Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “large data sets”

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 379 records · Page 21
All sourcesNASA NTRSDOE OSTI
NASA NTRS1995

Comparison of Retracking Algorithms Using Airborne Radar and Laser Altimeter Measurements of the Greenland Ice Sheet

This paper compares four continental ice sheet radar altimeter retracking algorithms using airborne radar and laser altimeter data taken over the Greenland ice sheet in 1991. The refurbished Advanced Application Flight Experiment (AAFE) airborne radar altimeter has a large range window and stores the entire return waveform during flight. Once the return waveforms are retracked, or post-processed to obtain the most accurate altitude measurement possible, they are compared with the high-precision Airborne Oceanographic Lidar (AOL) altimeter measurements. The AAFE waveforms show evidence of varying degrees of both surface and volume scattering from different regions of the Greenland ice sheet. The AOL laser altimeter, however, obtains a return only from the surface of the ice sheet. Retracking altimeter waveforms with a surface scattering model results in a good correlation with the laser measurements in the wet and dry-snow zones, but in the percolation region of the ice sheet, the deviation between the two data sets is large due to the effects of subsurface and volume scattering. The Martin et al model results in a lower bias than the surface scattering model, but still shows an increase in the noise level in the percolation zone. Using an Offset Center of Gravity algorithm to retrack altimeter waveforms results in measurements that are only slightly affected by subsurface and volume scattering and, despite a higher bias, this algorithm works well in all regions of the ice sheet. A cubic spline provides retracked altitudes that agree with AOL measurements over all regions of Greenland. This method is not sensitive to changes in the scattering mechanisms of the ice sheet and it has the lowest noise level and bias of all the retracking methods presented.

Ferraro, Ellen J.↗
DOE OSTI2023

Evaluating skill in predicting the Interdecadal Pacific Oscillation in initialized decadal climate prediction hindcasts in E3SMv1 and CESM1 using two different initialization methods and a small set of start years

Abstract It is a daunting challenge to conduct initialized hindcasts with enough ensemble members and associated start years to form a drifted climatology from which to compute the anomalies necessary to quantify the skill of the hindcasts when compared to observations. This limits the ability to experiment with case studies and other applications where only a few initial years are needed. Here we run a set of hindcasts with CESM1 and E3SMv1 using two different initialization methods for a limited set of start years and use the respective uninitialized free-running historical simulations to form the model climatologies. Since the drifts from the observed initial states in the hindcasts toward the uninitialized model state are large and rapid, after a few years the drifted initialized models approach the uninitialized model climatological errors. Therefore, hindcasts from the limited start years can use the uninitialized climatology to represent the drifted model states after about lead year 3, providing a means to compute forecast anomalies in the absence of a large hindcast sample. There is comparable skill for predicting spatial patterns of multi-year Pacific sea surface temperature anomalies in the domain of the Interdecadal Pacific Oscillation using this method compared to the conventional methodology with a large hindcast data set, though there is a model dependence to the drifts in the two initialization methods.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Carbonate clumped isotope analysis (Δ 47 ) of 21 carbonate standards determined via gas‐source isotope‐ratio mass spectrometry on four instrumental configurations using carbonate‐based standardization and multiyear data sets

Rationale Clumped isotope geochemistry examines the pairing or clumping of heavy isotopes in molecules and provides information about the thermodynamic and kinetic controls on their formation. The first clumped isotope measurements of carbonate minerals were first published 15 years ago, and since then, interlaboratory offsets have been observed, and laboratory and community practices for measurement, data analysis, and instrumentation have evolved. Here we briefly review historical and recent developments for measurements, share Tripati Lab practices for four different instrument configurations, test a recently published proposal for carbonate‐based standardization on multiple instruments using multi‐year data sets, and report values for 21 different carbonate standards that allow for recalculations of previously published data sets. Methods We examine data from 4628 standard measurements on Thermo MAT 253 and Nu Perspective IS mass spectrometers, using a common acid bath (90°C) and small‐sample (70°C) individual reaction vessels. Each configuration was investigated by treating some standards as anchors (working standards) and the remainder as unknowns (consistency standards). Results We show that different acid digestion systems and mass spectrometer models yield indistinguishable results when instrument drift is well characterized. For linearity correction, mixed gas‐and‐carbonate standardization or carbonate‐only standardization yields similar results. No difference is observed in the use of three or eight working standards for the construction of transfer functions. Conclusions We show that all configurations yield similar results if instrument drift is robustly characterized and validate a recent proposal for carbonate‐based standardization using large multiyear data sets. Δ 47 values are reported for 21 carbonate standards on both the absolute reference frame (ARF; also refered to as the Carbon Dioxide Equilibrated Scale or CDES) and the new InterCarb‐Carbon Dioxide Equilibrium Scale (I‐CDES) reference frame, facilitating intercomparison of data from a diversity of labs and instrument configurations and restandardization of a broad range of sample sets between 2006, when the first carbonate measurements were published, and the present.

58 GEOSCIENCES↗
DOE OSTI2022

Baryon acoustic oscillations in thin redshift shells from BOSS DR12 and eBOSS DR16 galaxies

ABSTRACT In an age of large astronomical data sets and severe cosmological tensions, the case for model independent analyses is compelling. We present a set of 14 baryon acoustic oscillations measurements in thin redshift shells with $3\,\mathrm{ per} \,\mathrm{ cent}$ precision that were obtained by analysing BOSS DR12 and eBOSS DR16 galaxies in the redshift range 0.32 < z < 0.66. Thanks to the use of thin shells, the analysis is carried out using just redshifts and angles so that the fiducial model is only introduced when considering the mock catalogues, necessary for the covariance matrix estimation and the pipeline validation. We compare our measurements, with and without supernova data, to the corresponding constraints from Planck 2018, finding good compatibility. A Monte Python module for this likelihood is available at github.com/ranier137/angularBAO.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1986

Plate motions and deformations from geologic and geodetic data

Research effort on behalf of the Crustal Dynamics Project focused on the development of methodologies suitable for the analysis of space-geodetic data sets for the estimation of crustal motions, in conjunction with results derived from land-based geodetic data, neo-tectonic studies, and other geophysical data. These methodologies were used to provide estimates of both global plate motions and intraplate deformation in the western U.S. Results from the satellite ranging experiment for the rate of change of the baseline length between San Diego and Quincy, California indicated that relative motion between the North American and Pacific plates over the course of the observing period during 1972 to 1982 were consistent with estimates calculated from geologic data averaged over the past few million years. This result, when combined with other kinematic constraints on western U.S. deformation derived from land-based geodesy, neo-tectonic studies, and other geophysical data, places limits on the possible extension of the Basin and Range province, and implies significant deformation is occurring west of the San Andreas fault. A new methodology was developed to analyze vector-position space-geodetic data to provide estimates of relative vector motions of the observing sites. The algorithm is suitable for the reduction of large, inhomogeneous data sets, and takes into account the full position covariances, errors due to poorly resolved Earth orientation parameters and vertical positions, and reduces baises due to inhomogeneous sampling of the data. This methodology was applied to the problem of estimating the rate-scaling parameter of a global plate tectonic model using satellite laser ranging observations over a five-year interval. The results indicate that the mean rate of global plate motions for that interval are consistent with those averaged over several million years, and are not consistent with quiescent or greatly accelerated plate motions. This methodology was also used to provide constraints on deformation in the western U.S. using very long baseline interferometry observations over a two-year period.

Jordan, T. H.↗
DOE OSTI2022

Practical galaxy morphology tools from deep supervised representation learning

Astronomers have typically set out to solve supervised machine learning problems by creating their own representations from scratch. We show that deep learning models trained to answer every Galaxy Zoo DECaLS question learn meaningful semantic representations of galaxies that are useful for new tasks on which the models were never trained. We exploit these representations to outperform several recent approaches at practical tasks crucial for investigating large galaxy samples. The first task is identifying galaxies of similar morphology to a query galaxy. Given a single galaxy assigned a free text tag by humans (e.g. ‘#diffuse’), we can find galaxies matching that tag for most tags. The second task is identifying the most interesting anomalies to a particular researcher. Our approach is 100 per cent accurate at identifying the most interesting 100 anomalies (as judged by Galaxy Zoo 2 volunteers). The third task is adapting a model to solve a new task using only a small number of newly labelled galaxies. Models fine-tuned from our representation are better able to identify ring galaxies than models fine-tuned from terrestrial images (ImageNet) or trained from scratch. We solve each task with very few new labels; either one (for the similarity search) or several hundred (for anomaly detection or fine-tuning). This challenges the longstanding view that deep supervised methods require new large labelled data sets for practical use in astronomy. To help the community benefit from our pretrained models, we release our fine-tuning code zoobot. Zoobot is accessible to researchers with no prior experience in deep learning.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

L-PBF High-Throughput Data Pipeline Approach for Multi-modal Integration

Abstract Metal-based additive manufacturing requires active monitoring solutions for assessing part quality. Multiple sensors and data streams, however, generate large heterogeneous data sets that are impractical for manual assessment and characterization. In this work, an automated pipeline is developed that enables feature extraction from high-speed camera video and multi-modal data analysis. The framework removes the need for manual assessment through the utilization of deep learning techniques and training models in a weakly supervised paradigm. We demonstrate this pipeline’s capability over 700,000 high-speed camera frames. The pipeline successfully extracts melt pool and spatter geometries and links them to corresponding pyrometry, radiography, and processparameter information. 715 individual prints are examined to reveal melt pool areas that exceeds 0.07 mm 2 and pyrometry signal over a threshold (375 pyrometry units) were more likely to have defects. These automated processes enable massive throughput of characterization techniques.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Audacity of huge: overcoming challenges of data scarcity and data quality for machine learning in computational materials discovery

Machine learning (ML)-accelerated discovery requires large amounts of high-fidelity data to reveal predictive structure–property relationships. For many properties of interest in materials discovery, the challenging nature and high cost of data generation has resulted in a data landscape that is both scarcely populated and of dubious quality. Data-driven techniques starting to overcome these limitations include the use of consensus across functionals in density functional theory, the development of new functionals or accelerated electronic structure theories, and the detection of where computationally demanding methods are most necessary. When properties cannot be reliably simulated, large experimental data sets can be used to train ML models. In the absence of manual curation, increasingly sophisticated natural language processing and automated image analysis are making it possible to learn structure–property relationships from the literature. Finally, models trained on these data sets will improve as they incorporate community feedback.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Classification of animal sounds in a hyperdiverse rainforest using convolutional neural networks with data augmentation

To protect tropical forest biodiversity, we need to be able to detect it reliably, cheaply, and at scale. Automated detection of sound producing animals from passively recorded soundscapes via machine-learning approaches is a promising technique towards this goal, but it is constrained by the necessity of large training data sets. Using soundscapes from a tropical forest in Borneo and a Convolutional Neural Network model (CNN), we investigate i) the minimum viable training data set size for accurate prediction of call types (‘sonotypes’), and ii) the extent to which data augmentation and transfer learning can overcome the issue of small and imbalanced training data sets. We found that even relatively high sample sizes (>80 per sonotype) lead to mediocre accuracy, which however improved significantly with data augmentation and transfer learning, including at extremely small sample sizes (3 per sonotype), regardless of taxonomic group or call characteristics. Neither transfer learning nor data augmentation alone achieved high accuracy. Our results suggest that transfer learning and data augmentation could make the use of CNNs to classify species’ vocalizations feasible even for small soundscape-based projects with many rare species. Retraining our open-source model requires only basic programming skills which makes it possible for individual conservation initiatives to match their local context, in order to enable more evidence-informed management of biodiversity.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Source term estimation using noble gas and aerosol samples

Algorithms that estimate the location, time, and magnitude of a point-source atmospheric release using remotely sampled air concentrations typically use data for a single chemical or radioactive isotope. Here, a Bayesian algorithm is presented that uses data from multiple radioactive isotopes that are all released in the same short-duration event. Data from noble gas and aerosol samplers can be used simultaneously in the model. Application to a large synthetic data set using four isotopes shows the new algorithm generally gives more accurate location and time estimates than a comparable model using a single isotope.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Ligand-Based Compound Activity Prediction via Few-Shot Learning

Predicting the activities of new compounds against biophysical or phenotypic assays based on the known activities of one or a few existing compounds is a common goal in early stage drug discovery. This problem can be cast as a “few-shot learning” challenge, and prior studies have developed few-shot learning methods to classify compounds as active versus inactive. However, the ability to go beyond classification and rank compounds by expected affinity is more valuable. We describe Few-Shot Compound Activity Prediction (FS-CAP), a novel neural architecture trained on a large bioactivity data set to predict compound activities against an assay outside the training set, based on only the activities of a few known compounds against the same assay. Our model aggregates encodings generated from the known compounds and their activities to capture assay information and uses a separate encoder for the new compound whose activity is to be predicted. The new method provides encouraging results relative to traditional chemical-similarity-based techniques as well as other state-of-the-art few-shot learning methods in tests on a variety of ligand-based drug discovery settings and data sets.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Stable Methane Isotopologues From Northern Lakes Suggest That Ebullition Is Dominated by Sub‐Lake Scale Processes

Abstract Stable isotopes have emerged as popular study targets when investigating emission of methane (CH 4 ) from lakes. Yet little is known on how isotopic patterns conform to variations in emission magnitudes—a highly relevant question. Here, we present a large multiyear data set on stable isotopes of CH 4 ebullition (bubbling) from three small adjacent subarctic lakes. The δ 13 C‐CH 4 and δD‐CH 4 range from −78.4‰ to −53.1‰ and from −369.8‰ to −218.8‰, respectively, and vary greatly among the lakes. The signatures suggest dominant hydrogenotrophic methanogenesis, particularly in the deep zones, but there are also signals of seemingly acetoclastic production in some high fluxing shallow areas, possibly fueled by in situ vegetation, but in‐sediment anaerobic CH 4 oxidation cannot be ruled out as an alternative cause. The observed patterns, however, are not consistent across the lakes. Neither do they correspond to the spatiotemporal variations in the measured bubble CH 4 fluxes. Patterns of acetoclastic and hydrogenotrophic production plus oxidation demonstrate that gains and losses of sediment CH 4 are dominated by sub‐lake scale processes. The δD‐CH 4 in the bubbles was significantly different depending on measurement month, likely due to evaporation effects. On a larger scale, our isotopic data, combined with those from other lakes, show a significant difference in bubble δD‐CH 4 between postglacial and thermokarst lakes, an important result for emission inventories. Although this characteristic theoretically assists in source partitioning studies, most hypothetical future shifts in δD‐CH 4 due to high‐latitude lake area or production pathway are too small to lead to atmospheric changes detectable with current technology.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

A Statistical Framework for Evaluating Rain Microphysics in Model Simulations and Disdrometer Observations

Abstract Statistical analyses of a large disdrometer data set and a diverse set of model simulations for convection using the Regional Atmospheric Modeling System were conducted, with the mutual goal of providing insights into precipitation formation and microphysical processes. We demonstrate that a two‐moment bulk microphysical model successfully captures the dominant observed modes of variability in rainfall related to rainfall intensity and raindrop size distributions. The model reproduced the general distribution of observed precipitation groups (PGs) derived from Principal Component Analysis. The multi‐variable analysis also uncovered some shortcomings in the model as well as limitations of the disdrometer data. The model solutions were constrained in their predicted drop size distributions (DSDs) due to the fixed DSD parameters assumed in a two‐moment microphysics scheme. A case study from the Mid‐latitude Continental Clouds and Convection Experiment field project demonstrated how model results can be used to contextualize the disdrometer observations which are limited in sample size, spatial coherence, and detection of small drops and low drop concentrations. The case study also showed that the spatial patterns of the statistically derived PGs revealed by the model are consistent with the hypothesized microphysical processes that determine surface rain DSDs. This work demonstrates how leveraging the strengths of observations and models together can improve our understanding and representation of rain microphysical processes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Inferring Plant Acclimation and Improving Model Generalizability With Differentiable Physics‐Informed Machine Learning of Photosynthesis

Net photosynthesis (A N ) is a key component of the global carbon cycle influencing climate feedback over decadal scales. Although plant acclimation to environmental changes can modify A N , traditional vegetation models in Earth system models (ESMs) often rely on plant functional type (PFT)-specific parameterizations or simplified acclimation assumptions limiting generalizability across time, space, and PFTs. In this study, we developed a differentiable photosynthesis model to learn the environmental dependencies of V c,max25 (maximum carboxylation rate at 25°C, representing photosynthetic capacity), as this genre of hybrid physics-informed machine learning can seamlessly train neural networks and process-based equations together. Compared to PFT-specific parameterization of V c,max25 , learning the environment dependencies of key photosynthetic parameters improved model spatiotemporal generalizability. Applying environmental acclimation to V c,max25 led to substantial variations in global mean A N indicating the need to address acclimation in ESMs. The model effectively captured multivariate observations (V c,max25 , A N , and stomatal conductance (g s )) simultaneously with multivariate constraints, improving generalization across space and PFTs. It also learned sensible acclimation relationships of V c,max25 to different environmental conditions. The model explained more than 54%, 57%, and 62% of the variance of A N , g s , and V c,max25 , respectively, presenting a first global-scale spatial test benchmark of A N and g s . These results highlight the potential for differentiable modeling to enhance process-based modules in ESMs and effectively leverage information from large, multivariate data sets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Discovering the Multisectoral Impacts of Global Energy Sector Outcomes Through Multiple Ensemble Aggregation Measures

Understanding complex human-Earth system interactions often involves analyzing large scenario ensembles that encompass a wide range of plausible futures. These ensembles often require aggregation to summarize information based on specific criteria or conditions. However, previous research using global change scenario ensembles has largely overlooked how the choice of aggregation method influences the interpretation of results. To address this gap, we leverage a large ensemble data set designed to capture broad energy system dynamics generated using the Global Change Analysis Model. We first explore how energy-related uncertainties are propagated to both global and regional water-energy-food sectors. We then conduct a rank correlation analysis across seven ensemble aggregation measures and demonstrate the need to consider multiple measures in global change scenarios. Our results suggest that global water and food sector outcomes in the 21st century vary widely depending on different scenario assumptions. The global energy productivity is projected to improve by the end of the century across all scenarios. Moreover, regions facing water scarcity challenges in 2100 do not always overlap with those facing extreme energy and food sector outcomes. Although rank correlations across seven aggregation measures are relatively stable across sectors, we identify cases where relying on a single measure leads to losing critical information in the full ensemble. Reliance on a single aggregation measure can distort the interpretation of global change scenario outcomes. Instead, adopting multiple ensemble aggregation measures provides a more holistic understanding of global change scenario ensembles.

Kim, Gijoo↗
DOE OSTI2025

A Greening Future Elevates Flash Drought Risk in Northern Mid‐to‐High Latitudes

Flash droughts have become a growing concern, as they can emerge rapidly and increase the risk of crop failure. Although past studies have investigated the meteorological drivers and future changes of flash drought, why flash drought is more frequent over humid and vegetated regions remains underexplored. This study delves further into the mechanism by which vegetation regulates flash drought and its future change using observations from multiple data sets and large ensemble simulations from three Earth system models. On an interannual timescale, both observations and simulations show robust increases in flash drought frequency and a higher flash-to-sub-seasonal drought ratio during spring or antecedent conditions with dense vegetation, supporting the important role of vegetation in flash drought occurrence, especially in the northern mid-to-high latitudes. In the latter regions, the large ensemble simulations show robust increases in flash drought (e.g., 67% and 46% increases in Eastern U.S. and North Asia in 2050–2100 relative to 1950–2000 under the high emission scenario), where the growing season is lengthening. Although greening might suggest reduced drought stress, it drives precipitation-soil moisture-evapotranspiration decoupling by increasing evapotranspiration partitioning to transpiration. As transpiration can access deep soil water through the plant root system, its increased portion can weaken the constraints of concurrent precipitation on evapotranspiration, thus accelerating soil moisture depletion under high evaporative demand, driving a slow-to-rapid drought transition. How vegetation regulates flash drought by regulating surface moisture budget is supported by observations and simulations. Although warming supports early planting, agriculture may increasingly be threatened by surging flash drought risk.

Drought↗
DOE OSTI2026

A Mountain Glacier Perspective on the Bipolar Seesaw

A global record of mountain glacier terminations during the last deglaciation (∼19–11 ka) dated by a large, uncurated data set of cosmogenic-nuclide exposure ages highlights a statistically significant asynchrony in termination ages between the Northern and Southern Hemispheres. This interhemispheric offset in the timing of glacier terminations is consistent with previously correlated ice core records that show a systematic interhemispheric lag in the timing of abrupt climate events, with the Southern Hemisphere leading the Northern Hemisphere by ∼300–3,500 years. Our analysis (a) aggregates cosmogenic-nuclide exposure ages from a global data set of moraines to discern climatically driven peaks in moraine emplacement events, and (b) utilizes a Monte Carlo simulation based on a null hypothesis that moraine emplacement is interhemispherically synchronous to estimate the statistical significance of the observed offset. The observed lag of Northern Hemisphere emplacement events compared to the Southern Hemisphere is statistically significant and is consistent with the “bipolar seesaw” pattern observed in ice core records.

58 GEOSCIENCES↗
DOE OSTI2021

Predicting fault slip via transfer learning

Abstract Data-driven machine-learning for predicting instantaneous and future fault-slip in laboratory experiments has recently progressed markedly, primarily due to large training data sets. In Earth however, earthquake interevent times range from 10’s-100’s of years and geophysical data typically exist for only a portion of an earthquake cycle. Sparse data presents a serious challenge to training machine learning models for predicting fault slip in Earth. Here we describe a transfer learning approach using numerical simulations to train a convolutional encoder-decoder that predicts fault-slip behavior in laboratory experiments. The model learns a mapping between acoustic emission and fault friction histories from numerical simulations, and generalizes to produce accurate predictions of laboratory fault friction. Notably, the predictions improve by further training the model latent space using only a portion of data from a single laboratory earthquake-cycle. The transfer learning results elucidate the potential of using models trained on numerical simulations and fine-tuned with small geophysical data sets for potential applications to faults in Earth.

58 GEOSCIENCES↗