Aiken, John Mark; Barras, Fabian; Renard, Francois; Hirth, Greg; Kelemen, Peter B. & Sohn, Robert A.
(2026).
Slowly Migrating Fracture Swarms in an Actively Serpentinizing Borehole.
Journal of Geophysical Research (JGR): Solid Earth.
ISSN 2169-9313.
131(2).
doi: 10.1029/2025jb032133.
Full text in Research ArchiveShow summary
Peridotite rocks are primary targets for engineered geological carbon sequestration efforts because the carbon in ‐bearing fluids is transferred to the rock in the form of carbonate minerals during alteration reactions. Sequestration efforts must necessarily open fractures in the rocks surrounding a pumped borehole, but the current understanding of fracture growth during serpentinization of peridotite is limited to theoretical models and laboratory experiments on small samples. We deployed hydrophone arrays in peridotite boreholes established by the Oman Drilling Program and detected downward migrating fracture swarms that represent the first field observations of fracture growth in a serpentinizing rock. More than 2 years after the boreholes were established, we detected four, downward migrating tensile fracture swarms during an interval of elevated pore pressure following large rain events. All of the swarms occurred within a partially confined section of the local aquifer, beginning at a depth of 170 m and migrating to the bottom of the 400 m‐deep hole at average rates of 6–20 cm.. We demonstrate that pore fluid processes can explain both the triggering of the tensile fracture swarms and their slow migration rates. Our results indicate that the tip of fractures propagating away from the borehole maintained near‐critical states over time such that relatively small increases in fluid pressure triggered tensile fracturing episodes, indicating that carbon sequestration efforts in mafic rocks should be able to open fractures and expose fresh rock to mineralization.
Liu, Tianze; Liang, Chao; Aiken, John Mark; Wu, Wenbo & Sohn, Robert
(2026).
Enigmatic Harmonic Tremor in the Samail Ophiolite, Oman: Evidence for Subsurface Gas Flow.
Journal of Geophysical Research (JGR): Solid Earth.
ISSN 2169-9313.
131(4).
doi: 10.1029/2025JB033691.
Full text in Research Archive
Aiken, John Mark; Dufournet, Elliot; Amiri, Hamed; Ternieten, Lotta & Plümper, Oliver
(2025).
An AI-Enabled Data Processing Pipeline for Ingesting Borehole Data in Peridotite Environments.
Journal of Geophysical Research (JGR): Machine Learning and Computation.
2(2).
doi: 10.1029/2025JH000666.
Full text in Research ArchiveShow summary
Researchers analyzing data collected from borehole drilling projects can face dozens of terabytes of seismic, hydrologic, geologic, and rock mechanics data, including complex imagery, physical measurements, and expert-written reports. These diverse data sets play a pivotal role in constraining solid-Earth processes. Ingesting and analyzing such data presents a colossal challenge that typically demands a team of experts and a lot of time. Artificial intelligence (AI) and machine learning have emerged as compelling approaches to tackle volume and complexity of drilling data. This paper presents an AI-based pipeline for ingesting data from the Oman Drilling Project's Multi-borehole Observatory. The study focuses on the alteration of peridotite core segments taken from Borehole BA1B, utilizing a gradient-boosted trees (CatBoost) regression model trained on an integrated data set of machine-learning segmented core images, physical measurements, geological, lithographic data, and AI-summarized expert texts and feature selection. This paper aims to establish a repeatable and efficient pattern for processing such multifaceted data from the well. We present results using the data set generated from BA1B. First, we examine the relationship between fracture/vein networks and peridotite alteration, a stand-in for historical fluid flows. Here we demonstrate that we do not find a strong relationship between these networks and alteration. Then we examine the very strong and also nonlinear relationship between alteration and the magnetic susceptibility and resistivity measured in BA1B.
Exploring the spatio-temporal variations of COVID-19 transmission and its potential determinants could provide a deeper understanding of the dynamics of disease spread. This study aimed to investigate the spatio-temporal spread of COVID-19 infections in England, and examine its associations with socioeconomic, demographic and environmental risk factors. We obtained weekly reported COVID-19 cases from 7 March 2020 to 26 March 2022 at Middle Layer Super Output Area (MSOA) level in mainland England from publicly available datasets. With these data, we conducted an ecological study to predict the COVID-19 infection risk and identify its associations with socioeconomic, demographic and environmental risk factors using a Bayesian hierarchical spatio-temporal model. The Bayesian model outperformed the ordinary least squares model and geographically weighted regression model in terms of prediction accuracy. The spread of COVID-19 infections over space and time was heterogeneous. Hotspots of infection risk exhibited inconsistent clustering patterns over time. Risk factors found to be positively associated with COVID-19 infection risk were: annual household income [relative risk (RR) = 1.0008, 95% Credible Interval (CI) 1.0005–1.0012], unemployment rate [RR = 1.0027, 95% CI 1.0024–1.0030], population density on the log scale [RR = 1.0146, 95% CI 1.0129–1.0164], percentage of Caribbean population [RR = 1.0022, 95% CI 1.0009–1.0036], percentage of adults aged 45–64 years old [RR = 1.0031, 95% CI 1.0024–1.0039], and particulate matter (PM2.5) concentrations [RR = 1.0126, 95% CI 1.0083–1.0167]. The study highlights the importance of considering socioeconomic, demographic, and environmental factors in analysing the spatio-temporal variations of COVID-19 infections in England. The findings could assist policymakers in developing tailored public health interventions at a localised level.
Bouchayer, Coline Lili Mathy; Aiken, John Mark; Thøgersen, Kjetil; Renard, Francois & Schuler, Thomas
(2022).
A Machine learning framework to automate the classification of surge-type glaciers in Svalbard.
Journal of Geophysical Research (JGR): Earth Surface.
ISSN 2169-9003.
127(7).
doi: 10.1029/2022JF006597.
Full text in Research ArchiveShow summary
Surge-type glaciers are present in many cold environments in the world. These glaciers experience a dramatic increase in velocity over short time periods, the surge, followed by an extended period of slow movement, the quiescence. This study aims at understanding why only few glaciers exhibit a transient behavior. We develop a machine learning framework to classify surge-type glaciers, based on their location, exposure, geometry, climatic mass balance and runoff. We apply this approach to the Svalbard archipelago, a region with a relatively homogeneous climate. We compare the performance of logistic regression, random forest, and extreme gradient boosting (XGBoost) machine learning models that we apply to a newly combined database of glaciers in Svalbard. Based on the most accurate model, XGBoost, we compute surge probabilities along glacier centerlines and quantify the relative importance of several controlling features. Results show that the surface and bed slopes, ice thickness, glacier width, climatic mass balance, and runoff along glacier centerlines are the most significant features explaining surge probability for glaciers in Svalbard. A thicker and wider glacier with a low surface slope has a higher probability to be classified as surge-type, which is in good agreement with the existing theories of surging. Finally, we build a probability map of surge-type glaciers in Svalbard. The framework shows robustness on classifying surge-type glaciers that were not previously classified as such in existing inventories but have been observed surging. Our methodology could be extended to classify surge-type glaciers in other areas of the world.
Wilson, Joseph; Pollard, Benjamin; Aiken, John; Caballero, Marcos & Lewandowski, H.J.
(2022).
Classification of open-ended responses to a research-based assessment using natural language processing.
Physical Review Physics Education Research.
18(1),
p. 010141-1–010141-16.
doi: 10.1103/PhysRevPhysEducRes.18.010141.
Full text in Research ArchiveShow summary
Surveys have long been used in physics education research to understand student reasoning and inform course improvements. However, to make analysis of large sets of responses practical, most surveys use a closed-response format with a small set of potential responses. Open-ended formats, such as written free response, can provide deeper insights into student thinking, but take much longer to analyze, especially with a large number of responses. Here, we explore natural language processing as a computational solution to this problem. We create a machine learning model that can take student responses from the Physics Measurement Questionnaire as input, and output a categorization of student reasoning based on different reasoning paradigms. Our model yields classifications with the same level of agreement as that between two humans categorizing the data, but can be done by a computer, and thus can be scaled for large datasets. In this work, we describe the algorithms and methodologies used to create, train, and test our natural language processing system. We also present the results of the analysis and discuss the utility of these approaches for analyzing open-response data in education research.
Aiken, John Mark; Sohn, Robert A.; Renard, Francois; Matter, Juerg; Kelemen, Peter & Jamtveit, Bjørn
(2022).
Gas Migration Episodes Observed During Peridotite Alteration in the Samail Ophiolite, Oman.
Geophysical Research Letters.
ISSN 0094-8276.
49(21).
doi: 10.1029/2022GL100395.
Full text in Research ArchiveShow summary
Serpentinization and carbonation of mantle rocks (peridotite alteration) are fundamentally important processes for a spectrum of geoscience topics, including arc volcanism, earthquake processes, chemosynthetic biological communities, and carbon sequestration. Data from a hydrophone array deployed in the Multi-Borehole Observatory (MBO) of the Oman Drilling Project demonstrates that free gas generated by peridotite alteration and/or microbial activity migrates through the formation in discrete bursts of activity. We detected several, minutes-long, swarms of gas discharge into Hole BA1B of the MBO over the course of a 9 month observation interval. The episodic nature of the migration events indicates that free gas accumulates in the permeable flow network, is pressurized, and discharges rapidly into the borehole when a critical pressure, likely associated with a capillary barrier at a flow constriction, is reached. Our observations reveal a dynamic mode of fluid migration during serpentinization, and highlight the important role that free gas can play in modulating pore pressure, fluid flow, and alteration kinetics during peridotite weathering.
McBeck, Jessica Ann; Aiken, John; Cordonnier, B.; Ben-Zion, Yehuda & Renard, Francois
(2021).
Predicting Fracture Network Development in Crystalline Rocks.
Pure and Applied Geophysics (PAGEOPH).
ISSN 0033-4553.
179,
p. 275–299.
doi: 10.1007/s00024-021-02908-7.
Full text in Research ArchiveShow summary
Abstract The geometric properties of fractures influence whether they propagate, arrest, or coalesce with other fractures. Thus, quantifying the relationship between fracture network characteristics may help predict fracture network development, and perhaps precursors to catastrophic failure. To constrain the relationship and predictability of fracture characteristics, we deform eight one centimeter tall rock cores under triaxial compression while acquiring in situ X-ray tomograms. The tomograms reveal precise measurements of the fracture network characteristics above the spatial resolution of 6.5 µm. We develop machine learning models to predict the value of each characteristic using the other characteristics, and excluding the macroscopic stress or strain imposed on the rock. The models predict fracture development more accurately in the experiments performed on granite and monzonite, than the experiments on marble. Fracture network development may be more predictable in these igneous rocks because their microstructure is more mechanically homogeneous than the marble, producing more systematic fracture development that is not strongly impeded by grain contacts and cleavage planes. The varying performance of the models suggest that fracture volume, length, and aperture are the most predictable of the characteristics, while fracture orientation is the least predictable. Orientation does not correlate with length, as suggested by the idea that the orientation evolves with increasing differential stress and thus fracture length. This difference between the observed and expected relationship between orientation and length highlights the influence of mechanical heterogeneities and local stress perturbations on fracture growth as fractures propagate, link, and coalesce.
Across the field of education research there has been an increased focus on the development, critique, and evaluation of statistical methods and data usage due to recently created, very large datasets and machine learning techniques. In physics education research (PER), this increased focus has recently been shown through the 2019 Physical Review PER Focused Collection examining quantitative methods in PER. Quantitative PER has provided strong arguments for reforming courses by including interactive engagement, demonstrated that students often move away from scientistlike views due to science education, and has injected robust assessment into the physics classroom via concept inventories. The work presented here examines the impact that machine learning may have on physics education research, presents a framework for the entire process including data management, model evaluation, and results communication, and demonstrates the utility of this framework through the analysis of two types of survey data.
Aiken, John & Lewandowski, H.J.
(2021).
Data sharing model for physics education research using the 70 000 response Colorado Learning Attitudes about Science Survey for Experimental Physics dataset.
Physical Review Physics Education Research.
17(2).
doi: 10.1103/PhysRevPhysEducRes.17.020144.
Full text in Research ArchiveShow summary
We present a model for sharing quantitative data in the field of physics education research and use it to present a newly available dataset as an example. This model is in line with calls from across physics and science more generally to democratize data and results through open access. The model includes suggestions for data collection, creation of a data schema, and data sharing. It attends to the specific needs of the physics education research community, such as anonymization of human subjects data. As an example of this model, we use the Colorado Learning Attitudes about Science Survey for Experimental Physics (E-CLASS) dataset, which includes over 70 000 responses to the E-CLASS survey. These data cover 133 institutions, 599 unique courses, and 204 instructors, and was collected between 2016 and 2019. These data are made available at the time of publication and can be used freely, without the need of any institutional review board approval.
The time it takes a student to graduate with a university degree is mitigated by a variety of factors such as their background, the academic performance at university, and their integration into the social communities of the university they attend. Different universities have different populations, student services, instruction styles, and degree programs, however, they all collect institutional data. This study presents data for 160,933 students attending a large American research university. The data includes performance, enrollment, demographics, and preparation features. Discrete time hazard models for the time-to-graduation are presented in the context of Tinto’s Theory of Drop Out. Additionally, a novel machine learning method: gradient boosted trees, is applied and compared to the typical maximum likelihood method. We demonstrate that enrollment factors (such as changing a major) lead to greater increases in model predictive performance of when a student graduates than performance factors (such as grades) or preparation (such as high school GPA).
McBeck, Jessica Ann; Aiken, John Mark; Mathiesen, Joachim; Ben-Zion, Yehuda & Renard, Francois
(2020).
Deformation precursors to catastrophic failure in rocks.
Geophysical Research Letters.
ISSN 0094-8276.
47(24).
doi: 10.1029/2020GL090255.
Full text in Research ArchiveShow summary
Forecasting the timing of catastrophic failure, such as crustal earthquakes, has been a central concern for centuries. Such forecasting requires identifying signals that evolve or accelerate in the precursory phase leading to failure, and the subset of signals that may be detected in the crust. We develop machine learning models to predict the proximity of catastrophic failure in synchrotron X‐ray tomography triaxial compression experiments on rocks using characteristics of evolving fracture networks. We then examine the characteristics that most strongly influence the model results, and thus may be considered the best predictors of the proximity of macroscopic failure. The resulting suite of predictive parameters underscores the importance of dilation in the precursory phase leading to catastrophic failure. The results indicate that the evolution of the strain energy density field may provide more robust predictions of the proximity of failure than other existing metrics of rock deformation.
McBeck, Jessica Ann; Aiken, John Mark; Ben-Zion, Yehuda & Renard, Francois
(2020).
Predicting the proximity to macroscopic failure using local strain populations from dynamic in situ X-ray tomography triaxial compression experiments on rocks.
Earth and Planetary Science Letters.
ISSN 0012-821X.
543.
doi: 10.1016/j.epsl.2020.116344.
Full text in Research ArchiveShow summary
Predicting the proximity of large-scale dynamic failure is a critical concern in the engineering and geophysical sciences. Here we use evolving contractive, dilatational, and shear strain deformation preceding failure in dynamic X-ray tomography experiments to examine which strain components best predict the proximity to failure. We develop machine learning models to predict the proximity to failure using time series of three-dimensional local incremental strain tensor fields acquired in rock deformation experiments under stress conditions of the upper crust. Three-dimensional scans acquired in situ throughout triaxial compression experiments provide a distribution of density contrasts from which we estimate the three-dimensional incremental strain that accumulates between each scan acquisition. Training machine learning models on multiple experiments of six rock types provides suites of feature importance that indicate the predictive power of each feature. Comparing the average importance of groups of features that include information about each strain component quantifies the ability of the contractive, dilatational and shear strain to predict the proximity of macroscopic failure. A total of 24 models of four machine learning algorithms with six rock types indicate that 1) the dilatational strain provides the best predictive power of the strain components, and 2) the intermediate values (25th-75th percentile) of the strain population provide the best predictive power of the statistics of the strain populations. In addition, the success of the predictions of models trained on one rock type and tested on other rock types quantifies the similarities and differences of the precursory strain accumulation process in the six rock types. These similarities suggest the potential existence of a unified theory of brittle rock deformation for a range of rock types.
Young, Nicholas; Allen, Grant; Aiken, John Mark; Henderson, Rachel & Caballero, Marcos Daniel
(2019).
Identifying features predictive of faculty integrating computation into physics courses.
Physical Review Physics Education Research.
15(1).
doi: 10.1103/PhysRevPhysEducRes.15.010114.
Full text in Research ArchiveShow summary
Computation is a central aspect of 21st century physics practice; it is used to model complicated systems, to simulate impossible experiments, and to analyze mountains of data. Physics departments and their faculty are increasingly recognizing the importance of teaching computation to their students. We recently completed a national survey of faculty in physics departments to understand the state of computational instruction and the factors that underlie that instruction. The data collected from the faculty responding to the survey included a variety of scales, binary questions, and numerical responses. We then used random forest, a supervised learning technique, to explore the factors that are most predictive of whether a faculty member decides to include computation in their physics courses. We find that experience using computation with students in their research, or lack thereof and various personal beliefs to be most predictive of a faculty member having experience teaching computation. Interestingly, we find demographic and departmental factors to be less useful factors in our model. The results of this study inform future efforts to promote greater integration of computation into the physics curriculum as well as comment on the current state of computational instruction across the United States.
McBeck, Jessica Ann; Kandula, Neelima; Aiken, John Mark; Cordonnier, Benoit & Renard, Francois
(2019).
Isolating the factors that govern fracture development in rocks throughout dynamic in situ X-ray tomography experiments.
Geophysical Research Letters.
ISSN 0094-8276.
46(20),
p. 11127–11135.
doi: 10.1029/2019GL084613.
Full text in Research ArchiveShow summary
Centuries of work have highlighted the importance of several characteristics on fracture propagation. However, the relative importance of each characteristic on the likelihood of propagation remains elusive. We rank this importance by performing dynamic X‐ray microtomography experiments that provide unique access to characteristics of evolving fracture networks as rocks are triaxially compressed toward failure. We employed a machine learning technique based on logistic regression analysis to predict whether or not a fracture grows from 14 fracture geometry and network characteristics identified throughout four experiments on crystalline rocks in which thousands of fractures propagated. The characteristics that best predict fracture growth are the length, thickness, volume, and orientation of fractures with respect to the external stress field and the distance to the closest neighboring fracture. Growing fractures tend to be more clustered, shorter, thinner, volumetrically smaller, and dipping closer to 30–60° from the maximum compression direction than closing fractures.
[This paper is part of the Focused Collection on Quantitative Methods in PER: A Critical Examination.] While other fields such as statistics and education have examined various issues with quantitative work, few studies in physics education research (PER) have done so. We conducted a two-phase study to identify and to understand the extent of these issues in quantitative PER. During phase 1, we conducted a focus group of three experts in this area, followed by six interviews. Subsequent interviews refined our plan. Both the focus group and interviews revealed issues regarding the lack of details in sample descriptions, lack of institutional or course contextual information, lack of reporting on limitation, and overgeneralization or overstatement of conclusions. During phase 2, we examined 72 manuscripts that used four conceptual or attitudinal assessments (Force Concept Inventory, Conceptual Survey of Electricity and Magnetism, Brief Electricity and Magnetism Assessment, and Colorado Learning Attitudes about Science Survey). Manuscripts were coded on whether they featured various sample descriptions, institutional or course context information, limitations, and whether they overgeneralized conclusions. We also analyzed the data to see if reporting has changed from the earlier periods to more recent times. We found that not much has changed regarding sample descriptions and institutional or course context information, but reporting and overgeneralizing conclusions have improved over time. We offer some questions for researchers, reviewers, and readers in PER to consider when conducting or using quantitative work.
Physics education research (PER) has used quantitative modeling techniques to explore learning, affect, and other aspects of physics education. However, these studies have rarely examined the predictive output of the models, instead focusing on the inferences or causal relationships observed in various data sets. This research introduces a modern predictive modeling approach to the PER community using transcript data for students declaring physics majors at Michigan State University. Using a machine learning model, this analysis demonstrates that students who switch from a physics degree program to an engineering degree program do not take the third semester course in thermodynamics and modern physics, and may take engineering courses while registered as a physics major. Performance in introductory physics and calculus courses, measured by grade as well as a students’ declared gender and ethnicity play a much smaller role relative to the other features included in the model. These results are used to compare traditional statistical analysis to a more modern modeling approach.
Solli, Robert; Aiken, John Mark; Henderson, Rachel & Caballero, Marcos
(2018).
Examining the relationship between student performance and video interactions.
Physics Education Research Conference Proceedings.
ISSN 1539-9028.
2018.
doi: 10.1119/perc.2018.pr.Solli.
Full text in Research Archive
Aiken, John Mark; Aiken, Chastity & Cotton, Fabrice
(2018).
A python library for teaching computation to seismology students.
Seismological Research Letters.
ISSN 0895-0695.
89(3),
p. 1165–1171.
doi: 10.1785/0220170246.
Full text in Research Archive
Knaub, Alexis V.; Aiken, John & Caballero, Marcos
(2019).
Editorial: Focused Collection: Quantitative Methods in PER: A Critical Examination.
Physical Review Physics Education Research.
15(2).
doi: 10.1103/PhysRevPhysEducRes.15.020001.
Full text in Research Archive
Aiken, John Mark
(2020).
Understanding University Student Pathways Towards Graduation with Machine Learning and Institutional Data.
Det matematisk naturvitenskapelige fakultet.
Full text in Research Archive
