Search

High-viscosity magma can form laccolith intrusions that deform and fracture the overburden, causing surface uplift and ground fracturing. Laccolith-induced deformation features have been described at well-exposed outcrops of long-solidified intrusions. The lack of recent geophysical data on rare laccolith emplacement events and the use of linearly elastic continuum-based numerical models precludes a clear understanding of the dynamic fracturing mechanisms. We present a new two-dimensional (2D) Discrete Element Method (DEM) approach to dynamic magma intrusion in a particle-based host medium. The model indicates highly discontinuous deformation and dynamic fracturing and visualizes the localization of subsurface strain. We calibrate the numerical rock strength parameters by performing numerical laboratory experiments to natural rock strength values. We systematically explored the effect of numerical parameters that govern host rock strength (bond cohesion, bond tensile strength, bond elastic modulus), and intrusion depth, on the spatial distribution of strain, stress, and fracturing. We find that high host rock stiffness results in widely distributed and dense fracturing associated with symmetrical dome-shaped surface uplift. Low host rock stiffness results in the concentration of central fracturing and narrow lateral shear bands and asymmetric evolution of the laccolith geometry and the surface deformation pattern. These patterns are affected by the intrusion depth. Our models help understand fracture distribution patterns above laccolith intrusions and open unprecedented opportunities for dynamically modelling intrusion-induced deformation in the upper few kilometers of the Earth’s crust.

Floors of impact craters on rocky planetary bodies in our Solar System are often fractured and bulged. Such deformation features are thought to form by the ascent of impact-generated magma and the inflation of laccolith-shaped magma bodies at a shallow depth below the crater floor. Only the final surface deformation features can be observed from space, and so modeling is the only manner to understand controls on magma emplacement depth and volume, and deformation of the overlying rock. The existing models of crater floor fracturing mostly assume linearly elastic deformation of the shallow planetary crust and are not capable of simulating dynamic opening and propagation of fractures. In contrast, magma-induced deformation on Earth often displays non-elastic deformation features. This mismatch between the realistic mechanical response of planetary crust to magma intrusion and the one assumed by numerical models leads to significant inaccuracies in the modeled magma intrusion characteristics. This has important consequences for volcanic unrest monitoring on Earth and our understanding of structural deformation generated by volcanism throughout the Solar System.We propose a new two-dimensional (2D) Discrete Element Method (DEM) approach to model dynamic fracturing and displacement in a particle-based host medium during the simulated inflation of a laccolith intrusion. The model indicates highly discontinuous deformation and dynamic fracturing and visualizes the localization of subsurface strain. We explored the effect of different gravitational conditions on the Moon, Mars and Earth on the spatial distribution of strain, stress, and fracturing above an inflating laccolith. Moreover, by systematically exploring a range of numerical parameters that govern host rock strength (bond cohesion, bond tensile strength, bond elastic modulus), and intrusion depth, we find complex controls of mechanical properties of planetary crust on the magma intrusion characteristics. Our models help understand fracture distribution patterns above laccolith intrusions in the shallow crust of rocky planetary bodies. We demonstrate that considering dynamic deformation and fracturing mechanisms in numerical models of magma-induced deformation is essential to better understand the formation of floor-fractured craters and the magmatic intrusions that lie beneath.

Many volcanoes exhibit evidence of long-term deformation driven by gravity. This deformation influences the development of magmatic plumbing systems and volcanic vent locations, and it manifests as earthquakes and ground surface motions. Two end-member deformation styles are volcano sagging (flexure) and volcano spreading. Previous modeling studies indicated that the mechanical and geometric properties of basement rocks below the volcano can control whether a volcano spreads or sags, but these works differed on exactly how. Furthermore, evidence in nature that a volcano may progress from sagging to spreading as it grows has not been tested experimentally. Using Digital Image Correlation (DIC) and a revised scaling approach, we here revisit the classic experiments in which a sand–plaster cone is emplaced upon a basement comprising an upper brittle layer and a lower ductile layer. In a first experiment series, we emplaced the cone instantaneously. By normalizing both brittle and ductile layer thicknesses to cone height, we provide new insight into the dimensionless geometric controls on spreading and sagging. In a novel second experiment series, we emplaced the cone incrementally. We show that as a growing cone migrates through the dimensionless geometric parameter space, its deformation style changes successively in time from sagging to spreading. DIC shows that the horizontal velocity of the cone flank increases linearly or non-linearly during cone construction, but it decreases exponentially after construction ceases. Despite their geometric simplification and the omission of several sedimentological, magmatic, and regional-tectonic processes, our models' results are compatible with a range of observations of volcanoes on Earth and Mars. In particular, they help to explain: (i) why lithosphere-scale loading results in sagging rather than spreading, and (ii) why ocean island volcanoes of >2-3 km in height develop stellate forms and rift zones.

Surface deformation patterns on terrestrial planetary bodies have been interpreted as magma emplacement features due to their similarity with Earth analogues or their association with inferred eruptive products. Dome-shaped surface uplift features were recognized for example at floor-fractured impact craters on the Moon (Jozwiak et al., 2012), and at the Tharsis volcanic province on Mars (Farrand et al., 2011). The lack of subsurface geophysical data however precludes a clear observation of the characteristics and emplacement mechanisms of the inferred underlying magma intrusions. On Earth, our insight into the characteristics of the magmatic plumbing systems that underly surface doming also remains cryptic due to the sparsity of observations. The most recent event was the hybrid explosive-effusive eruption at Cordón Caulle, Chile, in 2011. Pre- and syn-eruptive surface doming and pervasive ground fracturing observed there using satellite radar interferometry was interpreted there to result from the intrusion of a rhyolitic laccolith (Castro et al., 2016). Analytical and numerical models can use such geodetic observations to help estimate intrusion characteristics and assist in forecasting potential volcanic eruptions. The same models are used to infer shallow magma intrusion characteristics on terrestrial planetary bodies where the only observations available are surface topography and shallow ground-penetrating radar data. Existing models generally assume that planetary crusts deform linearly-elastic, which leads to a simplification of the magmatic source geometry and the related displacement and stress fields. On Earth, however, host rocks around exposed solidified and exposed dykes, sills and laccoliths often show indicators of non-elastic deformation. Strain can accumulate along large-scale discontinuities in the overburden rocks, making the investigation of the emplacement mechanisms by traditional continuum models difficult. To provide a more geologically realistic means of estimating magma intrusion characteristics and investigate the associated emplacement mechanisms, we have developed a numerical discontinuum model of laccolith emplacement within the framework of the “DeMo-Planet” project. We use the two-dimensional (2D) Discrete Element Method (DEM) model PFC2D (Itasca Consulting Group, Inc.) to represent the modelled medium as a particle-based network. We can use this model to indicate fracturing and highly discontinuous deformation, as well as visualizing the localization of subsurface strain and corresponding deformation. Two stages of the model application are now available. In a first stage, a laccolith-shaped pressure source is inflated at the base of the host medium with varied depth of the magmatic source. In a second stage, particles are injected into the laccolith-shaped fluid body to simulate constant magma supply into a growing intrusion. In this presentation, we will simulate laboratory tests to calibrate the mechanical behaviour and the relations between particle contact parameters of the numerical host rock and magma. To that end, we use the elastic property values (Young modulus) and tensile strength measured on host rocks estimated for the Moon and Mars (Heap et al., 2020). We then use elastic properties measured using uni-axial laboratory tests. The investigated rock samples were collected from exposed Permian sediments above Permian trachyandesite intrusions of the Intra-Sudetic Synclinorium in Poland. The calibration of the model with geological data then allows a comparison of the numerical results with field observations of the structural deformation in the intrusions’ overburden. In the future, we plan to analyze the effect of varying gravity, geometric and mechanical model parameters on the ratio of tensile to shear-mode fracturing, as well as investigating the magma-induced surface displacement in detail. DeMo-Planet demonstrates a promising multidisciplinary approach of informing numerical models of complex, discontinuous magma emplacement processes in the heterogeneous and fractured shallow crust of terrestrial planetary bodies. The validation of this new modelling application with detailed structural information from representative Earth analogue sites will allow us to investigate the characteristics of the cryptic magma bodies underlying surface doming on the Moon and Mars. The application of our model to other terrestrial planetary bodies will simultaneously improve the interpretation of volcanic unrest signals on Earth. References Castro, J.M. et al. (2016) “Rapid laccolith intrusion driven by explosive volcanic eruption,” Nature Communications, 7, p. 13585. doi:10.1038/ncomms13585. Farrand, W.H. et al. (2011) “Spectral evidence of volcanic cryptodomes on the northern plains of Mars,” Icarus, 211(1), pp. 139–156. doi:10.1016/j.icarus.2010.09.006. Heap, M.J. et al. (2020) “Towards more realistic values of elastic moduli for volcano modelling,” Journal of Volcanology and Geothermal Research, 390, p. 106684. doi:10.1016/j.jvolgeores.2019.106684. Jozwiak, L.M. et al. (2012) “Lunar floor-fractured craters: Classification, distribution, origin and implications for magmatism and shallow crustal structure,” Journal of Geophysical Research E: Planets, 117(11), pp. 1–23. doi:10.1029/2012JE004134.

For decades scaled analog experiments have improved the understanding of a broad range of multiphase volcanological processes in controlled laboratory environments. Successfully modeled processes include magma flow through magma reservoirs, conduits and sheets, associated crustal deformation, lava flow, volcanic plume dynamics, ash cloud dispersion, pyroclast sedimentation, pyroclastic density currents and debris flows. Prior to the advent of computational modeling in volcanology, analog experiments were the primary method used to test newly developed concepts. Over the past two decades, technological advances have led to increased quantification of model observables, including deformation fields, lava flow rheologies, bubble and particle suspension compositions, runout distances, plume geometries, and rates of ash cloud spreading and sedimentation. For experimental results to yield further insights in volcanic processes and observables directly useful to volcano monitoring efforts, we expect future progress to focus on three major fronts: 1) improved multimethod measurements in experiments; 2) upscaling to near-natural-scale experiments conducted by multidisciplinary teams at internationally shared facilities; and 3) integration with computational models that will guide future geophysical observations and predictions of volcanic activity. This way, analog experiments will bridge gaps between other techniques in volcanology and improve our understanding and forecasting of volcanic activity from the Earth’s mantle to the surface and into the atmosphere.

When magma ascends through the shallow parts of terrestrial planetary crust, it deforms the surrounding host rocks. The deformation patterns observed at the surface offer indirect means to characterize the position, geometry and volume of subsurface magmatic intrusions. To enable real-time eruption forecasting during volcano unrest, most volcano geodetic models assume that magma intrusion induces linearly elastic deformation of homogeneous shallow planetary crust. Other indirect geophysical volcano monitoring data (e.g., seismology, gravimetry) however offer only limited opportunity for validating geodetic model results. Moreover, recent geological observations at exhumed volcano plumbing systems and geophysical observations of recent intrusion events have shown that plastic behaviour can dominate in heavily fractured and heterogeneous volcanic edifices and tectonically active areas. The question remains how large the effect of unaccounted plastic deformation could be on estimated intrusion characteristics.Scaled laboratory experiments can be an innovative tool to assess by how much modelled magma intrusion characteristics – volume, geometry, position – deviate from reality in circumstances where plastic deformation processes are important. We used a tensile rectangular dislocation in a homogeneous, linearly elastic half-space to invert the three components of near-surface displacements extracted from X-ray Computed Tomography imagery of laboratory experiments of analogue dyke injection in cohesive mixtures of quartz sand and gypsum powder. The model results favored by the inversions are then compared to the three-dimensional characteristics of the analogue magma intrusions observed in the X-ray CT imagery. To further investigate the effect of more complex model geometry, we also used a tensile distributed-opening dislocation geometry. Preliminary results show that inversion results can be improved by fixing values of parameters that control the position of the modelled dislocation, but significant discrepancies remain between the modelled and observed intrusion geometry, orientation and volume. This test study helps gaining insight on the limitations of commonly used volcano geodetic modelling and inversion methods, and provides a novel basis for interpreting geological, geodetic and geophysical data related to volcanic deformation. The experimental results pave the way for developing complex forward models of magma-induced deformation in the heterogeneous shallow crust of terrestrial planets.

Magmatic sheet intrusions are commonly segmented, across multiple scales, with preserved segments typically interpreted as representing an early stage of intrusion growth. Intrusion propagation has long been associated with linear elastic host rock deformation, associated with tapered or elliptical tip zones. Many studies have identified intrusive segments with non-tapered (e.g., superelliptical) tip geometries, associated with a variety of non-brittle host rock deformation. This has led to development of several anelastic propagation models, including fluidisation, viscous indentation, brittle faulting, and ductile faulting and flow. These models are commonly inferred to represent the propagation mechanism throughout intrusion growth, in host rocks with constant material properties. However, non-brittle host rock deformation may overprint earlier emplacement mechanisms, hence tip geometries observed in the field may not be indicative of the entire emplacement process.Here we present a quantitative field study of segment tip geometry and associated host rock deformation using a segmented basaltic sill network at Neist Point, Isle of Skye, UK (part of the Little Minch Sill Complex), and static Finite Element (FE) Models of intrusion tip stress distributions, to define a new conceptual model for intrusion growth. The FE models highlight that as tip geometry changes from elliptical to rectangular, tensile and shear stress maxima move increasingly out-of-plane following the positions of maximum tip curvature, as would be expected for non-brittle propagation and matching field observations of host rock deformation. The studied sill segments are hosted in interbedded limestone, sandstone, siltstone, and mudstone units, and are mostly thin (2 m) and their tips transect multiple units. We identified 39 segments in total, 26 of which were geometrically characterised, and a total of 43 tips were measured. Segments with tapered tips were commonly associated with host rock bending (elastic-brittle emplacement), while superelliptical segments show a variety of host rock deformation (e.g., brecciation, faulting). Notably, this deformation is limited to the preserved segment tips, with no such features recorded along the length of the intrusions. Tip geometry and host rock deformation style are not linked to host rock lithology: local conditions of emplacement evolve to facilitate varying deformation mechanisms within a single intrusive network. Changes to magma viscosity (via crystallisation, volatile/heat loss) and host rock properties (heating, brecciation, fluidisation) may inhibit elastic-brittle fracture, and promote segment inflation and non-brittle propagation.We propose a multiphase conceptual model for basaltic segments in an initially brittle host. Segments are emplaced initially via elastic-brittle fracture followed by a transitional phase of segment inflation, tip rounding, and modification to the conditions of emplacement leading to a non-brittle propagation phase. Our model accounts for multiple segment geometries and styles of host rock deformation observed across many intrusive complexes and across an array of host rock lithologies. Intrusive segments preserved in outcrop primarily represent the final conditions of emplacement, rather than their growth.

Granular materials are a useful analogue for the Earth’s crust in laboratory models of deformation. Constraining their mechanical properties is critical for such model’s scaling and interpretation. Much information exists about monomineralic granular materials, such as quartz sand, but the mechanical characteristics of bimineralic mixtures, such as commonly-used quartz sand mixed with gypsum powder (i.e. plaster), are largely unconstrained. We used several mechanical tests (density, tensile, extension, shear) to constrain the failure envelope of various sand-plaster mixtures. We then fitted linear Coulomb and parabolic Griffith failure criteria to obtain cohesions and friction coefficients. Tests of the effects of emplacement technique, compaction and humidity demonstrated that the most reproducible rheology is given by oven-drying, pouring and mechanically compacting sand-plaster mixtures into their experimentation container. As plaster content increases, the tensile strength of dry sand-plaster mixtures increases from near zero (pure quartz sand) to 166±24 Pa (pure plaster). The cohesion increases from near zero to 250±21 Pa. The friction coefficient varies from 0.54±0.08 (sand) to 0.96±0.08 (20 weight% plaster). The mechanical behaviour of the resulting mixtures shifts at 20-35 weight% plaster from brittle Coulomb failure along a linear failure criterion, to more complex brittle-plastic Coulomb-Griffith failure along a non-linear failure criterion. With increasing plaster content, the brittle-plastic transition occurs at decreasing depth within a pile of sand-plaster mixture. We infer that the identified transitions in mechanical behaviour with increasing plaster content relate to (1) increasing porosities, (2) increasing grain size distributions, and (3) a decrease in sand-sand grain contacts and corresponding increase in contacts of anisotropic gypsum-gypsum grains. The presented characterisation enables a more quantitative scaling of the mechanical behaviour of sand-plaster mixtures, including their tensile strength. Sand-plaster mixtures can thereby realistically simulate brittle-plastic properties of the Earth’s crust in scaled laboratory models., info:eu-repo/semantics/published

Inferences about sheet intrusion emplacement mechanisms have been built largely on field observations of intrusion tip zones: magmatic systems that did not grow beyond their observed state. Here we use finite element simulation of elliptical to superelliptical crack tips, representing observed natural sill segments, to show the effect of sill tip shape in controlling local stress concentrations, and the potential propagation pathways. Stress concentration magnitude and distribution is strongly affected by the position and magnitude of maximum tip curvature κmax. Elliptical tips concentrate stress in-plane with the sill, promoting coplanar growth. Superelliptical tips concentrate maximum tensile stress (σmax) and shear stress out-of-plane of the sill, which may promote non-coplanar growth, vertical thickening, or coplanar viscous indentation. We find that σmax = Pe(1+ 2(√[aκmax]), where Pe is magma excess pressure and a is sill half length. At short length-scales, blunted tips locally generate large tensile stresses; at longer length-scales, elliptical-tipped sills become more efficient at concentrating stress than blunt sills., info:eu-repo/semantics/published

Lahars impose significant secondary hazards on highly populated volcanic islands by remobilizing volcanic ash deposits. Karthala, on Grande Comore Island, is a relatively young and poorly eroded basaltic shield volcano with sporadic occurrence of ash-forming phreatic eruptions. In 2005, two mildly explosive episodes emplaced tephra blankets on the summit area. Heavy precipitation subsequently triggered the repetitive occurrence of small-volume secondary lahars up to 2012. These lahars damaged roads and hundreds of houses, affecting thousands of inhabitants at the foot of Karthala volcano, but were poorly documented at the time of their occurrence. Their future hazard remains unclear as well. This study aims at gaining insights into the main characteristics of these lahars, as well as testing and comparing the effectiveness of two numerical tools to simulate the extent of these small-volume lahars. To understand the physical characteristics of the lahars, we first documented the spatial extent and characteristics of the debris deposits at the foot of Karthala volcano and in the ravines that guided the flows. Our observations suggest that the debris were emplaced by small-scale (volumes ≤ 105 m3), rain-triggered and predominantly low sediment concentration lahars. The spatial extent of the deposits served to calibrate and compare numerical lahar simulations from the widely used volume-limited LaharZ model with simulations from Q-LavHA, a probabilistic flow model originally developed for lava flows. Q-LavHA mitigates some limitations of LaharZ, such as its ability to simulate flow bifurcations and the transition from constrained to unconstrained flow but comparison demonstrates that Q-LavHA typically yielded lower simulation accuracies compared to LaharZ simulations. The obtained accuracy values represent a rather poor performance for both models compared to existing studies on larger-volume lahars on stratovolcanoes, and are inferred to result mostly from difficulties in delineating lahar flow paths on the smooth, poorly eroded topography of the volcanic edifice. We therefore also evaluated the potential to increase simulation accuracy by updating a 10 m resolution Digital Elevation Model (DEM) with channel topography measurements. By using such updated DEMs, the correctly delineated area improved for both models. This approach, however, did not prevent simulations to sometimes miss the hazard-prone position of settlements which were actually affected by the hazard in the past. Our study shows the limitations of numerical volcanic flow simulation strategies on young and poorly eroded volcanic edifices, such as active basaltic shields. The results indicate that accurate topographic representations and detailed documentation of spatial extent of the impacted area and lahar deposit thickness are needed to produce accurate lahar simulations, as well as the further adaptation of existing numerical simulation tools to better suit these particular environmental settings.

Volcanoes display diverse morphologies as a result of the complex interplay of several constructive and destructive processes. Scaled analogue models offer a tool to isolate and characterize the different processes. Here, we investigate the interplay between volcano growth and deformation caused by an underlying strike-slip fault through simple analogue models. In particular, we analyze the morphometry of analogue volcanoes resulting from different growth-to-deformation ratios. Deformation elongates the volcano edifice at an angle of 10-45º from the fault trace along the extensional quadrants and generates a summit graben structure oriented perpendicular to the edifice elongation. The overall steepness of the edifice decreases, but steep slopes are preserved or increased on the lower compressional flanks, commonly related to small avalanches. Growth can partially to totally mask these features depending on the growth rate to strike-slip velocity ratio. The summit graben is easily masked even by low growth rates, whereas edifice elongation has the best preservation potential. Scaling of the experiments suggests that at volcanoes with growth rate (km3/yr) to strike-slip velocity (km/yr) ratios ≤ 3.8 km3/km, deformation features should be clearly preserved, whereas at volcanoes with ratios ≥ 15 km3/km, deformation features should be completely masked. The typical growth rates of volcanoes (0.01 to 1 km3/ka) and the typical velocities of strike-slip faults (1 to 20 mm/yr) suggest that in nature, growth rate to strike-slip velocity ratios can range over 3 orders of magnitude, spanning both types of end-members. Using examples of both active and inactive volcanoes located on strike-slip faults with variable elongation intensities and orientations, we highlight that the analogue models account for some of the morphometric variability observed at volcanoes in nature, although the role of vent distribution can be a key factor. Fil: Grosse, Pablo. Fundación Miguel Lillo. Dirección de Geología; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Tucumán; Argentina Fil: Poppe, Sam. Vrije Universiteit Brussel; Bélgica Fil: Delcamp, Audray. Vrije Universiteit Brussel; Bélgica Fil: van Wyk de Vries, Benjamin. Université Clermont Auvergne; Francia Fil: Kervyn, Matthieu. Vrije Universiteit Brussel; Bélgica

Understanding the propagation and geometry of magmatic intrusions and the related surface displacement is critical for hazard and risk assessment at volcanoes. Magma ascends through the Earth’s crust towards its final subsurface position, or eventually towards its eruption site at the surface. Magma intrusions propagate and grow by deforming their crustal host rocks and by displacing the Earth’s surface. Because directly observing this process is impossible, accurate volcanic eruption forecasts depend on indirect geophysical monitoring of seismicity and surface deformation (e.g. Sigmundsson et al. 2018). The lack of direct observations makes the interpretation of geophysical monitoring data a challenge, and results in heavily debated questions such as: What are the physical mechanisms that control magma-induced deformation? How do magma intrusions grow to their final geometries? How can surface deformation patterns be interpreted in terms of the intrusive processes in the subsurface? To offer a new perspective on these questions, this presentation summarizes the observations and conclusions from the recently completed doctoral dissertation of the first author. This work compares observations from exposed volcanic plumbing systems and geophysical records of recent intrusions with novel 4D laboratory experiments. First, magma-host rock interactions were investigated at a small-scale outcrop in the Oslo Rift System, Norway (Figure 1). We infer from structural and geochemical analyses that the propagation of mafic dyklets there was controlled by regional, pre-existing structural weaknesses in the host rock and thermo-chemical interactions between the sedimentary host rock and the magma (Poppe et al., in prep. (A)). Thermal contact metamorphism produced a secondary fluid, comprising a pore-fluid-magma mixture, that likely propagated along and ahead of the magma body. The magma propagated by opening-mode fracturing and bending of the host rock. At locations in e.g. Australia, Argentina, on the contrary, mixed-mode or shear-mode fracturing has been inferred (e.g. Dering et al. 2019; Spacapan et al. 2017). Fig. 1 – Two overlapping dyklet tips intruded into sedimentary host rock on Hovedøya Island, Oslo Rift, Norway. Note the tapering geometry of the magma-filled fractures as well as the shear displacement along pre-existing fractures in the central, competent limestone layer (from Poppe et al., in prep. a). The ruler scales 30 cm. Second, a novel 4D laboratory modeling approach was developed to overcome limitations of past laboratory models regarding (1) mechanically relevant rheology of model rock materials, and (2) imaging subsurface magma-induced deformation. As for (1), a set of detailed material tests showed that dry sand-plaster mixtures display more complex mechanical behavior than hitherto appreciated, and that these mixtures are suitable as analogues for the Earth’s upper crust (Poppe et al., in prep.(B)). As for (2), 3D displacement and strain fields within sand-plaster host material, induced by analogue golden syrup intrusions, were quantified in laboratory models over time by using medical wide beam X-ray Computed Tomography (CT) and Digital Volume Correlation (DVC). The obtained displacement and strain fields show that analogue intrusions propagate by mixed-mode deformation of their sand-plaster host material (Figure 2). Fig. 2 – Cumulative maximum shear strains in the central vertical X-Y cross-section through the model domain, final 3D rendering of the analogue intrusion, and final model surface topography, extracted from the reconstructed imagery of laboratory models of golden syrup intrusion in sand-plaster host material, by using medical X-ray Computed Tomography (adapted from Poppe et al., 2019). The dominance of the fracturing mode is largely controlled by the host rock rheology, among other physical parameters, and leads to a spectrum of intrusion geometries as observed in nature. In low- to medium-strength host material, shear-mode fracturing dominates and thick cryptodomes or cup shapes form. In medium- to high-strength host material, opening-mode fracturing dominates and thin cone sheets, inclined sheets and dykes form (Figure 2; Poppe et al., 2019). Finally, we extracted surface displacements from the reconstructed CT imagery by using dedicated Matlab codes. Unlike previous models assuming linear elastic rheology and tensile fracturing, our models of golden syrup dykes and steeply inclined sheet intrusions induce dome- or bulges-and-through-shaped uplift of the model surface without net subsidence. Comparable surface displacement patterns were observed during recent sheet intrusion events. Our results suggest that mixed-mode magma-induced fracturing may occur in the weak, highly fractured upper portion of some volcanoes and may affect surface displacement patterns. Consequently, we propose that geodetic inversion models for such settings should incorporate more complex non-elastic behavior.

Geophysical observations provide insight into the deformation processes of the Earth’s upper crust induced by magma propagating towards the surface below active volcanoes. It is however impossible to directly observe these processes in the subsurface, and therefor volcanologists rely on a series of assumptions to construct theoretical and numerical models. To investigate assumptions regarding the geometrical development of magma intrusions and the related fracturing mechanisms in their host rocks, we developed a novel approach of imaging scaled laboratory experiments by using medical, broad beam X-ray Computed Tomography (CT) at the UZ Brussel. A temporal series of 3D scans is therein acquired by CT during injection of golden syrup – a magma analogue – into dry mixtures of quartz sand and gypsum powder – a brittle rock analogue. Extraction of the analogue intrusion geometry using image segmentation algorithms at a series of time steps then allows to study the geometrical growth of the analogue intrusions and deformation structures (i.e. fractures, faults) in the host material. New Matlab scripts further allow extraction of the model surface syn-intrusion, which allow direct comparison with surface displacement data collected at active volcanoes. Furthermore, Digital Volume Correlation (DVC) in the DaVis software (LaVision) is used to quantify the analogue magma-induced displacement and strain fields, which are then quantitatively analysed using dedicated Matlab scripts. The major novelty of our approach is that it extends 4D monitoring and quantification of deformation beyond the model surface deformation by enabling a full quantification of the analogue intrusion growth and related host rock deformation. By producing such unprecedented data off the experiments, we found that host rock strength controls intrusion geometry and propagation mode. Comparison with geophysical monitoring data from recent magma intrusion events, challenges the physical relevance of currently used conceptual models that are used to estimate subsurface magma intrusion geometries, and ultimately contributes to strengthening volcano eruption forecasts.

The study of rift basin's morphology can provide good insights into geological features influencing the development of rift valleys and the distribution of volcanism. The Kivu rift segment represents the central section of the western branch of the East African Rift and displays morphological characteristics contrasting with other rift segments. Differences and contradictions between several structural maps of the Kivu rift make it difficult to interpret the local geodynamic setting. In the present work, we use topographic and bathymetric data to map active fault networks and study the geomorphology of the Kivu basin. This relief-based fault lineament mapping appears as a good complement for field mapping or mapping using seismic reflection profiles. Results suggest that rifting reactivated NE-SW oriented structures probably related to the Precambrian basement, creating transfer zones and influencing the location and distribution of volcanism. Both volcanic provinces, north and south of the Kivu basin, extend into Lake Kivu and are connected to each other with a series of eruptive vents along the western rift escarpment. The complex morphology of this rift basin, characterized by a double synthetic half-graben structure, might result from the combined action of normal faulting, magmatic underplating, volcanism and erosion processes.

Natural disasters are too often presented as resulting from extreme natural phenomena affecting helpless populations, with people being insufficiently aware of the factors leading to disasters and of the existing strategies to mitigate their impacts. We developed a board game aimed at raising awareness about geohazards and disaster risk reduction strategies. The target groups are (1) secondary school students and citizens and (2) scientists and stakeholders involved in risk management activities. For the first group, the aim is to induce a better understanding of the geohazards and disasters they are confronted with in the media or in their daily lives; for the second, the objective is to generate discussion about risk management strategies. The game was tested with students in Belgium and with citizens, earth scientists, and risk managers in several African countries. Based on analysis of the most common game strategies observed, the players' reactions during the game, and their answers to a short questionnaire, we analyzed the main learning outcomes conveyed by this game. The game Hazagora appears to positively enhance the players' insights into processes involved in disasters. As such, the game is an effective, fun learning tool to introduce participants to the concepts of geohazards and disasters and to generate discussion.

Volcanic edifices can collapse during their lifetime, often more than once. Volcano collapses have been documented worldwide in various tectonic settings. They can affect small and large edifices and active or non-active volcanoes. Long- or short-term instabilities develop within or below the edifice, deform its shape and a sudden trigger might prompt volcano failure, resulting in a volcanic landslide. The volume portion of the edifice involved in the collapse can vary from ∼0.1 to 10 km 3 and depends on the size of the edifice and the location of the failure plane. A shallow-seated failure plane will likely not affect the volcanic plumbing system, while a deep-seated failure plane may impact the structure and the orientation of the volcanic and igneous plumbing system and can influence the post-collapse volcanic activity and type of erupted magma. This then influences how an active edifice can rebuild itself.

We present an integrated, spatially-explicit database of existing geochemical major-element analyses available from (post-) colonial scientific reports, PhD Theses and international publications for the Virunga Volcanic Province, located in the western branch of the East African Rift System. This volcanic province is characterised by alkaline volcanism, including silica-undersaturated, alkaline and potassic lavas. The database contains a total of 908 geochemical analyses of eruptive rocks for the entire volcanic province with a localisation for most samples. A preliminary analysis of the overall consistency of the database, using statistical techniques on sets of geochemical analyses with contrasted analytical methods or dates, demonstrates that the database is consistent. We applied a principal component analysis and cluster analysis on whole-rock major element compositions included in the database to study the spatial variation of the chemical composition of eruptive products in the Virunga Volcanic Province. These statistical analyses identify spatially distributed clusters of eruptive products. The known geochemical contrasts are highlighted by the spatial analysis, such as the unique geochemical signature of Nyiragongo lavas compared to other Virunga lavas, the geochemical heterogeneity of the Bulengo area, and the trachyte flows of Karisimbi volcano. Most importantly, we identified separate clusters of eruptive products which originate from primitive magmatic sources. These lavas of primitive composition are preferentially located along NE-SW inherited rift structures, often at distance from the central Virunga volcanoes. Our results illustrate the relevance of a spatial analysis on integrated geochemical data for a volcanic province, as a complement to classical petrological investigations. This approach indeed helps to characterise geochemical variations within a complex of magmatic systems and to identify specific petrologic and geochemical investigations that should be tackled within a study area.

Volumetric depletion of a subsurface body commonly results in the collapse of overburden and the formation of enclosed topographic depressions. Such depressions are termed sinkholes in karst terrains and pit craters or collapse calderas in volcanic terrains. This paper reports the first use of computed X-ray microtomography (mu CT) to image analog models of small-scale (similar to< 2 km diameter), high-cohesion, overburden collapse induced by depletion of a near-cylindrical ("stock-like") body. Time-lapse radiography enabled quantitative monitoring of the evolution of collapse structure, velocity, and volume. Moreover, mu CT scanning enabled nondestructive visualization of the final collapse volumes and fault geometries in three dimensions. The results illustrate two end-member scenarios: (1) near-continuous collapse into the depleting body; and (2) near-instantaneous collapse into a subsurface cavity formed above the depleting body. Even within near-continuously collapsing columns, subsidence rates vary spatially and temporally, with incremental accelerations. The highest subsidence rates occur before and immediately after a surface depression is formed. In both scenarios, the collapsing overburden column undergoes a marked volumetric expansion, such that the volume of subsurface depletion substantially exceeds that of the resulting topographic depression. In the karst context, this effect is termed "bulking," and our results indicate that it may occur not only at the onset of collapse but also during progressive subsidence. In the volcanic context, bulking of magma reservoir overburden rock may at least partially explain why the volume of magma erupted commonly exceeds that of the surface depression.

The Virunga Volcanic Province (VVP) represents the most active zone of volcanism in the western branch of the East African Rift System. While the VVP’s two historically active volcanoes, Nyamulagira and Nyiragongo, have built scoria cones and lava flows in the adjacent lava fields, several small phreatomagmatic eruptive centers lie along Lake Kivu’s northern shoreline, highlighting the potential for explosive magma-water interaction. Their presence in the densely urbanized Sake-Goma-Gisenyi area necessitates an assessment of their eruptive mechanisms and chronology. Some of these eruptive centers possess multiple vents, and depositional contacts suggest distinct eruptive phases within a single structure. Depositional facies range from polymict tuff breccia to tuff and loose lapilli, often impacted by blocks and volcanic bombs. Along with the presence of dilute pyroclastic density current (PDC) deposits, indicators of magma-water interaction include the presence of fine palagonitized ash, ash aggregates, cross-bedding, and ballistic impact sags. We estimate that at least 15 phreatomagmatic eruptions occurred in the Holocene, during which Lake Kivu rose to its current water level. Radiocarbon dates of five paleosols in the top of volcanic tuff deposits range between ∼2500 and ∼150 cal. year BP and suggest centennial- to millennial-scale recurrence of phreatomagmatic activity. A vast part of the currently urbanized zone on the northern shoreline of Lake Kivu was most likely impacted by products from phreatomagmatic activity, including PDC events, during the Late Holocene, highlighting the need to consider explosive magma-water interaction as a potential scenario in future risk assessments.

Geological deformation processes have been studied traditionally by observing the surface expression of the resultant structures, complemented since several decades with remote sensing and subsurface imaging by e.g. drilling and geophysical techniques. Scaled laboratory modeling is a common technique to simulate deformation processes and study their dynamics in a controlled environment. Whereas early efforts focused on qualitative descriptions of observed structures, model deformation is increasingly quantified by using advanced imaging and image analysis (laser scanning, Structure-from-Motion (SfM) photogrammetry and Digital Image Correlation (DIC); Galland et al., 2016). For granular crustal analogues, typical laboratory monitoring techniques are limited to quantifying 3D deformation at the model surface, to recording the final deformation state in sliced and digitized model cross-sections (e.g. Ravaglia et al., 2004), or to quantifying 2D deformation along a quasi-interior vertical cross-section at the model side-wall where friction effects could affect deformation rates (e.g. Ruch et al., 2012). The use of transparent gelatin as a crustal analogue enables an interior view, but this limits the model rheology to idealized elastic deformation (Kavanagh et al., 2013). Granular materials enable simulation of elasto-plastic crustal behavior, but the model interior itself remains a black box, its deformation studied only through indirect inferences, similar to nature. To surmount this issue, we used X-ray Computed Tomography (XCT) to dynamically image and quantify the interior deformation of laboratory experiments in 4D (i.e. 3D + time), as applied to analogue magma intrusion.

Scaled analogue models were used to study the structural influence of gravitational spreading and magma intrusion on the elongated volcanoes. A relation can be found between simulated structures and structural field observations of deformation at natural volcanoes. Intrusions are seen to diminish the expression of deformation structures associated with spreading, such as the basal thrust scarp, lower-flank horsts and grabens. The deformation velocity and direction were obtained by image analysis of images acquired during the simulations. In this manner, different zones of deformation types on a cone were identified.

Analogue sand box models have been widely used to study geo-tectonic deformation processes. While past outcomes of analogue models have commonly comprised qualitative structural descriptions, recent work has emphasized quantifying the geometric, kinematic and dynamic parameters of the deformation. This has led to two main trends in model development over the last number of years. Firstly, there has been an increased emphasis on 3D quantification of model deformation. While optical image analysis techniques (e.g. Structure-from-Motion (SfM) photogrammetry and Digital Image Correlation (DIC); see e.g. Galland et al. 2016) used for this purpose are limited to model surfaces, there is an emerging possibility of full 3D imaging of deformation of both the model surfaces and the model interior by X-ray Computed Tomography (XCT) (Adam et al., 2013). Complicated trade-off relations between image resolution (voxel size) and maximum object size, scanning time and high operational costs have led to very few applicationsof this powerful imaging technique to analogue models so far. Secondly, there has been deeper investigation of the mechanics of modeling materials, such as granular analogues for upper crustal rocks (e.g. Abdelmalak et al. 2016; Klinkmüller et al. 2016). Sand-filler mixtures used for this purpose, especially in volcano-tectonic analogue models, have not yet received detailed attention, however. In our study, we quantified the physical and X-ray imaging properties of a range of mixtures of fine-grained fillers and silica sand. We assessed the mechanical impact of powdered gypsum (plaster) and kaolin clay as fillers, but we also tested the effect of adding more dense garnet sand into the silica sand to provide texture in XCT images. Using a Schülze Ring-Shear Tester (RST) and a Hubert-type shear box, we estimated the density, cohesion and friction coefficient of the granular materials, both as end-members and when mixed at different volumetric ratios. We tested both pouring and sieving from 20 cm height as emplacement techniques. By using medical XCT scanners, we also estimated the X-ray attenuation characteristics of the four above granular materials and three liquid or visco-elastic materials (corn oil, golden syrup, silicon putty). We find that the handling technique (sieving vs. pouring) had a strong impact on the compositional homogeneity and mechanics of the mixed materials. In general, sieving resulted in layered (i.e. inhomogeneous) bulk materials because of the strong size and density contrast between the endmember materials (sand vs. filler). Pouring increased compositional homogeneity, but resulted in a less dense packing of the bulk material, causing a larger amount of distributed deformation to occur before strain localisation. RST results show that the mechanical behavior of pure fillers (plaster, kaolin clay) displays complex, non-brittle components and stick-slip behavior. Adding filler to sand progressively increases the mechanical heterogeneity of the mixture. Friction coefficients increased from ~0.6 to ~0.8 and cohesions from few tens to few hundreds Pa. Pouring the granular materials instead of sieving resulted in a progressive deviation of the mechanics of sand-filler mixtures from the brittle Mohr-Coulomb behavior typical for granular materials (Lohrmann et al. 2003). Both the emplacement technique and the amount of filler mixed into sand hence significantly influence the mechanics of the mixture, potentially due to the contrast in their respective mean grain sizes. The approximation of sand-filler mixtures as Mohr-Coulomb-type brittle material might therefore be an oversimplification of their true rheology. The X-ray attenuation of the tested materials varied over a range of X-ray energies and the nature of the attenuation curve differed largely between the liquids (near-linear curve) and granular materials which are of much higher density (sigmoidal curve). In this way, sand-filler mixtures provided a clear contrast with the liquid magma analogues at an X-ray energy of 100-120 KeV. These variations should be taken into account when setting up the optimal XCT scan acquisition parameters for obtaining sharp and contrast-rich XCT images of a sand box model. Finally, we obtained a time series of XCT scans of an analogue simulation of golden syrup injection in a granular sand-plaster host, analogue to the intrusion of low-viscous magma in dense volcanic rock. We demonstrate that the constrained physical properties help to produce contrast-rich images of an analogue experiment in which the host medium, the propagating liquid body and formed structures are efficiently imaged during the ongoing injection process. The outcome of our work should serve as a useful reference for quantification of analogue model deformation in 3D by using dynamic X-ray CT. References Abdelmalak, M. M., Bulois, C., Mourgues, R., Galland, O., Legland, J. B., Gruber, C., 2016. Description of new dry granular materials of variable cohesion and friction coefficient: Implications for laboratory modeling of the brittle crust.Tectonophysics, In press. Adam, J., Klinkmüller, M., Schreurs, G., Wieneke, B., 2013. Quantitative 3D strain analysis in analogue experiments simulating tectonic deformation: Integration of X-ray computed tomography and digital volume correlation techniques. Journal of Structural Geology, 55, 127–149. Galland, O., Bertelsen, H.S., Guldstrand, F., Girod, L., Johannessen, R., Bjugger, F., Burchardt, S., Mair, K., 2016. Application of open-source photogrammetric software MicMac for monitoring surface deformation in laboratory models. Journal of Geophysical Research: Solid Earth, 121:4, 2852-2872. Klinkmüller, M., Schreurs, G., Rosenau, M., Kemnitz, H., 2016. Properties of granular analogue materials: A community wide survey. Tectonophysics, 666, doi.: 10.1016/j.tecto.2016.01.017. Lohrmann, J., Kukowski, N., Adam, J., Oncken, O., 2003. The impact of analogue material properties on the geometry, kinematics, and dynamics of convergent sand wedges. Journal of Structural Geology, 25:10, 1691-1711.

The use of scaled laboratory modeling to study volcanic deformation processes increasingly involves quantitative image analysis. For granular crustal analogs, deformation is usually monitored at the model surface or along a vertical cross-section at the model side-wall, and only the final deformation state is recorded in model cross-sections. Transparent gelatin as an alternative crustal analog enables an interior view, but it only accommodates idealized elastic deformation. As a result, our understanding of magma propagation modes and the associated inelastic host rock deformation from such models in 4D (i.e. 3D + time) is limited. To surmount this issue, we dynamically image fluid intrusion in a granular crustal analog by using wide beam X-ray Computed Tomography (XCT). We use golden syrup as a magma analog, and granular sand-plaster mixtures at varying mixture ratios as upper crustal rock analogs of varying strengths. We simultaneously visualize the geometric development of the syrup body as well as that of subsurface fractures and surface deformation in the granular analog. We also use Digital Volume Correlation (DVC) to calculate incremental and cumulative 3D volume displacements and 3D strains in the granular host material over time. We image in 4D the development of surface displacement and subsurface strain fields related to the growth of various intrusion geometries – laccoliths, cone sheets and dykes -, which form as a function of the strength of the granular host material. Consequently, we show that the combination of XCT and analog modelling provides a novel basis for interpreting geological, geodetic and geophysical data related to volcanic deformation.

Although geophysical observations provide insight into the interaction of shallow magma intrusion and host rock deformation at active volcanic systems, we remain unable to observe these processes directly. Understanding these subsurface processes in all their dimensions, possibly by combining geophysical observables with analytical and numerical models, is nevertheless critical for ultimately forecasting volcanic eruption. We present a novel analogue modeling set-up for magma intrusion in brittle rock, based on X-ray Computed Tomography (XCT). Firstly, we quantified in detail the mechanical behavior of the often-used, but poorly-understood, brittle rock analogue sand-and-plaster mixed at a range of ratios. Secondly, we show how varying the mechanical strength of this brittle host material results in variations of magma analogue intrusion geometries. Thirdly, through medical wide beam XCT, we collected a series of 3D scan images of each entire experiment over a sequence of time increments. We then use this 4D (3D + time) imagery to study (1) the interaction of intrusion growth with deformation (i.e. fault growth and displacement) in the host and (2) the deformation at the model surface. The major novelty of our approach is that it extends 4D monitoring and quantification of deformation beyond the model surface deformation by enabling a full quantification of the entire intrusion and host rock system through time. In the example experiment series, we found that host rock strength controls intrusion geometry and propagation mode, and we document the development of the intrusion-induced system of faults and fractures in the host. Our modeling set-up envisions the interpretation of intrusion-induced deformation and seismicity, and potential validation of analytical and numerical models used in geodesy of active volcanic systems.

Analogue models of caldera collapse were imaged by computerized X-ray tomography (µCT). Interval µCT radiography sequences document '2.5D' surface and internal model deformation in an unprecedented way, and carry the potential for a better understanding of the kinematics of various volcano-tectonic processes, of which caldera collapse is a mere illustration. A semi-automatic subsidence velocity analysis was carried out on radiographs. The developed method is a step towards the quantitative documentation of volcano-tectonic modelling that would render data interpretations immediately comparable to monitoring data available from recent deformation at natural volcanoes.