Crystals within volcanic rocks record geochemical and textural signatures during magmatic evolution before eruption. Clues to this magmatic history can be examined using crystal size distribution (CSD) studies. The analysis of CSDs is a standard petrological tool, but laborious due to manual hand-drawing of crystal margins. The trainable Weka segmentation (TWS) plugin in ImageJ is a promising alternative. It uses machine learning and image segmentation to classify an image. We recorded back-scattered electron (BSE) images of three volcanic samples with different crystallinity (35, 50 and ≥85 vol. %), using scanning electron microscopes (SEM) of variable image resolutions, which we then tested using TWS. Crystal measurements obtained from the automatically segmented images are compared with those of the manual segmentation. Samples up to 50 vol. % crystallinity are successfully segmented using TWS. Segmentation at significantly higher crystallinities fails, as crystal boundaries cannot be distinguished. Accuracy performance tests for the TWS classifiers yield high F-scores (>0.930), hence, TWS is a successful and fast computing tool for outlining crystals from BSE images of glassy rocks. Finally, reliable CSD’s can be derived using a low-cost desktop SEM, paving the way for a wide range of research to take advantage of this new petrological method.
1 December 2018 in International Journal of Applied Earth Observation and Geoinformation
Accurate numerical simulation can provide crucial information useful for a greater understanding of destructive granular mass movements such as rock avalanches, landslides, and pyroclastic flows. It enables more informed and relatively low cost investigation of significant risk factors, mitigation strategy effectiveness, and sensitivity to initial conditions, material, or soil properties. In this paper, a granular avalanche experiment from the literature is reanalyzed and used as a basis to assess the accuracy of discrete element method (DEM) predictions of avalanche flow. Discrete granular approaches such as DEM simulate the motion and collisions of individual particles and are useful for identifying and investigating the controlling processes within an avalanche. Using a superquadric shape representation, DEM simulations were found to accurately reproduce transient and static features of the avalanche. The effect of material properties on the shape of the avalanche deposit was investigated. The simulated avalanche deposits were found to be sensitive to particle shape and friction, with the particle shape causing the sensitivity to friction to vary. The importance of particle shape, coupled with effect on the sensitivity to friction, highlights the importance of quantifying and including particle shape effects in numerical modeling of granular avalanches.
30 November 2022 in Earth and Planetary Science Letters
Probabilistic quantification of lahar hazard is an important component of lahar risk assessment and mitigation. Here we propose a new approach to probabilistic lahar hazard assessment through coupling a lahar susceptibility model with a shallow-layer lahar flow model. Initial lahar volumes and their probabilities are quantified using the lahar susceptibility model which establishes a relationship between the volume of mobilised sediment and exceedance probabilities from rainfall intensity-frequency-duration curves. Rainfall-triggered lahar hazard zones can then be delineated probabilistically by using the mobilised volumes as an input into lahar flow models. While the applicability of this model is limited to rain-triggered lahars, this approach is able to reduce the reliance on historic and empirical estimates of lahar hazard and creates an opportunity for the generation of purely quantitative probabilistic lahar hazard maps. The new approach is demonstrated through the generation of probabilistic hazard maps for lahars originating from the Mangatoetoenui Glacier, Ruapehu volcano, New Zealand.
Purpose: The purpose of this paper is to show how simulation of the flow of particulates and fluids using discrete element modelling (DEM) and smoothed particle dynamics (SPH) particle methods, offer opportunities for better understanding the dynamics of flow processes.Design/methodology/approach: DEM and SPH methods are demonstrated in a broad range of computationally‐demanding applications including comminution, biomedical, geophysical extreme flow events (risk/disaster modelling), eating of food by humans and elite water‐based sports.Findings: DEM is ideally suited to predicting industrial and geophysical applications where collisions between particles are the dominant physics. SPH is highly suited to multi‐physics fluid flow applications in industrial, biophysical and geophysical applications. The advantages and disadvantages of these particle methods are discussed.Research limitations/implications: Research results are limited by the numerical resolution that can currently be afforded.Practical implications: The paper demonstrates the use of particle‐based computational methods in a series of high value applications. Enterprises that share interests in these applications will benefit in their product and service development by adopting these methods.Social implications: The ability to model disasters provides governments and companies with the opportunity and obligation to use these to render knowable disasters which were previously considered unknowable. The ability to predict the breakdown of food during eating opens up opportunities for the design of superior performing foods with lower salt, sugar and fat that can directly contribute to improved health outcomes and can influence government food regulatory policy.Originality/value: The paper extends the scale and range of modelling of particle methods for demanding leading‐edge problems, of practical interest in engineering and applied sciences.