In this study, we aim to model strain buildup and release on a strike-slip fault which consists of elastic–brittle (upper crust) and elastic–viscous (lower crust and upper mantle) layers using a finite element model. The fault strength during strain buildup is controlled by the friction coefficient and cohesion, in addition to the viscoelastic parameter, as shown in the deformation model using Maxwell’s material. In the strain buildup model, we found that the differential stress on the elastic layer is larger than that on the viscoelastic layer and that the differential stress increases with the thickness of the elastic layer. When the viscoelastic layer is thinner, the deformation observed on the surface is larger. However, the differential of stress in the strain release model on the elastic layer is smaller than that on the viscoelastic layer, which shows the transfer stress from the lower crust and upper mantle to the upper crust. Using the knowledge gained by varying the thickness and frictional strength of the lithosphere, we discuss the seismicity pattern observed along the Great Sumatran Fault. Full article

Although accurate root cohesion model estimates are essential to quantify the effect of vegetation roots on shallow slope stability, few means exist to independently validate such model outputs. One validation approach for cohesion estimates is back-calculation of apparent root cohesion at a landslide site with well-documented failure conditions. The catchment named CB1, near Coos Bay, Oregon, USA, which experienced a shallow landslide in 1996, is a prime locality for cohesion model validation, as an abundance of data and observations from the site generated broad insights related to hillslope hydrology and slope stability. However, previously published root cohesion values at CB1 used the Wu and Waldron model (WWM), which assumes simultaneous root failure and therefore likely overestimates root cohesion. Reassessing published cohesion estimates from this site is warranted, as more recently developed models include the fiber bundle model (FBM), which simulates progressive failure with load redistribution, and the root bundle model-Weibull (RBMw), which accounts for differential strain loading. We applied the WWM, FBM, and RBMw at CB1 using post-failure root data from five vegetation species. At CB1, the FBM and RBMw predict values that are less than 30% of the WWM-estimated values. All three models show that root cohesion has substantial spatial heterogeneity. Most parts of the landslide scarp have little root cohesion, with areas of high cohesion concentrated near plant roots. These findings underscore the importance of using physically realistic models and considering lateral and vertical spatial heterogeneity of root cohesion in shallow landslide initiation and provide a necessary step towards independently assessing root cohesion model validity. Full article

In 2005, Hurricanes Katrina and Rita caused widespread destruction and displacement in parts of Louisiana, Alabama, and Mississippi. This research evaluates determinants of displacement and, conditional on being displaced, the duration of displacement for households living in areas affected by these hurricanes. Hurdle Models, which assume that different processes govern zero outcomes (i.e., no displacement) and positive outcomes (i.e., amount of time displaced), are used to model the likelihood of household displacement and its duration as a function of socioeconomic characteristics, hurricane-caused property and neighborhood damage, social support, and financial assistance. Results show that mobile home residence, marital status, educational attainment, the presence of children, and property and neighborhood damage affect the likelihood and expected length of displacement among sample respondents. Financial assistance and social support are also correlated with displacement and its duration, but endogeneity concerns complicate the interpretation of these results. The findings highlight the diversity of factors that slow households’ return following displacement and underscore the need for additional research on the role of social capital in determining hazard-related outcomes. Full article

There are still gaps in defining values and category classifications of exposed items in quantitative damage analysis. This paper proposes a framework that refines the development of flood risk analysis at a local scale. This study first performs a quantitative risk analysis, based mainly on secondary data; it then attempts to communicate the results graphically, aiming to reduce the financial and human resources required. We propose an easily standardized database in a GIS environment, analyzing the influence of a reservoir for flood control and the construction of replicable local-scale risk curves. Hydrological (HEC-HMS) and 2D hydrodynamic (HEC-RAS) models were used to simulate hydrographs considering different return periods. For damage estimation, the processing included vectorization of lots, building use definition with Google Street View, classification of standard designs, and a field survey to validate those classes. In monetary value, this study calculated the effect of the construction of a reservoir for damage reduction, showing the potential to determine the effectiveness of measures adopted to mitigate flood impacts. In addition, for each simulated return period, exposure, hazard, and damage maps can be established, making it possible to perform a complete risk analysis. Full article

Real-time local tsunami warnings embody uncertainty from unknowns in the source definition within the first minutes after the tsunami generates. In general, Tsunami Warning Systems (TWS) provide a quick estimate for tsunami action from deterministic simulations of a single event. In this study, variability in tsunami source parameters has been included by running 135 tsunami simulations; besides this, four different computational domains in the northeastern Atlantic ocean have been considered, resulting in 540 simulations associated with a single event. This was done for tsunamis generated by earthquakes in the Gulf of Cadiz with impact in the western Iberian peninsula and the Canary Islands. A first answer is provided after one minute, and 7 min are required to perform all the simulations in the four computational domains. The fast computation allows alert levels all along the coast to be incorporated into the Spanish National Tsunami Early Warning System. The main findings are that the use of a set of scenarios that account for the uncertainty in source parameters can produce higher tsunami warnings in certain coastal areas than those obtained from a single deterministic reference scenario. Therefore, this work shows that considering uncertainties in tsunami source parameters helps to avoid possible tsunami warning level underestimations. Furthermore, this study demonstrates that this is possible to do in real time in an actual TWS with the use of high-performance computing resources. Full article