SEARCH · Search NASA

Results for “spatial prediction”

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.

SourceResearch connectionFrom yearThrough year

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

At least 253 records · Page 14

All sources NASA NTRS DOE OSTI

DOE OSTI2024

The influence of burn severity on dissolved organic carbon concentrations across a stream network differs based on seasonal wetness conditions

Large, high-severity wildfires in many regions across the globe have increased concerns about their impacts on carbon cycling in watersheds. Altered sources of carbon and changes in catchment hydrology after wildfire can lead to shifts in dissolved organic carbon (DOC) concentrations in streams, which can have negative impacts on aquatic ecosystem health and downstream drinking-water treatment. Despite its importance, post-fire DOC responses remain relatively unconstrained in the literature, and we lack critical knowledge of how burn severity, landscape elements, and climate interact to affect DOC concentrations. To improve our understanding of the impact of burn severity on DOC concentrations, we measured DOC at 129 sites across a stream network extending upstream, within, and downstream of a large, high-severity wildfire in Oregon, USA. We collected samples across the study sub-basin during four distinct seasonal wetness conditions. We used our high-spatial-resolution data to develop spatial stream network (SSN) models to predict DOC across the stream network and to improve our understanding of the controls on DOC concentrations. Spatially, we found no obvious wildfire signal – instead, we observed a pattern of increasing DOC concentrations from the high-elevation headwaters to the sub-basin outlet, while the mainstem maintained consistently low DOC concentrations. This suggests that effects from large wildfires may be “averaged” out at higher stream orders and larger spatial scales. When we grouped DOC concentrations by burn severity group, we observed a significant decrease in the variability of DOC concentrations in the moderate and high burn severity sub-catchments. However, our SSN models were able to predict decreases in DOC concentrations with increases in burn severity across the stream network. Decreases in DOC concentrations were also highly variable across seasonal wetness conditions, with the greatest (–1.40 to –1.64 mg L –1 ) decrease occurring in the high-severity group during the wetting season. Additionally, our models indicated that in all seasons, baseflow index was more influential in predicting DOC concentrations than burn severity was, indicating that groundwater discharge can obscure the impacts of wildfire in a stream network. Overall, our results suggested that landscape characteristics can regulate the DOC response to wildfire. Moreover, our results also indicated that the seasonal timing of sampling can influence the observed response of DOC concentrations to wildfire.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

A cross-dimensional analysis of data-driven short-term load forecasting methods with large-scale smart meter data

Electricity load forecasting is essential to utility operation and power grid stability. A wide spectrum of data-driven methods, ranging from linear regression models to more recent deep learning models have been adopted to forecast electric load over the years. However, there still lacks a holistic evaluation of the applicability of conventional statistical and machine learning based algorithms with respect to different temporal and spatial scopes, computational requirements, and sensitivity of model-tuning. Enabled by a large-scale electricity load profile dataset of over 40,000 residential customers in a utility region, we conducted a cross-dimensional analysis of data-driven load forecasting methods. Three regression-based and seven deep learning algorithms with different model configurations were evaluated in terms of their overall and peak load prediction accuracy, and training burdens, across spatial aggregation levels ranging from the transformer, feeder, substation, to neighborhood. We found, first, the load forecasting accuracy is constrained by a predictability boundary, influenced by the forecasting horizon and spatial aggregation level. Specifically, RandomForest, XGBoost, TFT, TSMixer, and TiDE models achieved less than 10 % prediction error for up to 96-h ahead forecasting for district, substation, and feeder levels, while other models struggle at long-horizon predictions; Second, for winter and summer peak load dates, most models were able to predict the peak demand timing within ± 1 h, but the prediction percentage error varied by models, with TFT and TiDE models being the top performers; Third, models with similar prediction accuracy can differ in training burden by an order of magnitude. Therefore, choosing model configurations that balance prediction performance and computational resource is an important practical consideration for large-scale deployment of the machine learning based load forecasting. The outcome of this study can guide researchers and practitioners to choose the proper load forecasting algorithms based on their problem scope, required accuracy, and available resources. The predictability boundary can serve as a benchmark for electricity load forecasting problems with new algorithms and datasets.

Li, Han↗

NASA NTRS2021

Southern Idaho Health & Air Quality: Monitoring Atmospheric Mixing Heights Post-Wildfire Through the Use of NASA Earth Observations

Wildfire smoke has long-lasting impacts on public and environmental health. Currently, agencies that monitor smoke base their decisions on an analysis of how fires burn, the direction the smoke moves from the fire source, and unverified estimates of mixing height. Mixing heights describe the maximum altitude to which a smoke plume rises. Satellite imagery provides more continuous and accurate coverage of mixing heights than current in situ methods. Thus, the team developed a software tool that processes and extracts mixing height observations from Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) Vertical Feature Mask granules. The team partnered with the National Oceanic and Atmospheric Administration’s National Weather Service, the Bureau of Land Management’s National Interagency Fire Center, and the National Park Service’s Fire Management Program Center to analyze historic fire events in southern Idaho. To do so, they used Suomi National Polar-orbiting Partnership (Suomi NPP) Visible Infrared Imaging Radiometer Suite (VIIRS), and Terra and Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) to verify where a CALIPSO pass intersects thermal anomalies and smoke plumes. The software extracts features of relevance from the hdf file of each CALIPSO transect to locate layers of continuous aerosols. The maximum altitude at which the aerosol ends is recorded as the mixing height, along with a matching latitude and longitude. The satellite-derived values can be used to validate past mixing height predictions and evaluate the accuracy and systematic bias of different estimation methods. These results may allow agencies to make better comparisons and subsequent smoke pollution management, prevention, and public health decisions if the spatial and temporal differences between predictions and observations can be resolved.

DEVELOP Project Summary↗

DOE OSTI2024

SIGHT: Stacked Integration of Geospatial Hierarchical Typologies for Inferring Building Characteristics

Building characteristics are often absent in building stock datasets, particularly in regions most vulnerable to climate change and requiring effective disaster management strategies. Traditional machine learning approaches, while widely used to predict building attributes, typically neglect the spatial context of the data, leading to less accurate and reliable outcomes. To address these challenges, this paper introduces a novel algorithm, the Stacked Integration of Geospatial Hierarchical Typologies. This algorithm adapts a meta-learning framework to incorporate geospatial context into the predictive modeling process. We demonstrate the utility of the algorithm through two primary use cases: building use type classification and building height prediction. The algorithm consistently achieved or exceeded a 0.94 macro average F1 score across five geographically distinct countries for building use type classification. For building height prediction, it accurately predicted heights with a root mean square error of 3.01 in a comprehensive study using roughly 3.6 million buildings in Japan. These results underscore the benefits of integrating spatial hierarchies into machine learning models, enhancing both predictive accuracy and reliability in geospatial modeling. This work introduces a new algorithm to address the pervasive data sparsity issue in existing building stock datasets.

Adams, Daniel [ORNL] (ORCID:0000000196950577)↗

NASA NTRS1997

Response Classification Images in Vernier Acuity

Orientation selective and local sign mechanisms have been proposed as the basis for vernier acuity judgments. Linear image features contributing to discrimination can be determined for a two choice task by adding external noise to the images and then averaging the noises separately for the four types of stimulus/response trials. This method is applied to a vernier acuity task with different spatial separations to compare the predictions of the two theories. Three well-practiced observers were presented around 5000 trials of a vernier stimulus consisting of two dark horizontal lines (5 min by 0.3 min) within additive low-contrast white noise. Two spatial separations were tested, abutting and a 10 min horizontal separation. The task was to determine whether the target lines were aligned or vertically offset. The noises were averaged separately for the four stimulus/response trial types (e.g., stimulus = offset, response = aligned). The sum of the two 'not aligned' images was then subtracted from the sum of the 'aligned' images to obtain an overall image. Spatially smoothed images were quantized according to expected variability in the smoothed images to allow estimation of the statistical significance of image features. The response images from the 10 min separation condition are consistent with the local sign theory, having the appearance of two linear operators measuring vertical position with opposite sign. The images from the abutting stimulus have the same appearance with the two operators closer together. The image predicted by an oriented filter model is similar, but has its greatest weight in the abutting region, while the response images fall to nonsignificance there. The response correlation image method, previously demonstrated for letter discrimination, clarifies the features used in vernier acuity.

Ahumada, Albert J., Jr.↗

DOE OSTI2023

Optimizing Seasonal‐To‐Decadal Analog Forecasts With a Learned Spatially‐Weighted Mask

Abstract Seasonal‐to‐decadal climate prediction is crucial for decision‐making in a number of industries, but forecasts on these timescales have limited skill. Here, we develop a data‐driven method for selecting optimal analogs for seasonal‐to‐decadal analog forecasting. Using an interpretable neural network, we learn a spatially‐weighted mask that quantifies how important each grid point is for determining whether two climate states will evolve similarly. We show that analogs selected using this weighted mask provide more skillful forecasts than analogs that are selected using traditional spatially‐uniform methods. This method is tested on two prediction problems using the Max Planck Institute for Meteorology Grand Ensemble: multi‐year prediction of North Atlantic sea surface temperatures, and seasonal prediction of El Niño Southern Oscillation. This work demonstrates a methodical approach to selecting analogs that may be useful for improving seasonal‐to‐decadal forecasts and understanding their sources of skill.

Geology↗

DOE OSTI2025

Environmental Influences on Deep Convective Upscale Growth Rate in Central Argentina From a Convection‐Permitting Simulation

This study uses a convection‐permitting model simulation to describe the environmental conditions under which convective upscale growth occurs in central Argentina, particularly examining environmental parameters when deep convection initially forms that could differentiate the rate of initial upscale growth. Simulated mesoscale convective systems (MCSs) are separated into slow and rapid growth by the rate of spatial growth from convection initiation until reaching the MCS scale. A low‐level jet (LLJ) is found more frequently near the deep convection that experiences rapid growth to an MCS, but its presence alone is not predictive of rapid growth. Using spatially‐averaged parameters, we find that rapid growth to MCSs also occurs in environments that are significantly more thermodynamically favorable with greater low‐level moisture and instability. Fewer significant differences are found in the kinematic environment with only the 0–2 km vertical wind shear magnitude being significantly larger for rapid growth MCSs compared to slow growth MCSs, potentially related to LLJs often peaking near this height. When focusing only on MCSs with the slowest and fastest growth rates, elevated‐layer shear is significantly smaller for very rapid growth MCSs, suggesting elevated‐layer shear may help discriminate between the upper and lower bounds of growth rate. Finally, when upscale growth occurs near the Sierras de Córdoba (SDC) with a LLJ present, rapid growth is also supported by favorable wind shear orientation. However, this does not hold for upscale growth occurring away from the SDC, highlighting the importance of interpreting shear direction relative to the orientation of features initiating deep convection.

Sasaki, Clayton R. S. [Univ. of Washington, Seattl↗

DOE OSTI2021

A Deep Learning Modeling Framework to Capture Mixing Patterns in Reactive-Transport Systems

Prediction and control of chemical mixing are vital for many scientific areas such as subsurface reactive transport, climate modeling, combustion, epidemiology, and pharmacology. Due to the complex nature of mixing in heterogeneous and anisotropic media, the mathematical models related to this phenomenon are not analytically tractable. Numerical simulations often provide a viable route to predict chemical mixing accurately. However, contemporary modeling approaches for mixing cannot utilize available spatial-temporal data to improve the accuracy of the future prediction and can be compute-intensive, especially when the spatial domain is large and for long-term temporal predictions. To address this knowledge gap, in this work we will present in this paper a deep learning (DL) modeling framework applied to predict the progress of chemical mixing under fast bimolecular reactions. This framework uses convolutional neural networks (CNN) for capturing spatial patterns and long short-term memory (LSTM) networks for forecasting temporal variations in mixing. By careful design of the framework—placement of non-negative constraint on the weights of the CNN and the selection of activation function, the framework ensures non-negativity of the chemical species at all spatial points and for all times. Our DL-based framework is fast, accurate, and requires minimal data for training. The time needed to obtain a forecast using the model is a fraction (≈ O(-6)) of the time needed to obtain the result using a high-fidelity simulation. To achieve an error of 10% (measured using the infinity norm) for capturing local-scale mixing features such as interfacial mixing, only 24% to 32% of the sequence data for model training is required. To achieve the same level of accuracy for capturing global-scale mixing features, the sequence data required for model training is 64% to 70% of the total spatial-temporal data. Hence, the proposed approach—a fast and accurate way to forecast long-time spatial-temporal mixing patterns in heterogeneous and anisotropic media—will be a valuable tool for modeling reactive-transport in a wide range of applications.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2021

Using heterogeneous data to identify signatures of dengue outbreaks at fine spatio-temporal scales across Brazil

Dengue virus remains a significant public health challenge in Brazil, and seasonal preparation efforts are hindered by variable intra- and interseasonal dynamics. Here, we present a framework for characterizing weekly dengue activity at the Brazilian mesoregion level from 2010–2016 as time series properties that are relevant to forecasting efforts, focusing on outbreak shape, seasonal timing, and pairwise correlations in magnitude and onset. In addition, we use a combination of 18 satellite remote sensing imagery, weather, clinical, mobility, and census data streams and regression methods to identify a parsimonious set of covariates that explain each time series property. The models explained 54% of the variation in outbreak shape, 38% of seasonal onset, 34% of pairwise correlation in outbreak timing, and 11% of pairwise correlation in outbreak magnitude. Regions that have experienced longer periods of drought sensitivity, as captured by the “normalized burn ratio,” experienced less intense outbreaks, while regions with regular fluctuations in relative humidity had less regular seasonal outbreaks. Both the pairwise correlations in outbreak timing and outbreak trend between mesoresgions were best predicted by distance. Our analysis also revealed the presence of distinct geographic clusters where dengue properties tend to be spatially correlated. Forecasting models aimed at predicting the dynamics of dengue activity need to identify the most salient variables capable of contributing to accurate predictions. Our findings show that successful models may need to leverage distinct variables in different locations and be catered to a specific task, such as predicting outbreak magnitude or timing characteristics, to be useful. This advocates in favor of “adaptive models” rather than “one-size-fits-all” models. The results of this study can be applied to improving spatial hierarchical or target-focused forecasting models of dengue activity across Brazil.

59 BASIC BIOLOGICAL SCIENCES↗

NASA NTRS1984

Research Relative to High Spatial Resolution Passive Microwave Sounding Systems

Methods to obtain high resolution passive microwave weather observations, and understanding of their probable impact on numerical weather prediction accuracy were investigated. The development of synthetic aperture concepts for geosynchronous passive microwave sounders were studied. The effects of clouds, precipitation, surface phenomena, and atmospheric thermal fine structure on a scale of several kilometers were examined. High resolution passive microwave sounders (e.g., AMSU) with an increased number of channels will produce initialization data for numerical weather prediction (NWP) models with both increased spatial resolution and coverage. The development of statistical models for error growth in high resolution primitive equation NWP models which permit the consequences of various observing system alternatives, including sensors and assimilation times and procedures is discussed. A high resolution three dimensional primitive equation NWP model to determine parameters in an error growth model similar to that formulated by Lorenz, but with more degrees of freedom is utilized.

Staelin, D. H.↗

NASA NTRS2004

BASS Code Development

The BASS computational aeroacoustic code solves the fully nonlinear Euler equations in the time domain in two-dimensions. The acoustic response of the stator is determined simultaneously for the first three harmonics of the convected vortical gust of the rotor. The spatial mode generation, propagation and decay characteristics are predicted by assuming the acoustic field away from the stator can be represented as a uniform flow with small harmonic perturbations superimposed. The computed field is then decomposed using a joint temporal-spatial transform to determine the wave amplitudes as a function of rotor harmonic and spatial mode order. This report details the following technical aspects of the computations and analysis. 1) the BASS computational technique; 2) the application of periodic time shifted boundary conditions; 3) the linear theory aspects unique to rotor-stator interactions; and 4) the joint spatial-temporal transform. The computational results presented herein are twofold. In each case, the acoustic response of the stator is determined simultaneously for the first three harmonics of the convected vortical gust of the rotor. The fan under consideration here like modern fans is cut-off at +, and propagating acoustic waves are only expected at 2BPF and 3BPF. In the first case, the computations showed excellent agreement with linear theory predictions. The frequency and spatial mode order of acoustic field was computed and found consistent with linear theory. Further, the propagation of the generated modes was also correctly predicted. The upstream going waves propagated from the domain without reflection from the in ow boundary. However, reflections from the out ow boundary were noticed. The amplitude of the reflected wave was approximately 5% of the incident wave. The second set of computations were used to determine the influence of steady loading on the generated noise. Toward this end, the acoustic response was determined with three steady loading conditions: design, low-loading, high-loading. The overall trend showed significant (approximately 10 dB) increases in the generated noise for the highly loaded stator.

Sawyer, Scott↗

DOE OSTI2019

Characterizing the Influence of Fracture Density on Network Scale Transport

The topology of natural fracture networks is inherently linked to the structure of the fluid velocity field and transport therein. Here we study the impact of network density on flow and transport behaviors. We stochastically generate fracture networks of varying density and simulate flow and transport with a discrete fracture network model, which fully resolves network topology at the fracture scale. We study conservative solute trajectories with Lagrangian particle tracking and find that as fracture density decreases, solute channelization to large local fractures increases, thereby reducing plume spreading. Furthermore, in sparse networks mean particle travel distance increases and local network features, such as velocity zones where flow is counter to the primary pressure gradient, become increasingly important for transport. As the network density increases, network statistics homogenize and such local features have a reduced impact. We quantify local topological influence on transport behavior with an effective tortuosity parameter, which measures the ratio of total advective distance to linear distance at the fracture scale; large tortuosity values are correlated to slow-velocity regions. Additionally, these large tortuosity, slow-velocity regions delay downstream transport and enhance tailing on particle breakthrough curves. Finally, we predict transport with an upscaled, Bernoulli spatial Markov random walk model and parameterize local topological influences with a novel tortuosity parameter. Furthermore, Bernoulli model predictions improve when sampling from a tortuosity distribution, as opposed to a fixed value as has previously been done, suggesting that local network topological features must be carefully considered in upscaled modeling efforts of fracture network systems.

58 GEOSCIENCES↗

NASA NTRS2001

Design and Calibration of Autonomous Coherent Doppler Lidar for Space Missions

Developed a new algorithm for the simulation of three dimensional homogeneous turbulent velocity fields. For typical atmospheric conditions it is impossible to produce a simulated velocity field that simultaneously satisfy a given spatial correlation and the corresponding spatial spectrum because of spectral aliasing. The new algorithms produce a turbulent velocity field which has accurate spatial correlations which is required for performance predictions from space-based systems. Developed a new algorithm for extracting the spatial statistics of the atmospheric velocity field using coherent Doppler lidar. The performance of the algorithm was compared with past methods and the new algorithm produces useful results for space-based data, which was not possible before. Developed new methods for verification of the errors in ground-based and space-based Doppler lidar wind measurements. These new methods do not require independent in situ data. This is an important issue for the verification of space-based Doppler lidar measurements of the global wind field. The performance of the new algorithm was compared with past results for both space-based and ground-based operation. The new algorithm has the best performance and is the only algorithm that performed satisfactory for spacebased operation. The performance of coherent Doppler lidar for a space missions with various scanning geometries was determined using computer simulation which contained the effects of random instrumental velocity errors, wind shear, wind variability along the range-gate and from shot-to-shot, and random variations in atmospheric aerosol backscatter over the measurement volume. The bias in the velocity estimates was small and the accuracy in the is typically less than 0.5 m/s for high signal conditions. For a large number of shot per velocity estimate, the threshold signal level for acceptable estimates is proportional to the number of shots to the minus one half power. This agrees with previous results determined for ground-based measurements. The use of multi-element optical detectors for autonomous operation of coherent Doppler lidar was shown to be a very promising technique. Optimal detector geometries were determined by computer simulation of performance: for ground-based testing with a fixed calibration target and for space-based operation using the random surface returns. The effects of refractive turbulence on ground-based calibration of coherent Doppler lidar was determined by computer simulations and compared with theoretical predictions. New techniques were required to correctly predict performance for the focused beam geometry commonly used for verification of space-based operation. An improved velocity estimator was evaluated for space-based applications were signal shot measurements are used to produce vector wind measurements. This permits more accurate measurements when the signal level is not known a priori or not available from multiple shot measurements. The average Doppler lidar signal spectrum including the effects of velocity turbulence was derived and calculated. This permits new estimation algorithms for turbulence based on spectral estimates. In situ atmospheric measurements were conducted and analyzed using an instrumented kite-platform. This work helps provide the required in situ data for verification of Doppler lidar velocity statistics.

Frehlich, Rod G.↗

DOE OSTI2024

Reaction-diffusion modeling provides insights into biophysical carbon-concentrating mechanisms in land plants

Carbon-concentrating mechanisms (CCMs) have evolved numerous times in photosynthetic organisms. They elevate the concentration of CO 2 around the carbon-fixing enzyme rubisco, thereby increasing CO 2 assimilatory flux and reducing photorespiration. Biophysical CCMs, like the pyrenoid-based CCM (PCCM) of Chlamydomonas reinhardtii or carboxysome systems of cyanobacteria, are common in aquatic photosynthetic microbes, but in land plants appear only among the hornworts. To predict the likely efficiency of biophysical CCMs in C 3 plants, we used spatially resolved reaction-diffusion models to predict rubisco saturation and light use efficiency. We found that the energy efficiency of adding individual CCM components to a C 3 land plant is highly dependent on the permeability of lipid membranes to CO 2 , with values in the range reported in the literature that are higher than those used in previous modeling studies resulting in low light use efficiency. Adding a complete PCCM into the leaf cells of a C 3 land plant was predicted to boost net CO 2 fixation, but at higher energetic costs than those incurred by photorespiratory losses without a CCM. Two notable exceptions were when substomatal CO 2 levels are as low as those found in land plants that already use biochemical CCMs and when gas exchange is limited, such as with hornworts, making the use of a biophysical CCM necessary to achieve net positive CO 2 fixation under atmospheric CO 2 levels. This provides an explanation for the uniqueness of hornworts' CCM among land plants and the evolution of pyrenoids multiple times.

Kaste, Joshua A. M.↗

DOE OSTI2020

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

Remote sensing, or Earth Observation (EO), is increasingly used to understand Earth system dynamics and create continuous and categorical maps of biophysical properties and land cover, especially based on recent advances in machine learning (ML). ML models typically require large, spatially explicit training datasets to make accurate predictions. Training data (TD) are typically generated by digitizing polygons on high spatial-resolution imagery, by collecting in situ data, or by using pre-existing datasets. TD are often assumed to accurately represent the truth, but in practice almost always have error, stemming from (1) sample design, and (2) sample collection errors. The latter is particularly relevant for image-interpreted TD, an increasingly commonly used method due to its practicality and the increasing training sample size requirements of modern ML algorithms. TD errors can cause substantial errors in the maps created using ML algorithms, which may impact map use and interpretation. Despite these potential errors and their real-world consequences for map-based decisions, TD error is often not accounted for or reported in EO research. Here we review the current practices for collecting and handling TD. We identify the sources of TD error, and illustrate their impacts using several case studies representing different EO applications (infrastructure mapping, global surface flux estimates, and agricultural monitoring), and provide guidelines for minimizing and accounting for TD errors. To harmonize terminology, we distinguish TD from three other classes of data that should be used to create and assess ML models: training reference data, used to assess the quality of TD during data generation; validation data, used to iteratively improve models; and map reference data, used only for final accuracy assessment. We focus primarily on TD, but our advice is generally applicable to all four classes, and we ground our review in established best practices for map accuracy assessment literature. EO researchers should start by determining the tolerable levels of map error and appropriate error metrics. Next, TD error should be minimized during sample design by choosing a representative spatio-temporal collection strategy, by using spatially and temporally relevant imagery and ancillary data sources during TD creation, and by selecting a set of legend definitions supported by the data. Furthermore, TD error can be minimized during the collection of individual samples by using consensus-based collection strategies, by directly comparing interpreted training observations against expert-generated training reference data to derive TD error metrics, and by providing image interpreters with thorough application-specific training. We strongly advise that TD error is incorporated in model outputs, either directly in bias and variance estimates or, at a minimum, by documenting the sources and implications of error. TD should be fully documented and made available via an open TD repository, allowing others to replicate and assess its use. To guide researchers in this process, we propose three tiers of TD error accounting standards. Finally, we advise researchers to clearly communicate the magnitude and impacts of TD error on map outputs, with specific consideration given to the likely map audience.

58 GEOSCIENCES↗

DOE OSTI2020

Indicator Patterns of Forced Change Learned by an Artificial Neural Network

Abstract Many problems in climate science require the identification of signals obscured by both the “noise” of internal climate variability and differences across models. Following previous work, we train an artificial neural network (ANN) to predict the year of a given map of annual‐mean temperature (or precipitation) from forced climate model simulations. This prediction task requires the ANN to learn forced patterns of change amidst a background of climate noise and model differences. We then apply a neural network visualization technique (layerwise relevance propagation) to visualize the spatial patterns that lead the ANN to successfully predict the year. These spatial patterns thus serve as “reliable indicators” of the forced change. The architecture of the ANN is chosen such that these indicators vary in time, thus capturing the evolving nature of regional signals of change. Results are compared to those of more standard approaches like signal‐to‐noise ratios and multilinear regression in order to gain intuition about the reliable indicators identified by the ANN. We then apply an additional visualization tool (backward optimization) to highlight where disagreements in simulated and observed patterns of change are most important for the prediction of the year. This work demonstrates that ANNs and their visualization tools make a powerful pair for extracting climate patterns of forced change.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Machine learning-based microstructure prediction during laser sintering of alumina

Abstract Predicting material’s microstructure under new processing conditions is essential in advanced manufacturing and materials science. This is because the material’s microstructure hugely influences the material’s properties. We demonstrate an elegant machine learning algorithm that faithfully predicts the microstructure under new conditions, without the need of knowing the governing laws. We name this algorithm, RCWGAN-GP, which is regression-based conditional generative adversarial networks with Wasserstein loss function and gradient penalty. This algorithm was trained with experimental SEM micrographs from laser-sintered alumina under various laser powers. The RCWGAN-GP realistically regenerates the SEM micrographs under the trained laser powers. Impressively, it also faithfully predicts the alumina’s microstructure under unexplored laser powers. The predicted microstructure features, including the morphology of the sintered particles and the pores, match the experimental SEM micrographs very well. We further quantitatively examined the prediction accuracy of the RCWGAN-GP. We trained the algorithm with computer-created micrograph datasets of secondary-phase growth governed by the well-known Johnson–Mehl–Avrami (JMA) equation. The RCWGAN-GP accurately regenerates the micrographs at the trained time series, in terms of the grains’ shapes, sizes, and spatial distributions. More importantly, the predicted secondary phase fraction accurately follows the JMA curve.

08 HYDROGEN↗

DOE OSTI2020

Deep convolutional neural networks for multi-scale time-series classification and application to disruption prediction in fusion devices

The multi-scale, mutli-physics nature of fusion plasmas makes predicting plasma events challenging. Recent advances in deep convolutional neural network architectures (CNN) utilizing dilated convolutions enable accurate predictions on sequences which have long-range, multi-scale characteristics, such as the time-series generated by diagnostic instruments observing fusion plasmas. Here we apply this neural network architecture to the popular problem of disruption prediction in fusion tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. ECEi measures a fundamental plasma quantity (electron temperature) with high temporal resolution over the entire plasma discharge, making it sensitive to a number of potential pre-disruptions markers with different temporal and spatial scales. Promising, initial disruption prediction results are obtained training a deep CNN with large receptive field ({$\sim$}30k), achieving an $F_1$-score of {$\sim$}91\% on individual time-slices using only the ECEi data.

convolutional neural networks↗