Search NASA⌕ Search

SEARCH · Search NASA

Results for “machine learning, Random Forest”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 307 records · Page 17

Measuring Young Stars in Space and Time. II. The Pre-main-sequence Stellar Content of N44

The Hubble Space Telescope survey Measuring Young Stars in Space and Time (MYSST) entails some of the deepest photometric observations of extragalactic star formation, capturing even the lowest-mass stars of the active star-forming complex N44 in the Large Magellanic Cloud. We employ the new MYSST stellar catalog to identify and characterize the content of young pre-main-sequence (PMS) stars across N44 and analyze the PMS clustering structure. To distinguish PMS stars from more evolved line of sight contaminants, a non-trivial task due to several effects that alter photometry, we utilize a machine-learning classification approach. This consists of training a support vector machine (SVM) and a random forest (RF) on a carefully selected subset of the MYSST data and categorize all observed stars as PMS or non-PMS. Combining SVM and RF predictions to retrieve the most robust set of PMS sources, we find ∼26,700 candidates with a PMS probability above 95% across N44. Employing a clustering approach based on a nearest neighbor surface density estimate, we identify 16 prominent PMS structures at 1σ significance above the mean density with sub-clusters persisting up to and beyond 3σ significance. The most active star-forming center, located at the western edge of N44's bubble, is a subcluster with an effective radius of ∼5.6 pc entailing more than 1100 PMS candidates. Furthermore, we confirm that almost all identified clusters coincide with known H ii regions and are close to or harbor massive young O stars or YSOs previously discovered by MUSE and Spitzer observations.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

cTULIP: application of a human-based RNA-seq primary tumor classification tool for cross-species primary tumor classification in canine

The domestic dog, Canis familiaris, is quickly gaining traction as an advantageous model for use in the study of cancer, one of the leading causes of death worldwide. Naturally occurring canine cancers share clinical, histological, and molecular characteristics with the corresponding human diseases. In this study, we take a deep-learning approach to test how similar the gene expression profile of canine glioma and bladder cancer (BLCA) tumors are to the corresponding human tumors. We likewise develop a tool for identifying misclassified or outlier samples in large canine oncological datasets, analogous to that which was developed for human datasets. We test a number of machine learning algorithms and found that a convolutional neural network outperformed logistic regression and random forest approaches. We use a recently developed RNA-seq-based convolutional neural network, TULIP, to test the robustness of a human-data-trained primary tumor classification tool on cross-species primary tumor prediction. Our study ultimately highlights the molecular similarities between canine and human BLCA and glioma tumors, showing that protein-coding one-to-one homologs shared between humans and canines, are sufficient to distinguish between BLCA and gliomas. The results of this study indicate that using protein-coding one-to-one homologs as the features in the input layer of TULIP performs good primary tumor prediction in both humans and canines. Furthermore, our analysis shows that our selected features also contain the majority of features with known clinical relevance in BLCA and gliomas. Our success in using a human-data-trained model for cross-species primary tumor prediction also sheds light on the conservation of oncological pathways in humans and canines, further underscoring the importance of the canine model system in the study of human disease.

60 APPLIED LIFE SCIENCES↗

Identification of novel organic polar materials: A machine learning study with importance sampling

Recent advances in the synthesis of polar molecular materials have produced practical alternatives to ferroelectric ceramics, opening up exciting new avenues for their incorporation into modern electronic devices. However, in order to realize the full potential of polar polymer and molecular crystals for modern technological applications, it is paramount to assemble and evaluate all the available data for such compounds, identifying descriptors that could be associated with an emergence of ferroelectricity. In this paper, we utilized data-driven approaches to judiciously shortlist candidate materials from a wide chemical space that could possess ferroelectric functionalities. A machine learning study with importance sampling was employed to address the challenge of having a limited amount of available data on already-known organic ferroelectrics. Sets of molecular- and crystal-level descriptors were combined with a Random Forest Regression algorithm in order to predict the spontaneous polarization of the shortlisted compounds. First-principles simulations were performed to further validate the predictions obtained from the machine learning model.

36 MATERIALS SCIENCE↗

Spatial Mapping of Riverbed Grain-Size Distribution Using Machine Learning

Recent alluvial sediments in riverbeds play a significant role in controlling hydrologic exchange flows (HEFs) in river systems. The alluvial layer is usually associated with strong heterogeneity in physical properties (e.g., permeability and hydraulic conductivity), which affects local HEFs and therefore biogeochemical processes. The spatial distribution of these physical properties needs to be determined to inform the numerical models used to reveal the realistic hydro-biogeochemical behaviors. Such information can be obtained based on the intrinsic link between sediment grain-size distribution and hydraulic properties where sediment texture information is available. However, grain-size measurements are usually spatially sparse and do not have adequate coverage and resolution, particularly for a relatively large domain such as the Hanford Reach of the Columbia River. In this paper, we adopted machine learning (ML) approaches for categorizing and mapping the spatial distributions of riverbed substrate grain size and filling in missing areas of substrate data using the ML models along the reach. Such ML models for substrate size mapping were trained at 13,372 locations using measured substrate sizes along with observed and simulated attributes, including bathymetric attributes (e.g., elevation, slope, and aspect ratio) from LIDAR and bathymetric surveys, and hydrodynamic properties (e.g., water depth, velocity, shear stress, and their statistical moments). An ensemble bagging-based ML technique, Random Forest, was adopted to identify the most influential factors as predictors to develop the predictive models with over-fitting issues addressed. The models were evaluated with respect to each individual substrate size class and the lumped group, and then used to generate the final substrate size maps covering all the grid cells in the numerical modeling domain.

54 ENVIRONMENTAL SCIENCES↗

Field evaluation of semi‐automated moisture estimation from geophysics using machine learning

Geophysical methods can provide three-dimensional (3D), spatially continuous estimates of soil moisture. However, point-to-point comparisons of geophysical properties to measure soil moisture data are frequently unsatisfactory, resulting in geophysics being used for qualitative purposes only. This is because (1) geophysics requires models that relate geophysical signals to soil moisture, (2) geophysical methods have potential uncertainties resulting from smoothing and artifacts introduced from processing and inversion, and (3) results from multiple geophysical methods are not easily combined within a single soil moisture estimation framework. To investigate these potential limitations, an irrigation experiment was performed wherein soil moisture was monitored through time, and several surface geophysical datasets indirectly sensitive to soil moisture were collected before and after irrigation: ground penetrating radar, electrical resistivity tomography (ERT), and frequency domain electromagnetics (FDEM). Data were exported in both raw and processed form, and then snapped to a common 3D grid to facilitate moisture prediction by standard calibration techniques, multivariate regression, and machine learning. A combination of inverted ERT data, raw FDEM, and inverted FDEM data was most informative for predicting soil moisture using a random regression forest model (one-thousand 60/40 training/test cross-validation folds produced root mean squared errors ranging from 0.025–0.046 cm 3 /cm 3 ). This cross-validated model was further supported by a separate evaluation using a test set from a physically separate portion of the study area. Machine learning was conducive to a semi-automated model-selection process that could be used for other sites and datasets to locally improve accuracy.

54 ENVIRONMENTAL SCIENCES↗

Applications of Machine Learning to Predicting Core-collapse Supernova Explosion Outcomes

Most existing criteria derived from progenitor properties of core-collapse supernovae are not very accurate in predicting explosion outcomes. We present a novel look at identifying the explosion outcome of core-collapse supernovae using a machine-learning approach. Informed by a sample of 100 2D axisymmetric supernova simulations evolved with F ornax , we train and evaluate a random forest classifier as an explosion predictor. Furthermore, we examine physics-based feature sets including the compactness parameter, the Ertl condition, and a newly developed set that characterizes the silicon/oxygen interface. With over 1500 supernovae progenitors from 9-27 M ⊙ , we additionally train an autoencoder to extract physics-agnostic features directly from the progenitor density profiles. We find that the density profiles alone contain meaningful information regarding their explodability. Both the silicon/oxygen and autoencoder features predict the explosion outcome with ≈90% accuracy. In anticipation of much larger multidimensional simulation sets, we identify future directions in which machine-learning applications will be useful beyond the explosion outcome prediction.

79 ASTRONOMY AND ASTROPHYSICS↗

Machine Learning Prediction of H Adsorption Energies on Ag Alloys

Adsorption energies on surfaces are excellent descriptors of their chemical properties, including their catalytic performance. This has prompted a significant amount of work towards efficiently predicting adsorption energies without performing expensive quantum chemical calculations, based on semi-empirical or data scientific models. Here, we develop machine learning models to predict H adsorption energies on stepped Ag surface alloys using a database of energies obtained from first-principles electronic structure calculations based on density functional theory. Starting with a simple model and iteratively refining the features extracted from the calculations, we determine the key characteristics of successful models, and reduce the number of features required for accurate prediction of the adsorption energy. We find that nonlinear models are essential for accurate predictions. Moreover, detailed comparison of model predictions and calculations reveal a counterintuitive "selfpassivation" induced by some alloying elements, as well as unexpected discontinuous electronic structure effects. These findings are rationalized using tight-binding model hamiltonians and by analyzing the electronic density of states. Lastly, we show that we can estimate the reliability of a particular prediction, mitigating the effects of errors in the small fraction of cases where these errors are significant.

36 MATERIALS SCIENCE↗

THE MILKY WAY PROJECT: LEVERAGING CITIZEN SCIENCE AND MACHINE LEARNING TO DETECT INTERSTELLAR BUBBLES

We present Brut, an algorithm to identify bubbles in infrared images of the Galactic midplane. Brut is based on the Random Forest algorithm, and uses bubbles identified by >35,000 citizen scientists from the Milky Way Project to discover the identifying characteristics of bubbles in images from the Spitzer Space Telescope. We demonstrate that Brut's ability to identify bubbles is comparable to expert astronomers. We use Brut to re-assess the bubbles in the Milky Way Project catalog, and find that 10%-30% of the objects in this catalog are non-bubble interlopers. Relative to these interlopers, high-reliability bubbles are more confined to the mid-plane, and display a stronger excess of young stellar objects along and within bubble rims. Furthermore, Brut is able to discover bubbles missed by previous searches—particularly bubbles near bright sources which have low contrast relative to their surroundings. Brut demonstrates the synergies that exist between citizen scientists, professional scientists, and machine learning techniques. In cases where ''untrained' citizens can identify patterns that machines cannot detect without training, machine learning algorithms like Brut can use the output of citizen science projects as input training sets, offering tremendous opportunities to speed the pace of scientific discovery. A hybrid model of machine learning combined with crowdsourced training data from citizen scientists can not only classify large quantities of data, but also address the weakness of each approach if deployed alone.

79 ASTRONOMY AND ASTROPHYSICS↗

Dependence of Convective Cloud Microphysical Properties on Environmental Conditions during the TRACER and ESCAPE Field Campaigns: A Synergistic Approach of Observations, Machine Learning and Parcel Models

The sensitivity of convective clouds to aerosols and their interactions with environment, combined with limited observational constraints in parameterizations, introduces significant uncertainties in atmospheric models. Here, this study investigates the dependence of convective cloud microphysical properties on environmental conditions using a synergistic approach that combines unique observations from the TRACER and ESCAPE field campaigns, machine learning techniques, and parcel model simulations with a super-droplet microphysics scheme. A random forest algorithm identifies in-situ vertical velocity (w), temperature (T), and surface fine-mode aerosol mass concentration as the three most important environmental conditions influencing cloud properties including liquid water content (LWC), number concentration for particles with D max < 50 μm (N c ,<50), 50 μm ≤ D max ≤ 3000 μm (N c,50–3000 ), and droplet effective diameter (D e ). Results show that LWC, N c,<50 , and N c,50–3000 significantly increase with w in updrafts. Across w bins, as T decreases, LWC, D e , and N c,50–3000 increase, while N c,<50 decreases, which are closely linked to the distance above cloud bases. Warmer cloud bases yield higher LWC, greater N c,50–3000 , and smaller N c,<50 , while polluted environments produce greater N c,<50 . Parcel model simulations successfully replicate these observed dependencies. The simulation results indicate that warmer cloud bases enhance condensation generating larger droplets, and differences in droplet sizes are then amplified through collision-coalescence, resulting in a greater N c,50–3000 . Polluted conditions result in a greater N c,<50 primarily due to enhanced cloud condensation nuclei activation despite increased collision-coalescence rates compared to pristine conditions. This study provides observed quantitative patterns characterizing cloud microphysical properties as a function of key environmental parameters, offering valuable constraints for improving physics parameterizations and numerical models.

54 ENVIRONMENTAL SCIENCES↗

Objective Phenotyping of Root System Architecture Using Image Augmentation and Machine Learning in Alfalfa (Medicago sativa L.)

Active breeding programs specifically for root system architecture (RSA) phenotypes remain rare; however, breeding for branch and taproot types in the perennial crop alfalfa is ongoing. Phenotyping in this and other crops for active RSA breeding has mostly used visual scoring of specific traits or subjective classification into different root types. While image-based methods have been developed, translation to applied breeding is limited. This research is aimed at developing and comparing image-based RSA phenotyping methods using machine and deep learning algorithms for objective classification of 617 root images from mature alfalfa plants collected from the field to support the ongoing breeding efforts. Our results show that unsupervised machine learning tends to incorrectly classify roots into a normal distribution with most lines predicted as the intermediate root type. Encouragingly, random forest and TensorFlow-based neural networks can classify the root types into branch-type, taproot-type, and an intermediate taproot-branch type with 86% accuracy. With image augmentation, the prediction accuracy was improved to 97%. Coupling the predicted root type with its prediction probability will give breeders a confidence level for better decisions to advance the best and exclude the worst lines from their breeding program. This machine and deep learning approach enables accurate classification of the RSA phenotypes for genomic breeding of climate-resilient alfalfa.

59 BASIC BIOLOGICAL SCIENCES↗

A Novel Machine Learning Approach to Disentangle Multitemperature Regions in Galaxy Clusters

The hot intracluster medium (ICM) surrounding the heart of galaxy clusters is a complex medium that comprises various emitting components. Although previous studies of nearby galaxy clusters, such as the Perseus, the Coma, or the Virgo cluster, have demonstrated the need for multiple thermal components when spectroscopically fitting the ICM’s X-ray emission, no systematic methodology for calculating the number of underlying components currently exists. In turn, underestimating or overestimating the number of components can cause systematic errors in the emission parameter estimations. In this paper, we present a novel approach to determining the number of components using an amalgam of machine learning techniques. Synthetic spectra containing a various number of underlying thermal components were created using well-established tools available from the Chandra X-ray Observatory. The dimensions of the training set was initially reduced using principal component analysis and then categorized based on the number of underlying components using a random forest classifier. Our trained and tested algorithm was subsequently applied to Chandra X-ray observations of the Perseus cluster. Our results demonstrate that machine learning techniques can efficiently and reliably estimate the number of underlying thermal components in the spectra of galaxy clusters, regardless of the thermal model (MEKAL versus APEC). We also confirm that the core of the Perseus cluster contains a mix of differing underlying thermal components. We emphasize that although this methodology was trained and applied on Chandra X-ray observations, it is readily portable to other current (e.g., XMM-Newton, eROSITA) and upcoming (e.g., Athena, Lynx, XRISM) X-ray telescopes. The code is publicly available at https://github.com/XtraAstronomy/Pumpkin.

79 ASTRONOMY AND ASTROPHYSICS↗

Evaluation of Drilling Performance at The Geysers with Machine Learning Methods Using Geologic Data

A recent well, GDC-36, was drilled in The Geysers Geothermal Field served in a Department of Energy-industry to demonstrate improved drilling performance with polycrystalline diamond compact (PDC) bits. Both PDC and roller cone drill bits were used to drill this well. Key challenges encountered during drilling included lost circulation in the mud-drilled section, and bit damage interfacial severity in the deeper, air-drilled section. The objective of this study is to evaluate the drilling performance in relation to the local geological characteristics using machine learning methods. By applying K-clustering to the sonic log data, we were able to identify areas correlated with measured lost circulation. Also, the boundaries defined by clustering of the mineralogical and lithological data from the mud logs correlate well with interfacial severity during drilling. A random forest model was employed to build correlation between drilling data and rock strength. The confined compressive strength (CCS) of the rock in the training of the machine learning model was inferred from the dipole sonic log. The R-squared of the testing data is 0.78, and the RMSE (Root Mean Squared Error) is 0.06. The trained model was used to forecast rock strength for the section where sonic log data are not available. CCS could also be inferred from mud logs provided the relationship between mineralogy and rock strength is established through core testing data.

15 GEOTHERMAL ENERGY↗

Machine Learning–Augmented Laser-Induced Breakdown Spectroscopy for Spectral Discrimination of Iron Oxalates

Enhanced characterization and phase identification of post-PUREX Pu Oxalates (PuOXA) are pivotal for nonproliferation and pre-detonation nuclear forensics. Despite significant advances in the characterization of PuO 2 samples, little is known about the impact of both the chemical structure and oxidation states of PuOXA (i.e., Pu(III) and Pu(IV)) have on optical emission signatures. Here, we demonstrate the analytical capabilities of laser-induced breakdown spectroscopy (LIBS) applied to Fe(II) and Fe(III) oxalate samples as surrogates for PuOXA, highlighting the discriminating features in the LIBS emission spectra arising from differences in the oxidation states within mixed FeOXA samples. We report the enhancement of spectral feature selection using Principal Component Analysis (PCA), which enables the analytical superiority of machine learning algorithms such as Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), and Random Forest Regression (RFR) over conventional univariate techniques for phase discrimination and chemometric analysis. Cluster analysis revealed how both matrix effects and laser ablation influence cluster separability by introducing spectral artifacts that misdirect the maximization of variance. PCA-selected emission lines were used in the regression models, demonstrating that both univariate and multivariate linear regression models (i.e., PLSR and SVR) can achieve acceptable performance, with machine learning models outperforming conventional calibration regressions. Furthermore, the application of non-linearly activated PCA-selected emission lines illustrates how simplifying the data while retaining captured variance enables the use of less complex and more computationally efficient models. Furthermore, this is particularly evident in the underperformance of RFR, which suffers from increased computational costs and overfitting owing to its high complexity.

Oxalates↗

How Should Machine Learning Be Successfully Used for Wind Speed Vertical Extrapolation?

An accurate characterization of the wind resource available at hub-height is required for an efficient and bankable wind farm project. However, direct measurement of wind speed at the constantly increasing height of the hub of commercial wind turbines is oftentimes challenging and expensive, so that it is common practice to vertically extrapolate the wind resource from lower and more easily accessible levels. Conventional techniques for wind speed vertical extrapolation include the use of a power law and a logarithmic profile. While simple, the limits in accuracy of these methods have been shown in various studies. Recently, machine learning has been proposed as a new method to vertically extrapolate winds. All the published studies on the topic assess the performance of machine learning techniques in vertically extrapolating the wind resource at the same location where the algorithm has been trained. However, in real-world applications, the wind resource is measured at the instrument location, but it then needs to be extrapolated at hub height at the location of the wind turbines within the find farm. To be able to fully recommend the use of machine learning techniques over the simple power law and logarithmic law, the spatial variability of the performance improvements of the machine learning approaches needs to be assessed. Here, we propose a round-robin validation of a machine learning-based method for wind speed extrapolation. We use 20 months of observations at four locations spanning a 100 km wide region at the Southern Great Plains (SGP) atmospheric observatory, in north-central Oklahoma. At each location, we train a random forest to predict 30-min average wind speed at 143 m AGL. We use as input features lidar wind speed at 65 m AGL, time of day, sonic anemometer wind speed at 4 m AGL, turbulent kinetic energy, and Obukhov length. First, we perform a same-site comparison of the performance of the proposed random forest against the conventional techniques for wind speed extrapolation (namely power law and logarithmic profile, with widely accepted stability corrections). We find that the random forest outperforms the power law in vertically extrapolating wind speed in all the considered stability regimes, with a 33% reduction in MAE for stable conditions, and a 31% reduction in unstable conditions. Similar results are found when comparing predictions of extrapolated winds from the logarithmic profile and the random forest with the observed values. Next, we propose a round-robin validation, to use the random forest trained at each site to extrapolate wind speed at the remaining three sites. We find that the performance of the random forest approach degrades when the algorithm is tested at a site different than the training one. However, even under those circumstances, the machine learning-based approach still outperforms the conventional techniques for wind speed extrapolation, with, on average, a reduction in mean absolute error between 15 and 20% over the conventional methods, with the largest benefits obtained under stable conditions.

Monte Carlo↗

Six Machine-Learning Methods for Predicting Hospital-Stay Duration for Patients with Sepsis: A Comparative Study

Sepsis is a life-threatening medical condition that, if not treated promptly, can result in tissue damage, organ failure, and death. According to the Centers for Disease Control, about 270,000 individuals die of sepsis in the US each year. Further, sepsis expenditures accounted for 13% of total US hospital costs in 2013, totaling more than $24 billion. Our project objectives were to determine if Machine Learning algorithms could reliably predict hospital stay duration for patients with sepsis. The data set we used has been de-identified and is freely available through the BupaR package. The data includes 1050 cases, 15214 events, and 16 types of actions related to sepsis patient care. First, we used process mining to determine how long each patient was in the hospital. Using BupaR’s functions, we created several process model graphs. These process models depict the movement of patients at a hospital and provide duration data for each patent case. Second, we identified outlier data and created two dataset versions: one with and one without outliers. We then applied the following analysis methods: Linear Regression, Random Forest, K-Nearest Neighbors, Neural Networks, XGBoost, and lightGBM. We compared the model validations for the six machine learning models using the same data-splitting method. We found that the XGBoost model had the best prediction accuracy of 73.9 percent for cases with outliers, and 79 percent for cases without outliers. We also found that the lightGBM model had the lowest mean absolute error between prediction and actual duration in days with 3.66 days for the case with outliers, and 2.4 days for the case without outliers. These two models outperformed the other four models. This work will be enhanced in the future by exploring new prediction algorithms and comparing them with the results of this study.

Chen, Lingtao↗

Comparison of Machine Learning Algorithms for Natural Gas Identification with Mixed Potential Electrochemical Sensor Arrays

Mixed-potential electrochemical sensor arrays consisting of indium tin oxide (ITO), La 0.87 Sr 0.13 CrO 3 , Au, and Pt electrodes can detect the leaks from natural gas infrastructure. Algorithms are needed to correctly identify natural gas sources from background natural and anthropogenic sources such as wetlands or agriculture. We report for the first time a comparison of several machine learning methods for mixture identification in the context of natural gas emissions monitoring by mixed potential sensor arrays. Random Forest, Artificial Neural Network, and Nearest Neighbor methods successfully classified air mixtures containing only CH 4 , two types of natural gas simulants, and CH 4 +NH 3 with >98% identification accuracy. The model complexity of these methods were optimized and the degree of robustness against overfitting was determined. Finally, these methods are benchmarked on both desktop PC and single-board computer hardware to simulate their application in a portable internet-of-things sensor package. The combined results show that the random forest method is the preferred method for mixture identification with its high accuracy (>98%), robustness against overfitting with increasing model complexity, and had less than 10 ms training time and less than 0.1 ms inference time on single-board computer hardware.

03 NATURAL GAS↗

Uncertainty-aware mixed-variable machine learning for materials design

Abstract Data-driven design shows the promise of accelerating materials discovery but is challenging due to the prohibitive cost of searching the vast design space of chemistry, structure, and synthesis methods. Bayesian optimization (BO) employs uncertainty-aware machine learning models to select promising designs to evaluate, hence reducing the cost. However, BO with mixed numerical and categorical variables, which is of particular interest in materials design, has not been well studied. In this work, we survey frequentist and Bayesian approaches to uncertainty quantification of machine learning with mixed variables. We then conduct a systematic comparative study of their performances in BO using a popular representative model from each group, the random forest-based Lolo model (frequentist) and the latent variable Gaussian process model (Bayesian). We examine the efficacy of the two models in the optimization of mathematical functions, as well as properties of structural and functional materials, where we observe performance differences as related to problem dimensionality and complexity. By investigating the machine learning models’ predictive and uncertainty estimation capabilities, we provide interpretations of the observed performance differences. Our results provide practical guidance on choosing between frequentist and Bayesian uncertainty-aware machine learning models for mixed-variable BO in materials design.

36 MATERIALS SCIENCE↗

Failure Analysis–Informed Risk Assessment Framework for Geological Carbon Storage Using Numerical Simulation and Machine Learning

Geological carbon storage (GCS) is recognized as a critical technology for achieving large-scale reductions in anthropogenic carbon dioxide (CO 2 ) emissions. Ensuring long-term containment and safety requires robust risk assessment frameworks that account for geological uncertainty and identify potential failure scenarios. Among various indicators, the area of review (AoR) serves as a key metric for evaluating storage performance, regulatory compliance, and monitoring design, as it delineates the spatial extent impacted by pressure buildup and plume migration. However, conventional AoR-based risk assessments typically perturb parameters within narrow uncertainty bounds, potentially overlooking rare but high-impact events arising from extreme geological conditions. In this study, we present a failure analysis–informed risk assessment framework for large-scale GCS projects to improve site prescreening and monitoring design. A suite of 300 numerical simulations was generated using stochastic geological models that vary five key parameters: net-to-gross ratio, anisotropy azimuth, porosity multiplier, permeability multiplier, and vertical-to-horizontal permeability ratio. Among these, 200 realizations represent normal geological uncertainty, while 100 additional cases explore extreme yet plausible conditions for failure-case analysis. The AoR was simulated and computed from pressure and CO 2 saturation fields, where the baseline AoR boundary, representing the extent predicted under typical geological uncertainty, was defined as the union of 200 normal-range simulations, and failure was identified when extreme-range cases exceeded this baseline. Results show that incorporating broader parameter uncertainty produces significantly larger AoR extents, underscoring the potential underestimation of risk under conventional uncertainty ranges. Furthermore, spatial probability maps derived from failure-induced AoR exceedance identify regions requiring enhanced monitoring attention. Various machine learning (ML)–based classifiers were developed to predict failure occurrence from geological parameters, with the random forest model achieving the highest performance (F1-score of 0.986). Consistent findings from correlation coefficient, feature importance, and Sobol sensitivity analyses reveal that low net-to-gross ratios and permeability multipliers are the dominant risk drivers, reflecting reduced reservoir connectivity and limited pressure dissipation. Altogether, these results provide a novel framework for risk-informed site prescreening and monitoring design that explicitly considers rare but high-impact geological scenarios in GCS projects.

25 ENERGY STORAGE↗