Application of Machine Learning to Hydrothermal System Analysis: Geochemical Insights from the Bektakari–Bneli Khevi Ore Knot, Southern Georgia

Mantle plume plays an important role in modern seafloor hydrothermal mineralization system

Introduction: concept of TAIGA.- Geochemical constraints on potential biomass sustained by subseafloor water-rock interactions.- Microbial cell densities, community structures, and growth in the hydrothermal plumes of subduction hydrothermal systems.- Systematics of distributions of various elements between ferromanganese oxides and seawater from natural observation, thermodynamics, and structures.- Evaluating hydrothermal system evolution using geochronological dating and biological diversity analyses.- Quantification of microbial communities in hydrothermal vent habitats of the Southern Mariana Trough and the Mid-Okinawa Trough.- Development of hydrothermal and frictional experimental systems to simulate sub-seafloor water-rock-microbe interactions.- Experimental hydrogen production in hydrothermal and fault systems: Significance for habitability of subseafloor H2 chemoautotroph microbial ecosystems.- Experimental assessment of microbial effects on chemical interaction between seafloor massive sulfides and seawater at 4 .- A compilation of the stable isotopic compositions of carbon, nitrogen, and sulfur in soft body parts of animals collected from deep-sea hydrothermal vent and methane seep fields: variations in energy source and importance of subsurface microbial processes in the sediment-hosted systems.- Tectonic background of four hydrothermal fields along the Central Indian Ridge.- Indian Ocean hydrothermal systems: seafloor hydrothermal activities, physical and chemical characteristics of hydrothermal fluids, and vent-associated biological communities.- Petrology and geochemistry of mid-ocean ridge basalts from the southern Central Indian Ridge.- Petrology of peridotites and related gabbroic rocks around the Kairei-hydrothermal field in the Central Indian Ridge.- Distribution and Biogeochemical Properties of Hydrothermal Plumes in the Rodriguez Triple Junction.- Vent fauna in the Central Indian Ridge.- The mantle dynamics, the crustal formation, and the hydrothermal activity of the Southern Mariana Trough back-arc Basin.- Seismic structure and seismicity in the Southern Mariana Trough and their relation to hydrothermal activity.- Electrical resistivity structure of the Snail site at the Southern Mariana Trough spreading center.- Asymmetric seafloor spreading of the southern Mariana Trough back-arc basin.- Geochemical characteristics of active backarc basin volcanisms at the southern end of Mariana Trough.- Mineralogical and geochemical characteristics of hydrothermal minerals collected from hydrothermal vent fields in the Southern Mariana spreading center.- Dating of hydrothermal mineralization in active hydrothermal fields in the Southern Mariana Trough.- Intra-field variation of prokaryotic communities on and below the seafloor in the back-arc hydrothermal system of the Southern Mariana Trough.- Vent fauna in the Mariana Trough.- Population history of a hydrothermal vent-endemic snail Alviniconcha hessleri in the Mariana Trough.- Hydrothermal activity in the Okinawa Trough backarc basin -geological background and hydrothermal mineralization-.- Active rifting structures in Iheya Graben and adjacent area of the mid-Okinawa Trough observed through seismic reflection surveys.- ESR dating of barite in sea-floor hydrothermal sulfide deposits in the Okinawa Trough.- Fluid geochemistry of high-temperature hydrothermal fields in the Okinawa Trough.- Sediment-pore water system associated with native sulfur formation at Jade hydrothermal field in Okinawa Trough.- Comparative investigation of microbial communities associated with hydrothermal activities in the Okinawa Trough.- In situ determination of bacterial growth in mixing zone of hydrothermal vent field on the Hatoma Knoll, Southern Okinawa Trough.- Vent Fauna in the Okinawa Trough.- Brief report of side-scan sonar observations around the Yokoniwa NTO massif.- Examination of volcanic activity: AUV and submersible observations of fine-scale lava flow distributions along the Southern Mariana Trough spreading axis .- Brief report of side-scan sonar imagery observations of the Archaean, Pika, and Urashima hydrothermal sites.- The Yoron Hole: the shallowest hydrothermal site in the Okinawa Trough.- The Irabu Knoll: Hydrothermal site at the eastern edge of the Yaeyama Graben.- Tarama Knoll: Geochemical and biological profiles of hydrothermal activity.- Petrography and geochemistry of basement rocks drilled from Snail, Yamanaka, Archean, and Pika hydrothermal fields at the Southern Mariana Trough by Benthic Multi-coring System (BMS).- Pore fluid chemistry beneath active hydrothermal fields in the mid-Okinawa Trough: Results of shallow drilling by BMS during TAIGA11 cruise.- The characteristics of the seafloor massive sulfide deposits at the Hakurei Site in the Izena Hole, the Middle Okinawa Trough.- Occurrence of hydrothermal alteration minerals at the Jade hydrothermal field, in the Izena Hole, mid-Okinawa Trough.- Geochemistry of hydrothermal fluids collected from active hydrothermal systems in the southern Mariana Trough backarc spreading center.- Gamma ray doses in water around sea floor hydrothermal area in South Mariana.- 226Ra-210Pb and 228Ra-228Th dating of barite in submarine hydrothermal sulfide deposits collected at Okinawa Trough and South Mariana Trough.- OSL dating of sea floor sediments at the Okinawa Trough.- Immediate change of radiation doses from hydrothermal deposits.- Periodic behavior of deep sea current in the Hatoma Knoll hydrothermal system.- The gelatinous macroplankton community at the Hatoma Knoll hydrothermal vent.

Abstract. Hydrothermal alteration and mineralization processes can affect the physical and chemical properties of volcanic rocks. Aggressive acidic degassing and fluid flow often also lead to changes in the appearance of a rock, such as changes in surface coloration or intense bleaching. Although hydrothermal alteration can have far-reaching consequences for rock stability and permeability, limited knowledge exists on the detailed structures, extent, and dynamic changes that take place near the surface of hydrothermal venting systems. By integrating drone-based photogrammetry with mineralogical and chemical analyses of rock samples and surface gas flux, we investigate the structure of the evolving volcanic degassing and alteration system at the La Fossa cone on the island of Vulcano, Italy. Our image analysis combines principal component analysis (PCA) with image classification and thermal analysis through which we identify an area of approximately 70 000 m2 that outlines the maximum extent of hydrothermal alteration effects at the surface, represented by a shift in rock color from reddish to gray. Within this area, we identify distinct gradients of surface coloration and temperature that indicate a local variability in the degassing and alteration intensity and define several structural units within the fumarole field. At least seven such larger units of increased activity could be constrained. Through mineralogical and geochemical analysis of samples from the different alteration units, we define a relationship between surface appearance in drone imagery and the mineralogical and chemical composition. Gradients in surface color from reddish to gray correlate with a reduction in Fe2O3 from up to 3.2 % in the unaltered regime to 0.3 % in the altered regime, and the latter coincides with the area of increased diffuse acid gas flux. As the pixel brightness increases towards higher alteration gradients, we note a loss of the initial (igneous) mineral fraction and a change in the bulk chemical composition with a concomitant increase in sulfur content from close to 0 % in the unaltered samples to up to 60 % in samples from the altered domains. Using this approach of combined remote-sensing and in situ analyses, we define and spatially constrain several alteration units and compare them to the present-day thermally active surface and degassing pattern over the main crater area. The combined results permit us to present a detailed anatomy of the La Fossa fumarole field, including high-temperature fumaroles and seven larger units of increased alteration intensity, surface temperature, and variably intense surface degassing. Importantly, we also identify apparently sealed surface domains that prevent degassing, likely as a consequence of mineral precipitation from degassing and alteration processes. By assessing the thermal energy release of the identified spatial units quantitatively, we show that thermal radiation of high-temperature fumaroles accounts for < 50 % of the total thermal energy release only and that the larger part is emitted by diffuse degassing units. The integrated use of methods presented here has proven to be a useful combination for a detailed characterization of alteration and activity patterns of volcanic degassing sites and has the potential for application in alteration research and for the monitoring of volcanic degassing systems.

Geothermal fields and hydrothermal mineral deposits are manifestations of the interaction between heat transfer and fluid flow in the Earth’s crust. Understanding the factors that drive fluid flow is essential for managing geothermal energy production and for understanding the genesis of hydrothermal mineral systems. We provide an overview of fluid flow drivers with a focus on flow driven by heat and hydraulic head. We show how numerical simulations can be used to compare the effect of different flow drivers on hydrothermal mineralisation. We explore the concepts of laminar flow in porous media (Darcy’s law) and the non-dimensional Rayleigh number (Ra) for free thermal convection in the context of fluid flow in hydrothermal systems in three dimensions. We compare models of free thermal convection to hydraulic head driven flow in relation to hydrothermal copper mineralisation at Mount Isa, Australia. Free thermal convection occurs if the permeability of the fault system results in Ra above the critical threshold, whereas a vertical head gradient results in an upward flow field.

Hydrothermal mineralization at seafloor spreading centers

Fe-Pb-Sr isotopic systematics of the hydrothermal chimney from the Minami-Ensei hydrothermal field, middle Okinawa Trough: Constraint on hydrothermal mineralization process in incipient back-arc basin

Géochimie des interactions entre les Eaux, les minéraux et les roches: Yves Tardy (Editor), 1980. S.A.R.L. Eléments, 7, Place Parmentier, Terbes, France. Price FF80, 239 pp., paperback

A Primer on Machine Learning.

Foreword by Peter A. Cawood Acknowledgements. 1. Water and hydrothermal fluids on Earth 2. Hydrothermal processes and wall rock alteration 3. Tectonic settings, geodynamics and temporal evolution of hydrothermal mineral systems 4. Intrusion-related hydrothermal mineral systems 5. Porphyry systems fossil and active epithermal systems 6. Skarn systems 7. Submarine hydrothermal mineral systems 8. Metalliferous sediments and sedimentary rock-hosted stratiform and/or stratabound hydrothermal mineral systems 9. Orogenic, amagmatic and hydrothermal mineral systems of uncertain origin 10. Hydrothermal systems and the biosphere 11. Hydrothermal processes associated with meteorite impacts 12. Hydrothermal processes and systems on other planets and satellites 13. Uranium hydrothermal mineral systems References. Index

A comparative study of independent component analysis with principal component analysis in geological objects identification. Part II: A case study of Pinghe District, Fujian, China

Integrating soil geochemistry and machine learning for enhanced mineral exploration at the dayu gold deposit, south China block

As our understanding of seafloor hydrothermal systems grows, we recognize they are not always stable and sometimes show dramatic changes. In this review, the authors present a compilation of geochemical and geochronological studies that are helpful when investigating the evolving processes of submarine hydrothermal systems. Chapter II describes the systematics and methodology of three dating techniques with discussions on their application to minerals formed by seafloor hydrothermal activities. The K-Ar (Ar-Ar) technique is popular for dating igneous rocks, but it is not appropriate for dating hydrothermal minerals because potassium is a trace component of sulfide/sulfate minerals. Following recent progress, micro-analytical techniques applying laser fusion are applicable for dating fluid inclusions and/or hydrothermal alteration minerals, which could provide important geochronological information. Uranium and thorium series disequilibrium dating have been employed for previous geochronological studies of hydrothermal minerals obtained from submarine ore deposits. To cover a wide time range, it is necessary to use various combinations of parent and daughter nuclides. Applying ESR dating to hydrothermal minerals is a rather new challenge. Although it needs several investigations to establish the methodology, it could be a useful rapid dating technique for a time range of less than one thousand years. Chapter III introduces studies focusing on the evolution of seafloor hydrothermal activities over a short time scale (one week to a few years). Detection of event plumes associated with seafloor lava eruption brought an awareness of episodic hydrothermal activity triggered by magmatic perturbation. Subsequent dive studies revealed evolving geochemical processes, such as major changes of volatiles and elemental species concentrations of venting fluid. With remote real-time monitoring of acoustic T-waves generated by seafloor seismic activities, event detection and response cruises have been conducted successfully to investigate various evolving processes in more detail. Chapter IV introduces studies focusing on the evolution of seafloor hydrothermal activities over a long time scale (tens of thousands of years). Radiometric dating studies of hydrothermal minerals such as sulfide and manganese oxide collected from the TAG mound, which is one of the largest hydrothermal mound structures, reveal an age distribution over at least 15000 years separated by quiescent intervals lasting up to 2000 years. On slow spreading ridges such as the Mid-Atlantic ridge, major fracture systems focus the hydrothermal discharge at one place for more than one thousand years with repeated reactivation. In Chapter V, the authors discuss the direction of future studies. Although hydrothermal systems on mid-oceanic ridges have been well studied, those related to arc-backarc magmatic activities could provide more appropriate fields for studying the evolutionary process of submarine hydrothermal systems. Combining geochronological studies with geochemical and mineralogical studies would be important for reconstructing the evolution process in more detail.

Plants meet machines: Prospects in machine learning for plant biology

ABSTRACT: Machine learning (ML) has revolutionized petrophysical analysis by providing advanced tools to efficiently interpret well-log data and predict reservoir properties. In this study, key well logs, including Gamma Ray (GR), Resistivity (LLD, LLS, and MSFL), Neutron Porosity (NPHI), Bulk Density (RHOB), and Sonic (DT), were utilized to evaluate reservoir characteristics in the Meyal Oilfield. Supervised ML algorithms, such as Random Forest Regressor (RFR), Extra Trees Regressor (ETR), and Support Vector Machines (SVM), were deployed to predict critical properties, including porosity, permeability, water saturation, and shale volume. Dimensionality reduction via Principal Component Analysis (PCA) and clustering techniques like K-Means further enhanced feature selection and geological interpretation. The application of ensemble learning and artificial neural networks (ANNs) demonstrated exceptional accuracy in automating well-log interpretation, surpassing traditional methods in efficiency and precision. In addition, seismic data analysis was conducted using 2D seismic lines, integrating ML-predicted petrophysical properties with structural interpretation. Horizons corresponding to the Sakesar and Chorgali formations were delineated, revealing structural traps and fault systems crucial for hydrocarbon accumulation. This study underscores the transformative role of ML in subsurface reservoir characterization, highlighting its potential to optimize hydrocarbon exploration and production strategies in complex geological environments like the Potwar Basin. 1. INTRODUCTION Artificial neural networks (NNs) have been used in a study to forecast how well CO2 foam flooding will work to improve oil recovery on a lab scale. Using petrophysical data, created a model that uses artificial intelligence (AI) to forecast the porosity and permeability of petroleum reservoirs in addition to reservoir characteristics modeling, many scientists have created data-driven methods for wax deposition prediction and applied advanced machine learning techniques to the problems encountered in engineering, construction, and other industries (Akkurt et al., 2018). A comparative analysis of the integration of different machine-learning methods used to estimate the energy efficiency of buildings’ heating loads for smart city design. ML-based models to forecast permeability impairment due to scale deposition were studied, along with a comparison of various ML techniques for estimating the permeability and porosity of oil reservoirs using petrophysical logs (Al-Khalifa et al., 2020). When compared to traditional methods, it was found that the Extra Trees Regressor performed exceptionally well in estimating the volume of shale and porosities, while RFR and DTC were the most effective in modeling Sw and facies. This is ascribed to its capacity to accurately detect patterns in the training data and hence model reservoir features (Zhang et al., 2021). With a time-efficient approach and optimized results, suggested machine learning algorithms have effectively addressed the drawbacks of traditional methods, including generalization and data range, for petrophysical prediction without requiring extensive use of geological or lithological characteristics of the reservoir formation as shown in Figure 1 By adding more realistic data (core) and comprehensive data sets from all over the world with difficult reservoirs, this strategy can be further enhanced (Najwa et al., 2023) With a variety of inputs, machine learning algorithms can gain experience without being specifically designed to do so. Classification, continuous value prediction, and performance or event forecasting are among the predictions that may be made using the built model (Bader et al., 2019).

In many signal processing, machine learning and computer vision applications, one often has to deal with high dimensional and big datasets such as images, videos, web content, etc. The data can come in various forms, such as univariate or multivariate time series, matrices or high dimensional tensors. The goal of the data mining community is to reveal the hidden linear or non-linear structures in the datasets. Over the past couple of decades matrix factorization, owing to its intrinsic association with dimensionality reduction has been adopted as one of the key methods in this context. One can either use a single linear subspace to approximate the data (the standard Principal Component Analysis (PCA) approach) or a union of low dimensional subspaces where each data class belongs to a different subspace. In many cases, however, the low dimensional data follows some additional structure. Knowledge of such structure is beneficial, as we can use it to enhance the representativity of our models by adding structured priors. A nowadays standard way to represent pairwise affinity between objects is by using graphs. The introduction of graph-based priors to enhance matrix factorization models has recently brought them back to the highest attention of the data mining community. Representation of a signal on a graph is well motivated by the emerging field of signal processing on graphs, based on notions of spectral graph theory. The underlying assumption is that high-dimensional data samples lie on or close to a smooth low-dimensional manifold. Interestingly, the underlying manifold can be represented by its discrete proxy, i.e. a graph. A primary limitation of the state-of-the-art low-rank approximation methods is that they do not generalize for the case of non-linear low-rank structures. Furthermore, the standard low-rank extraction methods for many applications, such as low-rank and sparse decomposition, are computationally cumbersome. We argue, that for many machine learning and signal processing applications involving big data, an approximate low-rank recovery suffices. Thus, in this thesis, we present solutions to the above two limitations by presenting a new framework for scalable but approximate low-rank extraction which exploits the hidden structure in the data using the notion of graphs. First, we present a novel signal model, called `Multilinear low-rank tensors on graphs (MLRTG)' which states that a tensor can be encoded as a multilinear combination of the low-frequency graph eigenvectors, where the graphs are constructed along the various modes of the tensor. Since the graph eigenvectors have the interpretation of \textit{non-linear} embedding of a dataset on the low-dimensional manifold, we propose a method called `Graph Multilinear SVD (GMLSVD)' to recover PCA based linear subspaces from these eigenvectors. Finally, we propose a plethora of highly scalable matrix and tensor based problems for low-rank extraction which implicitly or explicitly make use of the GMLSVD framework. The core idea is to replace the expensive iterative SVD operations by updating the linear subspaces from the fixed non-linear ones via low-cost operations. We present applications in low-rank and sparse decomposition and clustering of the low-rank features to evaluate all the proposed methods. Our theoretical analysis shows that the approximation error of the proposed framework depends on the spectral properties of the graph Laplacians