Supplemental Material: Prediction of CO2 Content in Mid-Ocean Ridge Basalts Via a Machine Learning Approach

Tian-Ting Lei,Liu jia,Qunke Xia,et al.

DOI: https://doi.org/10.1130/geol.s.26948278.v1

2024-01-01

Abstract:Details on data collection and machine learning analysis process.

What problem does this paper attempt to address?

Machine Learning for Reconstructing the Primary Carbon Contents of Mid-Ocean Ridge Basalts

Tian-Ting Lei,Jia Liu,Qun-Ke Xia

DOI: https://doi.org/10.5194/egusphere-egu24-7266

2024-01-01

Abstract:One of the primary locations of mafic magma production on Earth is the global mid-ocean ridge system, with basalts erupted from these ridges serving as valuable probes for assessing the compositional variability of the upper mantle and exploring the deep carbon cycle. However, directly measuring carbon contents in mid-ocean ridge basalts (MORBs) has proven challenging due to degassing during magma ascent. Early investigations indicate that some incompatible-trace-element- depleted and -enriched MORBs avoid heavy degassing, and show a narrow range of CO2/Ba ratios, which was generally applied to reconstruct the primitive CO2 content of global MORBs. However, increasing studies reveal significant variability in the CO2/Ba ratios of MORBs. Here, we compiled a dataset including the geochemical compositions of MORB glasses and melt inclusions for which studies supported no significant degassing. Based on it, we constructed a supervised machine learning (ML) model capable of accurately predicting CO2 contents in individual samples using the selected elemental contents. Applying our model to a global MORB database reveals that CO2 contents and CO2/Ba ratios of global MORBs are highly variable, highlighting the significance of mantle heterogeneity, which can be attributed to the interactions with deep-sourced plume, or the recycled components associated with the big subduction zone. Our findings underscore the potential of ML as a powerful tool for uncovering hidden structural patterns in complex geological data, shedding light on the intricate interplay between carbon, mantle composition, and Earth's long-term geological processes.
Prediction of CO2 content in mid-ocean ridge basalts via a machine learning approach

Tian-Ting Lei,Jia Liu,Qun-Ke Xia,Jing-Jun Zhou,Zhi-Kang Luan

DOI: https://doi.org/10.1130/g52466.1

IF: 5.8

2024-09-13

Geology

Abstract:One of the primary locations of mafic magma production on Earth is the global mid-ocean ridge system. The basalts erupted along ridges probe the upper mantle and can be used to explore the deep carbon cycle. However, mid-ocean ridge basalts (MORBs) degas heavily during magma ascent. Some incompatible-trace-element−depleted and −enriched MORBs avoid heavy degassing, and show a narrow range of CO2/Ba, which have been used to reconstruct the pre-eruptive CO2 content of primitive MORB. With an increasing amount of data, however, it has become apparent that the CO2/Ba ratios of MORBs vary significantly. We compiled a data set of the geochemical compositions of MORB glasses and melt inclusions that are not degassed significantly and used a supervised machine learning model to accurately predict CO2 contents of individual samples from the concentrations of selected elements. This approach reveals that predicted CO2 contents and CO2/Ba ratios of global MORBs are highly variable, highlighting the significance of mantle heterogeneity, which can be attributed to the interactions with deep-sourced plumes or recycled crust (oceanic crust with or without sediments). Our findings underscore the potential of machine learning as a powerful tool for investigating the intricate interplay between carbon, mantle composition, and Earth’s long-term geological processes.

geology
A Machine Learning Based‐Approach to Predict the Water Content of Mid‐Ocean Ridge Basalts

Jingjun Zhou,Jia Liu,Qunke Xia,Cheng Su,Takeshi Kuritani,Eero Hanski

DOI: https://doi.org/10.1029/2023gc010984

2023-01-01

Geochemistry Geophysics Geosystems

Abstract:Abstract Water is critical in the evolution of the mantle due to its strong influence on the physicochemical properties of mantle rocks. Mid‐ocean ridge basalts (MORBs) are commonly used to study the compositional characteristics of the convecting upper mantle. However, there remains abundant samples in the global MORB data sets without direct measurements of water contents. The commonly observed good correlation between H2O and other incompatible trace components, such as Ce, has been applied to quantify water contents of MORBs. However, this approach assumes constant H2O/Ce in the target samples, which is not always true as the H2O/Ce ratios of MORBs could be rather heterogeneous even in some short ridge segments. Utilizing the present compositional data of global MORB glasses with measured water contents (n = 1,467), we construct a Random Forest Regression model based on machine learning, which can predict water concentrations of samples based on selected major and trace element data, without assuming a ratio between H2O and other trace elements. This model allows us to precisely recover water contents for MORBs with comparable accuracy with traditional analytical methods. The predicted results of MORB glasses from this model (n = 1,931) expand the water content database of global MORBs and indicate a broad existence of high‐H2O MORBs. This new approach allows us to investigate the water content of MORBs from some ridges lacking previous water content measurements (e.g., the Chile Ridge and the Pacific‐Antarctic Ridge) and infer changes in the water content of MORB sources through applying the model to transform fault samples.
Recover the Water Content of Mid-Ocean Ridge Basalts by a Machine Learning Method

Jingjun Zhou,Jia Liu,Qunke Xia,Cheng Su,Takeshi Kuritani,Eero Hanski

DOI: https://doi.org/10.5194/egusphere-egu24-7440

2024-01-01

Abstract:Water's impact on the physicochemical attributes of mantle rocks makes it a pivotal factor in mantle evolution. Mid-ocean ridge basalts (MORBs) are essential for analyzing the upper mantle's composition, yet many global MORB samples lack direct water content assessment. The common method, using a correlation between H2O and trace elements like Ce to estimate MORB water contents, often presume a constant H2O/Ce ratio. Sometimes this assumption is unreliable due to the heterogeneity in H2O/Ce ratios, even within short ridge segments. For addressing this gap, we utilize compositional data from 1,467 global MORB glasses with measured water contents to develop a Random Forest Regression model. This machine learning-based model can predict water concentrations of MORB glasses based on major and trace element data, without the need for a fixed H2O/trace element ratio. Our model accurately recovers water contents of MORB glasses, showing comparable precision to traditional analytical methods. Applying this model to 1,931 MORB glass samples has significantly expanded the global MORB water content database, revealing the widespread presence of high-water MORBs. Importantly, this innovative approach enables the exploration of water content in MORBs from regions previously without such data, like the Chile Ridge and the Pacific-Antarctic Ridge. Moreover, it allows us to deduce variations of water contents of MORB sources by applying the model to transform fault samples, thereby offering novel insights into the dynamics of the mantle.
Geochemical fingerprinting of continental and oceanic basalts: A machine learning approach

Luc S. Doucet,Michael G. Tetley,Zheng-Xiang Li,Yebo Liu,Hamed Gamaleldien

DOI: https://doi.org/10.1016/j.earscirev.2022.104192

2022-10-01

Abstract:Basalts are ubiquitous, mantle-derived mafic rocks found within all tectonic settings. Studying the chemical composition of basalts has proven to be an effective way to understand tectonics-related mantle processes for more recent periods of Earth history when other geological and tectonic records are still well preserved. However, using basalt geochemistry to decipher ancient tectonic environments and mantle processes on Earth remains a significant challenge; interpretations are often non-unique, and weathering, erosion, and orogenic processes can modify the chemical composition of basalts, and eliminate or obscure other complementary geotectonic records. This is most apparent in the oceanic realm, where subduction-related processes have erased much of the geological record for times older than ∼200 Ma, only leaving small and rare dismembered blocks of oceanic lithosphere as ophiolite mélanges along tectonic sutures. As a result, workers must rely heavily on analyzing the chemical composition of preserved basalts to identify the tectonic settings of their original melt extraction from the mantle using various tectonic discrimination diagrams from the literature (e.g., Pearce and Cann, 1973). In this paper, we demonstrate that such widely used classic discrimination diagram approaches often suffer from the intrinsic shortfall of large ambiguity. Instead, we apply here a big-data approach to characterize basalts generated in some typical tectonic settings. We start with a bespoke data correction and machine learning workflow. Our results show that mid-ocean ridge basalts, ocean island basalts, continental flood basalts, arc basalts (from both the oceanic and continental realm), and oceanic flood basalts are statistically chemically different, thus presenting a novel and unique chemical ‘fingerprinting’ approach to more accurately discriminate basalts. The method successfully predicts the tectonic settings of basalt emplacement at a prediction accuracy of >96%.

geosciences, multidisciplinary
Machine Learning for Identification of Primary Water Concentrations in Mantle Pyroxene

Huan Chen,Cheng Su,Yong-Quan Tang,An-Zhou Li,Sen-Sen Wu,Qun-Ke Xia,J. ZhangZhou

DOI: https://doi.org/10.1029/2021gl095191

IF: 5.2

2021-01-01

Geophysical Research Letters

Abstract:The approach of estimating the H2O content of basaltic magmas via clinopyroxene (cpx) phenocrysts is a potentially effective way to glimpse the deep Earth water cycle. However, it is difficult to ascertain using traditional geochemical methods whether hydrogen (H) measured in cpx phenocrysts represents a primary signature that can ultimately inform estimates of the mantle water content. In this study, we conducted machine learning on the major element compositions and H2O content of cpx phenocrysts (1904 samples in total). Using the support vector machine (SVM), we defined a classifier (overall accuracy >92%) that can separate cpx that have undergone H diffusion, and thus modification of their original water content, from those that have not experienced H diffusion. Our trained SVM model has broad implications for understanding the primary water content of magma, the variations in water content during magma evolution, and the water cycle in the deep Earth.
Machine Learning Reveals Source Compositions of Intraplate Basaltic Rocks

Peng Guo,Ting Yang,Wen-Liang Xu,Bin Chen

DOI: https://doi.org/10.1029/2021gc009946

2021-01-01

Abstract:Recycling of crustal material is thought to introduce pyroxenite to the peridotite mantle. Mapping such lithological heterogeneity within the mantle is crucial to understanding the mantle's chemical evolution but remains challenging. By sampling the mantle source, intraplate basaltic melts provide a unique chance to reveal lithological heterogeneity within the mantle. We train machine learning (ML) models with major oxide data of experimental peridotite and pyroxenite melts to help reveal the mantle source lithology of basaltic rocks. The ML models can predict source lithologies from major oxide information with an accuracy larger than 94%. As a case study, we predict source lithology of the Cenozoic intraplate basaltic rocks in Northeast China. Our ML models suggest that pyroxenite dominates the mantle source of basaltic rocks sitting above the stagnant Pacific slab while peridotite dominates the source of the basaltic rocks located west of the slab tip, consistent with previous studies using other approaches. Our ML models could potentially be used to infer mantle source lithologies of basaltic rocks from other regions around the world.
Machine Learning Reveals the Influence of the Changbaishan Mantle Plume Sourced from the Mantle Transition Zone on Cenozoic Intraplate Magmatism in NE China

Yuning Qi,Huan Chen,Sensen Wu,Takeshi Kuritani,Zhenhong Du,Qunke Xia,Renyi Liu

DOI: https://doi.org/10.1016/j.chemgeo.2024.122282

IF: 3.9

2024-01-01

Chemical Geology

Abstract:This study investigates the influence of the Changbaishan mantle plume, formed by dehydration of stagnant slab, on Cenozoic intraplate magmatism in Northeast China, focusing on the plume's dominant influence zone. Compared to prior research, this paper includes a more extensive published dataset of major and trace element compositions of basalts and adopts machine learning algorithms with better generalization capabilities for the data being analyzed. This approach measures the geochemical similarity of basalts in Northeast China to training data from different geological backgrounds, determining their degree of influence by subduction-derived fluids. Global ocean island basalts (OIBs) and East African continental rift basalts (CRBs) serve as training data for subduction-distal basalts, while global island arc basalts (IABs) represent subduction-proximal basalts. A binary classification model is constructed through ridge logistic regression and support vector machine to establish a hyperplane in high-dimensional space. The greater the positive distance of the basalt data from the hyperplane, the closer it is to subduction-proximal basalts, and vice versa. Model feature importances correlate with their enrichment capacity in fluids, and the application of well-trained models to basalts with known water content shows a positive correlation between model results and H2O/Ce ratios, collectively demonstrating the models' ability to capture fluid signals. Model results of all Cenozoic basalts in Northeast China reveal that within a 600-km range centered on Changbaishan, fluid signals attenuate radially outward, rather than a continuous decrease along the subduction direction of the Pacific plate. This indicates a fluid-rich upwelling centered around Changbaishan, with fluid distribution decreasing from the center to the periphery. This upwelling plays a primary role in controlling the generation of Cenozoic basalts within a specific geographic range.
Explainable Machine Learning to Uncover Hydrogen Diffusion Mechanism in Clinopyroxene

Anzhou Li,Sensen Wu,Huan Chen,Zhenhong Du,Qunke Xia

DOI: https://doi.org/10.1016/j.chemgeo.2023.121776

IF: 3.9

2023-01-01

Chemical Geology

Abstract:Estimating the water content of mantle-derived magma using clinopyroxene (cpx) phenocrysts serves as a valuable constraint on the water budget in deep Earth. Intricate magma processes and the high hydrogen diffusion rate necessitate careful evaluations of whether the water content in cpx preserves its original state. Machine learning (ML) has been utilized to develop a classifier for judging hydrogen diffusion in cpx. Nevertheless, the opaqueness and complexity of most ML models hinder a clear understanding of their classification principles. To elucidate the mechanistic basis of the ML model, the Shapley theory is integrated to determine the contributions of major elements of cpx as features in a linear additive manner. This study achieves superior classification performance using an extreme gradient boosting model and innovatively presents a quantitative evaluation of feature importance at the sample level for each observation. The results indicate that Na plays a predominant role in the diffusion process surpassing other major elements and its associated hydrogen can easily diffuse out of cpx. Our model also identifies various hydrogen association modes in different elemental compositions and puts constraints on the properties of incorporated hydrogen with non-lattice forming elements in cpx. The findings demonstrate that the application of explainable ML methods in mineralogy holds significant potential for advancing the comprehension of geological phenomena.
Water Concentration Prediction of MORB and BABB by a Machine Learning Approach

Jingjun Zhou,Jia Liu,Lu Wang

DOI: https://doi.org/10.46427/gold2022.10778

2022-01-01

Abstract:Water is a key component which can change chemical and physical properties of the mantle materials greatly.It's amount and concentration in the mantle are essential to learn mantle dynamics.Mid-ocean ridge basalts (MORB) and back-arc basin basalts (BABB), sampling the depleted upper mantle and the latter including some signals of the subducted slabs, provide abundant information of the upper mantle.In this study, we've collected MORB and BABB glass data with water concentration from all over the world and introduced an approach of multivariate polynomial regression using a machine learning algorithm to estimate water concentration from the abundances of major oxides and some trace element.Coefficient of determination (R 2 ) and root-mean-square error (RMSE) are applied to evaluate the performance of our model.Recently, the best-performed models show R 2 of 0.822 and a RMSE of 0.181 wt% for the test set.This result shows the feasibility of predicting the water abundance from major and trace element.The proposed model has important implication for our understanding of mantle dynamics, mantle heterogeneity, and crust-mantle water recycling.
Machine learning with model selection to predict TOC from mineralogical constituents: case study in the Sichuan Basin

C. M. Saporetti,D. L. Fonseca,L. C. Oliveira,E. Pereira,L. Goliatt

DOI: https://doi.org/10.1007/s13762-022-04081-3

2022-04-04

International Journal of Environmental Science and Technology

Abstract:The total organic carbon content from rock samples is the fundamental quantitative and qualitative indicator of the existing organic matter in a reservoir. Generally, it is calculated manually through the analysis of rock samples of origin. However, this procedure demands time and resources since it depends on samples obtained from several intervals of wells in source rocks. Consequently, efforts on research have been conducted to assist this task. Machine learning approaches arise as an alternative to producing estimates for total organic carbon grounded on data well logs and stratigraphic analysis. Given this context, the present paper proposes using machine learning techniques to automate total organic carbon estimation. In order to provide flexibility to the model, a grid search procedure was combined with cross-validation to perform the model selection. This computational approach allows finding models that produced the best generalization capacity. Three methods were applied: Support Vector Machines, Extreme Learning Machine, and Ridge Regression. The proposed methodology was validated on core samples of the shale gas field YuDongNan area, Sichuan Basin. The Support Vector Machine method outperformed the other methods in several metrics analyzed, producing accurate predictions, showing that the approach present in this paper can be used as a surrogate model to assist geologists and petrologists in estimating total organic carbon values.

environmental sciences
Quantifying calcium carbonate and organic carbon content in marine sediments from XRF-scanning spectra with a machine learning approach

An-Sheng Lee,Weng-Si Chao,Sofia Ya Hsuan Liou,Ralf Tiedemann,Bernd Zolitschka,Lester Lembke-Jene

DOI: https://doi.org/10.1038/s41598-022-25377-x

2022-12-02

Abstract:Geochemical variations of sedimentary records contain vital information for understanding paleoenvironment and paleoclimate. However, to obtain quantitative data in the laboratory is laborious, which ultimately restricts the temporal and spatial resolution. Quantification based on fast-acquisition and high-resolution provides a potential solution but is restricted to qualitative X-ray fluorescence (XRF) core scanning data. Here, we apply machine learning (ML) to advance the quantification progress and target calcium carbonate (CaCO3) and total organic carbon (TOC) for quantification to test the potential of such an XRF-ML approach. Raw XRF spectra are used as input data instead of software-based extraction of elemental intensities to avoid bias and increase information. Our dataset comprises Pacific and Southern Ocean marine sediment cores from high- to mid-latitudes to extend the applicability of quantification models from a site-specific to a multi-regional scale. ML-built models are carefully evaluated with a training set, a test set and a case study. The acquired ML-models provide better results with R2 of 0.96 for CaCO3 and 0.78 for TOC than conventional methods. In our case study, the ML-performance for TOC is comparably lower but still provides potential for future optimization. Altogether, this study allows to conveniently generate high-resolution bulk chemistry records without losing accuracy.
Applications of machine learning in mantle geochemistry research

ZHANG Zhou,XIA Qun-ke

DOI: https://doi.org/10.19658/j.issn.1007-2802.2023.42.004

2023-01-01

Abstract:Thermodynamic equilibrium under high-pressure and high-temperature conditions are crucial for mantle geochemistry research. At present, there is still a lack of understanding of the thermodynamic equilibrium of minerals and melts under mantle conditions. The mantle geochemical data（pressure, temperature, and chemical compositions） are scattering points in high dimensional space. Machine learning is an effective approach to exploring the correlation of the scattering points in high space. Machine learning application includes geothermobarometer, elemental solubility in melts, ferric iron content in minerals, discrimination of mantle xenoliths with or without metasomatism, and discrimination of samples with or without hydrogen diffusion, etc. Development of this research direction requires a broader application of machine learning, enhancement of data accuracy rate in cyberinfrastructures, improvement of data quality, and modification of algorithms.
Lithogeochemistry Of The Mid-Ocean Ridge Basalts Near The Fossil Ridge Of The Southwest Sub-Basin, South China Sea

Kai Sun,Tao Wu,Xuesong Liu,Xue-Gang Chen,Chun-Feng Li

DOI: https://doi.org/10.3390/min10050465

IF: 2.5

2020-01-01

Minerals

Abstract:Mid-ocean ridge basalts (MORB) in the South China Sea (SCS) record deep crust-mantle processes during seafloor spreading. We conducted a petrological and geochemical study on the MORBs obtained from the southwest sub-basin of the SCS at site U1433 and U1434 of the International Ocean Discovery Program (IODP) Expedition 349. Results show that MORBs at IODP site U1433 and U1434 are unaffected by seawater alteration, and all U1433 and the bulk of U1434 rocks belong to the sub-alkaline low-potassium tholeiitic basalt series. Samples collected from site U1433 and U1434 are enriched mid-ocean ridge basalts (E-MORBs), and the U1434 basalts are more enriched in incompatible elements than the U1433 samples. The SCS MORBs have mainly undergone the fractional crystallization of olivine, accompanied by the relatively weak fractional crystallization of plagioclase and clinopyroxene during magma evolution. The magma of both sites might be mainly produced by the high-degree partial melting of spinel peridotite at low pressures. The degree of partial melting at site U1434 was lower than at U1433, ascribed to the relatively lower spreading rate. The magmatic source of the southwest sub-basin basalts may be contaminated by lower continental crust and contributed by recycled oceanic crust component during the opening of the SCS.
Heterogeneous Distribution of Water in Wet Mantle Plume Beneath Changbaishan, NE China: Constraints from Machine Learning

Yuning Qi,Huan Chen,Sensen Wu,Zhen-Hong Du,Qun-Ke Xia

DOI: https://doi.org/10.46427/gold2022.10428

2022-01-01

Abstract:The genesis of the Cenozoic intraplate basalts in Changbaishan area is considered to be related to a low-velocity zone in the underlying mantle.Previous studies suggested that this low-velocity zone corresponds to a wet mantle plume which originated from mantle transition zone.However, the origin and distribution of water in the wet mantle plume still remain unclear.In this study, the new approach of machine learning is used to provide a perspective on this issue.The major and trace element compositions of island arc basalts (IABs) and ocean island basalts (OIBs) are collated from GEOROC as training data.Based on the support vector machine (SVM) model, a hyperplane in high-dimensional space for binary classification is established.The signed distance of each sample to the hyperplane (i.e. the greater the distance is, the more similar the sample is to each class), is taken as the output value of the model.Positive values represent a similarity to IABs, and vice versa.Then we apply the trained SVM model to basalts in Changbaishan area.The predicted values of samples are close to IABs, indicating that the water in the mantle source of basalts in Changbaishan area is probably from the subducted Pacific slab.More importantly, the predicted values of samples display a negative relationship with their distances to the center of Changbaishan.That is to say, the water content in the mantle source of basalts is decreasing from the center of Changbaishan area to the edge.This finding further confirms the existence of the wet mantle plume beneath Changbaishan and indicates the geographical characteristics of water distribution in the wet mantle plume.
The Role of Continental Alkaline Magmatism in Mantle Carbon Outflux Constrained by a Machine Learning Analysis of Zircon

Lu Wang,Jia Liu,Christopher J. Spencer,Sensen Wu,Anzhou Li,Chengfeng Qiu,Qi Wu,Zubing Jia,Zizhen Wang,Hao Sun,Qun-Ke Xia

DOI: https://doi.org/10.1029/2023gl106847

IF: 5.2

2024-01-01

Geophysical Research Letters

Abstract:Continental alkaline magmatism has been suggested to play a significant role in releasing deep mantle carbon into the atmosphere, which can greatly impact the global climate. However, the temporal variations of alkaline magmatism and their potential to modulate climate over geologic time remain poorly constrained. The detrital zircon record is a frequently used proxy for tracking secular variations in particular magmatism. Here, we use a novel machine-learning technique to discriminate zircon from carbonatites, kimberlites, and other alkaline rocks. A global compilation of detrital zircon yields continental alkaline magmatic flare-ups between 1,050-850, 650-500, 250-200, and 50-0 Ma. Our estimates indicate relatively elevated contributions of total magmatic carbon outgassing from alkaline magmatism during the aforementioned magmatic flare-ups. We infer that anomalous alkaline magmatism may influence global warming during specific intervals of geologic time, but when they are not that voluminous or persistent extensive arc magmatism may drive warming conditions. The potential of volcanic CO2 emissions to modulate atmospheric CO2 levels and affect the environment of our planet has been recognized. Some specific volcanic types, such as continental arc volcanism, have been hypothesized to play a dominant role in driving long-term climate change. Recently, the efficiency of continental alkaline magmatism in releasing carbon from the deep mantle to the atmosphere and its ability to influence Earth's climate is proposed for certain timescales and supported by empirical data. However, the quantitative estimation of the alkaline magmatic activity over geologic time and its general link to the global climate change remains poorly constrained. Here, we assess the alkaline magmatic variations based on the detrital zircon record and a novel machine-learning model which could discriminate zircon from carbonatites, kimberlites and other related alkaline silicate rocks. The predictive result shows several peaks at 1,050-850, 650-500, 250-200, and 50-0 Ma, which is considered a minimum estimation due to the preservation bias of the detrital zircon record. Our estimates indicate that continental alkaline magmatism may influence global warming during specific intervals of geologic time such as the early Paleozoic and early Mesozoic. Frequency of magmatic events can be assessed by a machine-learning analysis of zircon Continental alkaline magmatism peaked between 1,050-850, 650-500, 250-200, and 50-0 Ma Anomalous alkaline magmatism may contribute to global warming at certain times
Prediction of TOC in Lishui–Jiaojiang Sag Using Geochemical Analysis, Well Logs, and Machine Learning

Xu Han,Dujie Hou,Xiong Cheng,Yan Li,Congkai Niu,Shuosi Chen

DOI: https://doi.org/10.3390/en15249480

IF: 3.2

2022-12-14

Energies

Abstract:Total organic carbon (TOC) is important geochemical data for evaluating the hydrocarbon generation potential of source rocks. TOC is commonly measured experimentally using cutting and core samples. The coring process and experimentation are always expensive and time-consuming. In this study, we evaluated the use of three machine learning (ML) models and two multiple regression models to predict TOC based on well logs. The well logs involved gamma rays (GR), deep resistivity (RT), density (DEN), acoustic waves (AC), and neutrons (CN). The ML models were developed based on random forest (RF), extreme learning machine (ELM), and back propagation neural network (BPNN). The source rock of Paleocene Yueguifeng Formation in Lishui–Jiaojiang Sag was taken as a case study. The number of TOC measurements used for training and testing were 50 and 27. All well logs and selected well logs (including AC, CN, and DEN) were used as inputs, respectively, for comparison. The performance of each model has been evaluated using different factors, including R2, MAE, MSE, and RMSE. The results suggest that using all well logs as input improved the TOC prediction accuracy, and the error was reduced by more than 30%. The accuracy comparison of ML and multiple regression models indicated the BPNN was the best, followed by RF and then multiple regression. The worst performance was observed in the ELM models. Considering the running time, the BPNN model has higher prediction accuracy but longer running time in small-sample regression prediction. The RF model can run faster while ensuring a certain prediction accuracy. This study confirmed the ability of ML models for estimating TOC using well logs data in the study area.

energy & fuels
A machine learning based line-by-line absorption coefficient model for the application of atmospheric carbon dioxide remote sensing

Fengxin Xie,Tao Ren,Ziqing Zhao,Changying Zhao

DOI: https://doi.org/10.1016/j.jqsrt.2022.108441

IF: 2.342

2022-11-30

Journal of Quantitative Spectroscopy and Radiative Transfer

Abstract:Space-based remote sensing, which is used to infer CO 2 concentration from the satellite-measured atmospheric spectral absorption signals, is an effective way to obtain CO 2 concentration data for greenhouse gas monitoring. The next-generation greenhouse gases monitoring satellites mainly address the challenge of improving the spatial and temporal resolutions of observations, which will dramatically increase the computational power required for the CO 2 retrievals. One of the bottlenecks is the high computational cost for the radiative heat transfer calculations, which involve inefficient high-resolution (usually requires a line-by-line spectral resolution) spectral modeling. Therefore, developing a fast and accurate spectral modeling method becomes necessary to tackle this problem. In the present study, we presented a machine learning based line-by-line absorption coefficient calculation (prediction) method for CO 2 in the applications of atmospheric remote sensing. By training an artificial neural network with data randomly generated from a line-by-line CO 2 absorption coefficient look-up table, a compact, accurate and efficient absorption coefficient prediction model can be developed, which only takes the CO 2 thermodynamic states as input. The proposed method has been tested by developing an absorption coefficient prediction model for the CO 2 1.6 μm spectral band, which was later used to simulate the measured spectra for clear-sky conditions from the Greenhouse gases Observing SATellite (GOSAT) for several different locations around the world. Results have shown that the model is both accurate and efficient. In addition, the same approach has been applied to fit the absorption coefficient tables provided by the Orbiting Carbon Observatory (OCO)-2 mission and an accurate prediction model was also presented.

optics,spectroscopy
Machine Learning Investigation of Clinopyroxene Compositions to Evaluate and Predict Mantle Metasomatism Worldwide

Ben Qin,Fang Huang,Shichun Huang,Andre Python,Yunfeng Chen,J. ZhangZhou

DOI: https://doi.org/10.1029/2021jb023614

2021-01-01

Abstract:Plain Language Summary Clinopyroxene is a major mineral in Earth's upper mantle. Previous studies have attempted to discriminate between reactions modifying the mantle by plotting clinopyroxene major and trace element compositions in two‐dimensional (2‐D) diagrams. However, these 2‐D methods show poor accuracy when applied to global datasets. Therefore, we suggest a machine learning approach to evaluate clinopyroxene compositional data in higher dimensions. Our results demonstrate that machine learning can significantly improve the accuracy of clinopyroxene compositional predictions over classical methods utilizing elemental ratios. Furthermore, the application of our algorithm to a global clinopyroxene dataset suggests that mantle metasomatism is globally widespread.
Machine Learning Fusion Multi-Source Data Features for Classification Prediction of Lunar Surface Geological Units

Wei Zuo,Xingguo Zeng,Xingye Gao,Zhoubin Zhang,Dawei Liu,Chunlai Li

DOI: https://doi.org/10.3390/rs14205075

IF: 5

2022-10-12

Remote Sensing

Abstract:Taking the Chang'e-4 and Chang'e-5 landing areas as the study areas, this study extracts the geological unit information from the regional USGS geological map, as well as the feature information such as topography and geomorphology, material composition and mineral abundance from Chang'e-2 DOM and DEM, wide angle camera (WAC) and Kaguya multi-band imager data. By applying methods including the statistical-based estimation of mutual information of data and the integrated-algorithmic-model-based evaluation of feature importance to this extracted information, we screen the significant features and construct a high-precision classification model by combining machine learning algorithm with important features of sample data. The practical application of the multi-classification prediction on the complex geological units in the two study areas achieves 97.9% and 95.1% accuracy. At the same time, the significant characteristics of the study area are mined, and the rules and knowledge associated with the geological evolution of the study area are obtained. In this study, we carry out research on quantitative prediction and identification of lunar surface geological units based on large samples and construct a high-precision multi-classification model to achieve automatic classification and prediction on large sample geological units with high accuracy. This method provides a new idea for the predicted mapping of geological units of lunar global digital mapping. In addition, it helps to fully exploit the useful information in the data and enrich the knowledge regarding the formation and evolution of the Moon.

environmental sciences,imaging science & photographic technology,remote sensing,geosciences, multidisciplinary

Supplemental Material: Prediction of CO2 Content in Mid-Ocean Ridge Basalts Via a Machine Learning Approach

Machine Learning for Reconstructing the Primary Carbon Contents of Mid-Ocean Ridge Basalts

Prediction of CO2 content in mid-ocean ridge basalts via a machine learning approach

A Machine Learning Based‐Approach to Predict the Water Content of Mid‐Ocean Ridge Basalts

Recover the Water Content of Mid-Ocean Ridge Basalts by a Machine Learning Method

Geochemical fingerprinting of continental and oceanic basalts: A machine learning approach

Machine Learning for Identification of Primary Water Concentrations in Mantle Pyroxene

Machine Learning Reveals Source Compositions of Intraplate Basaltic Rocks

Machine Learning Reveals the Influence of the Changbaishan Mantle Plume Sourced from the Mantle Transition Zone on Cenozoic Intraplate Magmatism in NE China

Explainable Machine Learning to Uncover Hydrogen Diffusion Mechanism in Clinopyroxene

Water Concentration Prediction of MORB and BABB by a Machine Learning Approach

Machine learning with model selection to predict TOC from mineralogical constituents: case study in the Sichuan Basin

Quantifying calcium carbonate and organic carbon content in marine sediments from XRF-scanning spectra with a machine learning approach

Applications of machine learning in mantle geochemistry research

Lithogeochemistry Of The Mid-Ocean Ridge Basalts Near The Fossil Ridge Of The Southwest Sub-Basin, South China Sea

Heterogeneous Distribution of Water in Wet Mantle Plume Beneath Changbaishan, NE China: Constraints from Machine Learning

The Role of Continental Alkaline Magmatism in Mantle Carbon Outflux Constrained by a Machine Learning Analysis of Zircon

Prediction of TOC in Lishui–Jiaojiang Sag Using Geochemical Analysis, Well Logs, and Machine Learning

A machine learning based line-by-line absorption coefficient model for the application of atmospheric carbon dioxide remote sensing

Machine Learning Investigation of Clinopyroxene Compositions to Evaluate and Predict Mantle Metasomatism Worldwide

Machine Learning Fusion Multi-Source Data Features for Classification Prediction of Lunar Surface Geological Units