Search

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 computed tomography 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. [ABSTRACT FROM AUTHOR]

Detailing the relationship between stress and reactions in metamorphic rocks has been controversial, and much of the debate has centered on theory. Here, we add to this discussion and make a major advance by showing in time-resolved synchrotron microtomography experiments that a reacting and deforming sample experiencing an elastic differential stress produces a fabric orthogonal to the largest principal stress. This fabric forms very early in the reaction and can be shown to be unrelated to strain. The consequences of this are significant because a non-hydrostatic stress state is a very common geological occurrence. Our data provide the basis for new interpretations of the classical, and enigmatic, serpentine fabrics of Val Malenco, Italy, and Cerro del Almirez, Spain, where we relate the reported fabrics to transient, and cyclical, differential stresses from magma intrusion and the earthquake cycle. [ABSTRACT FROM AUTHOR]

Third- and fourth-generation synchrotron light sources with high fluxes and beam energies enable the use of innovative X-ray translucent experimental apparatus. These experimental devices access geologically relevant conditions whilst enabling in situ characterization using the spatial and temporal resolutions accessible at imaging beamlines. Here, Heitt Mjölnir is introduced, a heated miniature triaxial rig based on the design of Mjölnir, but covering a wider temperature range and larger sample volume at similar pressure capacities. This device is designed to investigate coupled thermal, chemical, hydraulic and mechanical processes from grain to centimetre scales using cylindrical samples of 10 mm × 20 mm (diameter × length). Heitt Mjölnir can simultaneously reach confining (hydraulic) pressures of 30 MPa and 500 MPa of axial stress with independently controlled sample pore fluid pressure < 30 MPa. This internally heated apparatus operates to temperatures up to 573 K with a minimal vertical thermal gradient in the sample of <0.3 K mm-1. This new apparatus has been deployed in operando studies at the TOMCAT (Swiss Light Source), I12 JEEP (Diamond Light Source) and PSICHE' (Synchrotron SOLEIL) beamlines for 4D X-ray microtomography with scan intervals of a few minutes. Heitt Mjölnir is portable and modular, allowing a wide range of 4D characterizations of low-grade metamorphism and deformational processes. It enables spatially and temporally resolved fluid-rock interaction studies at conditions of crustal reservoirs and is suitable for characterization of material properties in geothermal, carbonation or subsurface gas storage applications. Technical drawings and an operation guide are included in this publication. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

In experiments designed to understand deep shear zones, we show that periodic porous sheets emerge spontaneously during viscous creep and that they facilitate mass transfer. These findings challenge conventional expectations of how viscosity in solid rocks operates and provide quantitative data in favour of an alternative paradigm, that of the dynamic granular fluid pump model. On this basis, we argue that our results warrant a reappraisal of the community's perception of how viscous deformation in rocks proceeds with time and suggest that the general model for deep shear zones should be updated to include creep cavitation. Through our discussion we highlight how the integration of creep cavitation, and its Generalised Thermodynamic paradigm, would be consequential for a range of important solid Earth topics that involve viscosity in Earth materials like, for example, slow earthquakes. [ABSTRACT FROM AUTHOR]

In experiments designed to understand deep shear zones, we show that periodic porous sheets emerge spontaneously during viscous creep, forming a hydro-mechanical anisotropy that influences mass transfer. These findings challenge the current paradigm of viscosity in solid rocks. In particular, they showcase how shear zones may actively focus mass transport and highlight the possibility that viscous rocks could locally transition from flow to fracture. Our work demonstrates that viscosity in solids is not directly comparable to viscosity in fluids and this is consequential for a range of important solid Earth topics, like slow earthquakes, the flow of glacial ice and the tectonics of exoplanets. [ABSTRACT FROM AUTHOR]

Creep cavities are increasingly recognized as an important syn‐kinematic feature of shear zones, but much about this porosity needs investigation. Largely, observations of creep cavities are restricted to very fine grained mature ultramylonites, and it is unclear when they developed during deformation. Specifically, a question that needs testing is should grain size reduction during deformation produce creep cavities? To this end, we have reanalyzed the microstructure of a large shear strain laboratory experiment that captures grain size change by dynamic recrystallization during mylonitization. We find that the experiment does contain creep cavities. Using a combination of scanning electron microscopy and spatial point statistics, we show that creep cavities emerge with, and because of, subgrain rotation recrystallization during ultramylonite formation. As dynamic recrystallization is ubiquitous in natural shear zones, this observation has important implications for the interpretation of concepts such as the Goetze criterion, paleopiezometery, and phase mixing. Plain Language Summary: At great depths inside the Earth, rocks called mylonites slowly deform and accommodate tectonic forces. Generally, these rocks are considered to have no porosity because the pressure they experience is very large. However, it is frequently documented that these mylonites focus the transport of mass, both fluid and solid, through the crust. This implies that mylonites host a permeable porosity. To better understand this paradox, we reanalyzed an old laboratory experiment that documented the formation of a mylonite. We showed that a porosity, known as creep cavities, forms synchronously with the mylonite. This is an important experimental finding because it suggests that creep cavities are a fundamental feature of mylonites. Our results showcase a rare snapshot into the dynamics of rocks important for tectonics and advance larger questions about their transport properties. Key Points: Creep cavities emerge with grain size reduction by subgrain rotation recrystallizationPorosity driven by creep can be opened and sustained at high confining pressuresThere is a direct and spontaneous physical path for single‐phase rocks to transition to polyphase rocks during deformation [ABSTRACT FROM AUTHOR]

The dispersal of monomineralic quartz domains in a quartzofeldspathic ultramylonite is interpreted to be the result of the emergence of syn-kinematic pores, called creep cavities. The cavities can be considered the product of two distinct mechanisms that formed hierarchically: Zener-Stroh cracking and viscous grain boundary sliding. In initially thick and coherent quartz ribbons deforming by grain size-insensitive creep, cavities were generated by the Zener-Stroh mechanism on grain-boundaries aligned with the YZ plane of finite strain. The opening of creep cavities promoted the ingress of fluids to sites of low stress. The local addition of a fluid lowered the adhesion and cohesion of grain-boundaries and promoted viscous grain boundary sliding. With the increased contribution of viscous grain boundary sliding, a second population of cavities formed to accommodate strain incompatibilities. Ultimately, the emergence of creep cavities is interpreted to be responsible for the transition of quartz domains from a grain size-insensitive, to a grain size-sensitive rheology. [ABSTRACT FROM AUTHOR]

Mylonitic shear zones are important fluid conduits in the Earth’s crust. They host transient and permeable porosity that facilitates fluid transfer and controls fluid-rock interaction. Here we present microstructural observations from a mid-crustal ultramylonite with very large pores that occupy the strain shadows of albite porphyroclasts. Our non-invasive three-dimensional X-ray microtomographic data show that the largest of these strain shadow megapores have substantial volumes of as much as ∼1.7 × 105 μm³. Given that the sample shows no signs of retrogressive overprint or weathering, these pores must be synkinematic. Importantly, the close proximity of the pores to creep cavities in dynamically recrystallized quartz ribbon grains suggests a potential hydraulic link between fluid in the strain shadow megapores and fluid in the creeping rock matrix. The evolving megapores constitute very large syndeformational local fluid reservoirs in mylonites that likely fed into the granular fluid pump established by the dynamically evolving creep cavities. Our findings add to an emerging picture of the dynamic transport properties of ultramylonitic shear zones, where the formation and destruction of porosity are intrinsically linked to microscale deformation processes. They also suggest that despite many studies on porphyroclast systems, open questions remain, especially concerning the interaction of clasts with their matrix. [ABSTRACT FROM AUTHOR]

Here we will address two possible applications of wavelet analysis for deformation microstructures: firstly, the isolation of features of interest and secondly the quantification of the frequency of those features.In microstructural analysis data is often complex with many features that have various fractal or periodic characteristics. Consequentially, when data is acquired it often contains an excess of data for the questions at hand. In image analysis of microstructures, data is often excluded qualitatively by eye. This is a time consuming exercise and might succumb to selection and confirmation biases. Here we show how wavelets can be used to find statistically significant features and isolate them. Furthermore, by virtue of the wavelet technique this analysis also yields the frequency of those features. Even more critically, the technique preserves the spatial arrangement of the features of interest. In this way, wavelet analysis overcomes some of the abstraction and loss of spatial information that classical Fourier frequency analysis is encumbered with. Wavelet analysis is routinely used in Geophysics and other wave based fields of study and is emerging in more of Earth sciences. Our analysis highlights two possibilities where the technique enhances the knowledge that can be obtained from quantitative microstructural analysis. [ABSTRACT FROM AUTHOR]

We present 230Th-normalized dust and export production fluxes for two contrasted marine sediment cores spanning the Antarctic Polar Front, close to the Kerguelen Plateau in the Southern Indian Ocean, covering the last glacial cycle. We report glacial lithogenic fluxes comparable to the South Atlantic and higher than in the South Pacific sectors of the Southern Ocean. Structural and temporal discrepancies with dust reconstructions from Antarctic ice cores and the Atlantic and Pacific sectors of the Subantarctic Zone (SAZ) point towards Southern Africa and/or the Kerguelen Plateau as an additional source of lithogenic material to the Southern Indian Ocean during the last ice age. In the SAZ, export production proxies respond to iron (Fe) fertilization with total organic carbon (TOC) fluxes as high as those previously reported from the Atlantic sector of the Southern Ocean. However, the correlation between export production and dust proxies is weaker than in the other sectors, and shows a muted response of export production during peak glacials. We hypothesize that this muted response may be related to macronutrient (co-)limitation imposed on phytoplankton growth possibly induced by a northward displacement of wind-driven upwelling and/or the polar frontal system during peak glacials. The Antarctic Zone (AZ) record depicts the typical pattern of enhanced export production during interglacials and comparatively low productivity during glacials suggesting a decrease in the supply of macronutrients to the AZ surface during ice ages compared to warm periods, as previously proposed. However, a muted response of opal fluxes during marine isotope stage (MIS) 5e argues for a southward migration of the frontal system during warmer MIS 5e, possibly causing silicic acid (co-)limitation at this site. These results illustrate the important role of Fe fertilization and changes in wind-driven upwelling intensities in modulating the strength and efficiency of the biological carbon pump in the Southern Ocean, but also highlight the potential role of frontal movements in explaining export production variability at sites located close to the Antarctic Circumpolar Current (ACC) fronts. Overall, our new datasets provide new insights on the regional differences in dust and export production patterns that need to be considered in order to evaluate the efficiency of carbon sequestration in the different sectors of the Southern Ocean during ice ages. • Multiple sources of dust supplied to the Southern Indian Ocean during ice ages. • Iron fertilization may have been impeded by nutrient (co-)limitation. • Nutrient (co-)limitation may be related to frontal migration. • Iron fertilization enhanced carbon export in all three basins of the Southern Ocean. [ABSTRACT FROM AUTHOR]