ORCID Profile
0000-0002-1722-4081
Current Organisations
University of Edinburgh School of GeoSciences
,
University of Manchester
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Copernicus GmbH
Date: 15-05-2023
DOI: 10.5194/EGUSPHERE-EGU23-3424
Abstract: The 3rd and 4th generation of synchrotron light sources with their high brilliance, fluxes and beam energies allow the development of innovative X-ray translucent rock deformation apparatus that maximise these capabilities. Following on from the development of the Mjolnir triaxial deformation rig (Butler et al., 2020), we present an upscaled design: Heitt Mjolnir, covering a wider temperature range and larger s le volume while operating at similar pressure, enabling a wide range of time-resolved investigations. This device is designed to characterise coupled hydraulic, chemical and mechanical processes, occurring at various temperatures, from the & #181 m to the centimetre scale in cylindrical s les of 10 mm diameter and 20 mm length. Heitt Mjolnir can simultaneously reach confining pressures of & #8804 MPa (hydraulic), 500 MPa of axial stress while the s le& #8217 s pore fluid pressure is controlled in a dedicated fluid channel and can reach 30 MPa. This apparatus has an internal heating system and is able to reach temperatures of 573 K in the s le with a minimal vertical thermal gradient of .5 K/mm. This portable and modular device has been successfully deployed in operando studies at TOMCAT (SLS) and I12 JEEP (DLS) beamlines for 4D X-ray microtomography with scan intervals of a few minutes. Heitt Mjolnir allows the 4D characterisation of low-grade metamorphism, fluid-rock interaction and deformation processes. It enables spatially and temporally resolved fluid-rock interaction studies at a wide range of conditions and, by covering most geological reservoirs, will be particularly valuable for geothermal, carbonation or subsurface gas storage research.
Publisher: Copernicus GmbH
Date: 14-08-2023
DOI: 10.5194/EGUSPHERE-2023-1819
Abstract: Abstract. X-ray computed tomography has established itself as a crucial tool in the analysis of rock materials, providing the ability to visualise intricate 3D microstructures and capture quantitative information about internal phenomena such as structural damage, mineral reactions, and fluid-rock interactions. The efficacy of this tool, however, depends significantly on the precision of image segmentation, a process that has seen varied results across different methodologies, ranging from simple histogram thresholding to more complex machine learning and deep learning strategies. The irregularity in these segmentation outcomes raises concerns about the reproducibility of the results, a challenge that we aim to address in this work. In our study, we employ the mass balance of a metamorphic reaction as an internal standard to verify segmentation accuracy and shed light on the advantages of deep learning approaches, particularly their capacity to efficiently process expansive datasets. Our methodology utilises deep learning to achieve accurate segmentation of time-resolved volumetric images of the gypsum dehydration reaction, a process that traditional segmentation techniques have struggled with due to poor contrast between reactants and products. We utilise a 2D U-net architecture for segmentation and introduce machine learning-obtained labelled data (specifically, from random forest classification) as an innovative solution to the limitations of training data obtained from imaging. The deep learning algorithm we developed has demonstrated remarkable resilience, consistently segmenting volume phases across all experiments. Furthermore, our trained neural network exhibits impressively short run times on a standard workstation equipped with a Graphic Processing Unit (GPU). To evaluate the precision of our workflow, we compared the theoretical and measured molar evolution of gypsum to bassanite during dehydration. The errors between the predicted and segmented volumes in all time-series experiments fell within the 2 % confidence intervals of the theoretical curves, affirming the accuracy of our methodology. We also compared the results obtained by the proposed method with standard segmentation methods and found a significant improvement in precision and accuracy of segmented volumes. This makes the segmented CT images suited for extracting quantitative data, such as variations in mineral growth rate and pore size during the reaction. In this work, we introduce a distinctive approach by using an internal standard to validate the accuracy of a segmentation model, demonstrating its potential as a robust and reliable method for image segmentation in this field. This ability to measure the volumetric evolution during a reaction with precision paves the way for advanced modelling and verification of the physical properties of rock materials, particularly those involved in tectono-metamorphic processes. Our work underscores the promise of deep learning approaches in elevating the quality and reproducibility of research in the geosciences.
Publisher: Copernicus GmbH
Date: 15-05-2023
DOI: 10.5194/EGUSPHERE-EGU23-7041
Abstract: Fluid-rock interaction relies on the fluid& #8217 s ability to migrate through rocks, utilising permeable pore space. While we understand permeability in rocks that interact with fluids to evolve dynamically, e.g. in dehydration or carbonation reactions, we have very little quantitative information on these dynamics, as direct measurements of permeability in reacting rocks are inherently difficult.Here, we present a series of permeability measurements that capture the evolving fluid transport properties of dehydrating gypsum s les. To derive these measurements, we used an X-ray transparent deformation rig to document gypsum dehydration in 4-dimensional & #181 CT datasets and then modelled the permeability evolution for a segmented sub-volume numerically. In doing so, we were able to characterise the grain-scale porosity and permeability evolution of a dehydration reaction for the first time. We present analyses from two experimental time-series run at a fixed confining pressure, temperature and pore fluid pressure (Pc = 20 MPa T = ~125 C Pf = 5 MPa) but contrasting stress states: one with the largest principal stress (& #916 & #963 = 16.1 MPa) parallel to the s le cylinder axis and another the largest principal stress (& #916 & #963 = 11.3 MPa) being radial. In both cases, as pore space formed due to the negative change in the solid molar volume during the reaction, permeability evolved and increased congruently with porosity in time until ultimately reaching average values of 3.14E-13 m& #178 and 4.55E-13 m& #178 , respectively. A clear spatial heterogeneity of fluid flow develops at the grain-scale along with the fabrics in the s les. Importantly,& the calculated permeability tensors are anisotropic from the onset, but& develop over different spatiotemporal trajectories and have different preferred orientations in the two experimental geometries: If the anisotropy is expressed as 1-(min_eigenvalue/max_eigenvalue) of the permeability tensor (where isotropy = 0), then the experiment with the largest principal stress applied radially has a final anisotropy of 0.45, with fluid flow efficiently focussed into a vertical lineation. In the case with an axial largest principal stress, the final anisotropy of permeability is 0.30 with fluid flow being channelled along a foliation that developed orthogonally to & #963 .Our results suggest that the spatial and temporal developments of permeability during a dehydration reaction are controlled by the orientation and relative magnitudes of the principal stresses of a tectonic environment, and that these two parameters exert a strong control on the efficiency of drainage and thus reaction progress. This has consequences for our understanding of fluid movements in thrust tectonics and subduction zones, but also in applications such as the in-situ carbonation of ultramafic rocks.
Publisher: Copernicus GmbH
Date: 15-05-2023
DOI: 10.5194/EGUSPHERE-EGU23-6483
Abstract: Many metamorphic rocks have a fabric. What is often not clear is how much deformational or metamorphic processes contributed to the formation of these fabrics. Are foliations always the result of strain? When does intrinsic crystallographic anisotropy alone lead to the formation of structural elements? Understanding the relative contributions of deformation and metamorphism in rock fabrics is fundamentally important because it is foundational to understanding the role of stress in reacting and deforming rocks.To this end, we make a major advance in our understanding of fabric development in reacting rocks by showing in time-resolved (4D) synchrotron microtomography (& #181 CT) experiments that when a gypsum dehydration reaction occurs in a differentially stressed s le the reaction products develop orthogonally to the largest principal stress. This is an important finding because we can show with our & #181 CT data that this preferred orientation forms early in the reaction and at very small strains ( %). Using a simple kinematic model we can demonstrate that it cannot have formed because of reorientation during mechanical compaction. It remains to be established if it is nucleation or growth of bassanite that is being affected by the stress or both. Our experiments suggest that metamorphic transformations may be inherently anisotropic when reacting under the influence of a non-hydrostatic stress state.& The consequences of this are many. For ex le, there will be cases in natural rocks where the interpretation of a lineation, foliation or crystallographic preferred orientation as formed by strain may be incorrect. Moreover, the physical properties (e.g. hydraulic and mechanics) of metamorphic rocks could also be significantly anisotropic from early in a transformation. Mass transport pathways might initialise as channelled or partitioned conduits which would have an impact during subduction and in thin-skinned tectonics. Our data reveal a critical new finding related to the very common geological occurrence of reacting rocks experiencing a differential stress.
Publisher: Copernicus GmbH
Date: 15-05-2023
DOI: 10.5194/EGUSPHERE-EGU23-7120
Abstract: X-ray tomographic imaging has become a very valuable tool for the analysis of (rock) materials, both for visualising complex 3D microstructures and for imaging internal features such as damage, mineral reaction, and fluid/rock interactions quantitatively. The validity of the results derived from X-ray tomography, however, hinge on the& accuracy of the image segmentation. There are many methods for image segmentation (from simple manual thresholding to machine learning and deep learning approaches), which can produce a high range of variability in the segmentation results. Accuracy of segmentation results is seldom checked and thus calling the reproducibility of the results into question. In this contribution we show how metamorphic reactions themselves can be used to constrain accuracy and highlight the benefits of deep learning methods to extend this over many large datasets efficiently.Here, we demonstrate a methodology that uses deep learning to achieve reliable segmentation of time-series volumetric images of gypsum dehydration reaction, on which standard segmentation approaches fail due to insufficient contrast. We implement 2D U-net architecture for segmentation, and, to overcome the limitations of training data obtained experimentally through imaging, we show how labelled data obtained via machine learning (i.e., Random Forest Classification) can be used as input data and enhance the neural network performances. The developed deep learning algorithm proves to be incredibly robust, as it is able to consistently segment volume phases within the whole suite of experiments. In addition, the trained neural network exhibits short run times ( minutes for ~250 MB of image volumes) on a local workstation equipped with a GPU card.& & To confirm the precision achieved by our workflow, we consider the theoretical and measured molar evolution of gypsum (CaSO4.2H2O) to bassanite (CaSO4.& #189 H2O) during the dehydration. Within all time-series experiments, errors between the predicted theoretical and the segmented volumes fall within the 5% confidence intervals of the theoretical curves. Thus, the segmented CT images are very well suited for extracting quantitative information, such as mineral growth rate and pore size variations during the reaction. To our knowledge, this is the first time an internal standard is used to unequivocally measure the accuracy of a segmentation model.& Being able to accurately and unambiguously measure the volumetric evolution during a reaction enables high-level modelling and verification of the physical (hydraulic and mechanical) properties of rock materials involved in tectono-metamorphic processes.
Publisher: Copernicus GmbH
Date: 28-03-2022
DOI: 10.5194/EGUSPHERE-EGU22-8048
Abstract: & & Tectonic-scale features happening at convergent plates are ultimately the outcome of microscopic, grain scale processes. In collision zones, prograde metamorphism occurs by gradual increase of pressure and temperature [1 2]. Among the most important prograde mineral reactions are dehydration reactions, which are characterized by solid volume reduction, porosity creation, fluid release and high pore fluid pressures [3]. Most models linking dehydration and mechanical instabilities [4-6] involve feedback loops between coupled chemical, hydraulic and mechanical processes. Feedbacks control pore fluid pressure build-up and drainage, and provide efficient pathways for the transport of chemical components. Gypsum dehydration is crucial in the formation of detachment faults thin-skinned tectonics [7]. It is also used as a proxy for serpentine dehydration and the generation of intermediate depth seismic events/aseismic slip activity [8].& & & & We performed a set of experimental gypsum dehydrations both at the TOMCAT microtomography beamline at the Swiss Light Source, and in the laboratory. Using a modified version of the Mjolnir triaxial rig [9] that allowed control of pore fluid pressure in the synchrotron microtomography setup enabled us to document how differential stress (& #8710 & #963 ) and pore fluid pressure (P& sub& f& /sub& ) influence the dehydration of Volterra alabaster gypsum to bassanite at a constant confining pressure and temperature in 4D.& & & & We derived data on mineral phase transformation and formation of pore networks by applying a deep-learning algorithm in ORS Dragonfly& #174 software, which reduced data processing times, minimized interpretation biases, and allowed analysing larger volumes. The results exhibit an extremely high accuracy compared to standard procedures. The analysis of phase proportions (gypsum, bassanite and porosity) of segmented volumes correlates very well to theoretical predictions indicating a correct segmentation from the algorithms and self-consistency of the generated datasets. Comparing results obtained& at various & #8710 & #963 and P& sub& f& /sub& to the light of mechanical data and additional in-house experiments allows us to better interpret their effect on reaction duration, magnitude and textural evolution of the rock. Transient phenomena as well as in idual grain transformation and growth are now traceable in a fully automated way.& & & & Our data further our understanding of gypsum dehydration: We found that & #8710 & #963 greatly influences the assemblage of the bassanite needles, which tend to grow nearly vertical at & #8710 & #963 & #8773 0. Increasing & #8710 & #963 significantly increases s le compaction. On the contrary, increasing P& sub& f & /sub& decreases the bulk deformation and slows down the reaction. As pores grow around bassanite needles, the control of the orientation of needles by differential stress can influence the overall pore network and thus introduce anisotropies during transient and final stages of the reaction. Our data confirm that & #8710 & #963 and P& sub& f& /sub& greatly influence transient and final rock texture, which has implications on drainage during nappe emplacements.& & & & & strong& References& /strong& : [1] Hacker et al., 2003, /10.1029/2001JB001129 [2] Peacock, 2001, 10.1130/0091-7613(2001)029& :ATLPOD& .0.CO [3] Llana-Funez et al. 2012, /10.1007/s00410-012-0726-8 [4] Raleigh and Paterson, 1965 /10.1029/JZ070i016p03965& [5] Dobson et al., 2002 /10.1126/science.1075390 [6] Jung et al., 2004 /10.2747/0020-6814.46.12.1089 [7] Hubbert and Rubey, 1959 /10.1130/0016-7606(1959)70[115:ROFPIM]2.0.CO [8] Rutter et al. 2009 /10.1016/j.jsg.2008.09.008 [9] Butler 2020, /10.1107/S160057752001173X.& &
Location: United Kingdom of Great Britain and Northern Ireland
Location: United Kingdom of Great Britain and Northern Ireland
No related grants have been discovered for Damien Freitas.