Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “predictive state estimation”

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 1,009 records · Page 56
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Battery Life Prediction Using Reduced-Order Physics Models and Machine Learning (CRADA Final Report)

Phase 1 (Original CRADA, plus no-cost extension modifications #1-3, 6/1/2017 to 3/13/2021): The Australian Department of Defence (AUDoD) is performing accelerated aging tests of Li-ion batteries to benchmark their reliability and degradation characteristics. Using its previously developed battery lifetime predictive model framework, the National Laboratory of the Rockies (NLR) will develop analytical models based the AUDoD data to predict lifetime of the multiple Li-ion battery chemistries under real-world use scenarios of interest to AUDoD. The NLR model is based on physical degradation mechanisms encountered by Li-ion batteries and has been previously validated. Phase 2 (CRADA modification #4, plus no-cost extension modification #5, 2/22/2021 to 3/30/2025): Train and support Australian Department of Defence personnel to use NLR software for model-based estimation of Li-ion battery lifetime using accelerated battery aging data collected by the Australian Department of Defence. Under separate DOE funding from 2019 to 2021, NLR enhanced its battery life-prediction software using machine learning algorithms to automate portions of the model-fitting process, requiring significantly less labor and expert judgment and also adding uncertainty quantification, increasing statistical rigor. Under Phase 2, NLR will customize NLR Software and provide it to AuDoD. NLR will enhance its NLR Model to capture aging modes of AuDoD's multi-cell modules, including cell-balancing effects. NLR will develop example single-cell and multi-cell models based on one AuDoD battery aging dataset. NLR will train AuDoD personnel on NLR Software. By the conclusion of the project, NLR will have provided AuDoD the training materials, a user manual and software needed to perform their own analysis of additional and/or future battery aging datasets.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2022

Plutonium reactive transport in fractured granite: Multi-species experiments and simulations

Plutonium (Pu) in the subsurface environment can transport in different oxidation states as an aqueous solute or as colloidal particles. The transport behavior of Pu is affected by the relative abundances of these species and can be difficult to predict when they simultaneously exist. This study investigates the concurrent transport of Pu intrinsic colloids, Pu(IV) (aq) and Pu(V-VI) (aq) through a combination of controlled experiments and semi-analytical dual-porosity transport modeling. Pu transport experiments were conducted in a fractured granite at high and low flow rates to elucidate sorption processes and their scaling behavior. In the experiments, Pu(IV) (aq) was the least mobile of the Pu species, Pu(V-VI) (aq) had intermediate mobility, and the colloidal Pu, which consisted mainly of precipitated and/or hydrolyzed Pu(IV), was the most mobile. The semi-analytical modeling revealed that the sorption of each Pu species was rate-limited, as the sorption could not be described by assuming local equilibrium in the experiments. The model was able to describe the sorption of the different Pu species that occurring either on fracture surfaces, in the pores of the rock matrix, or simultaneously in both locations. While equally good fits to the data could be achieved using any of these assumptions, a fracture-dominated process was considered to be the most plausible because it provided the most reasonable estimates of sorption rate constants. Importantly, a key result of this work is that the sorption rate constant of all Pu species tends to decrease with increasing time scales, which implies that Pu will tend to be more mobile at longer time scales than observations at shorter time scales suggest. This result has important implications for predicting the environmental impacts of Pu in the safety assessments of geologic repositories for radioactive waste disposal, and we explore potential mechanistic bases for upscaling the sorption rate constants to time and distance scales that cannot be practically evaluated in experiments.

58 GEOSCIENCES↗
DOE OSTI2024

Weak-form inference for hybrid dynamical systems in ecology

Species subject to predation and environmental threats commonly exhibit variable periods of population boom and bust over long timescales. Understanding and predicting such behaviour, especially given the inherent heterogeneity and stochasticity of exogenous driving factors over short timescales, is an ongoing challenge. A modelling paradigm gaining popularity in the ecological sciences for such multi-scale effects is to couple short-term continuous dynamics to long-term discrete updates. We develop a data-driven method utilizing weak-form equation learning to extract such hybrid governing equations for population dynamics and to estimate the requisite parameters using sparse intermittent measurements of the discrete and continuous variables. The method produces a set of short-term continuous dynamical system equations parametrized by long-term variables, and long-term discrete equations parametrized by short-term variables, allowing direct assessment of interdependencies between the two timescales. We demonstrate the utility of the method on a variety of ecological scenarios and provide extensive tests using models previously derived for epizootics experienced by the North American spongy moth ( Lymantria dispar dispar ).

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

PV Degradation Modeling: Applying Geospatial Workflows with "PVDeg"

Accurate degradation modeling is essential for predicting photovoltaic (PV) module performance, estimating longevity and informing design decisions. With degradation rates varying significantly by location, geospatial analysis is critical for PV and broader applications, such as agrivoltaics, weathering and environmental data analysis. This work presents PVDeg, an open-source tool designed for geospatial degradation analysis. PVDeg integrates meteorological data from global sources, including the National Solar Radiation Database (NSRDB) and Photovoltaic Geographical Information System (PVGIS), with degradation models. The toolkit enables users to customize geospatial workflows by integrating weather data, material parameters, and user-defined Python functions. It facilitates accelerated downloads of NSRDB and PVGIS datasets and optimizes geospatial point selection to preserve data density in regions of interest. Additionally, PVDeg provides a local database for storage and spatial queries, supporting large-scale analyses without the need for high-performance computing (HPC) resources. PVDeg provides a foundational workflow that extends its utility beyond PV applications, enabling researchers to analyze geospatial processes across discipline.

14 SOLAR ENERGY↗
DOE OSTI2026

The Dependence of Electron-only Magnetic Reconnection on Plasma Beta and Magnetic Shear: Theory and Observations

We investigate the onset conditions for electron-only magnetic reconnection through an analysis of 86 reconnection events observed by the Magnetospheric Multiscale mission in magnetosheath and magnetospheric current sheets. Our study presents the first observational evidence that the occurrence of electron-only reconnection depends on both the magnetic-field shear angle (θ) across the current sheet and the cross-sheet electron beta jump, quantified either by the inflow electron beta difference (Δβ e ) or by a modified electron beta jump (Δβ$^*_e$) defined by the cross-sheet electron pressure difference normalized by the average of the inflow magnetic pressure in the two inflow regions. We extend the diamagnetic-drift suppression mechanism from standard reconnection to electron-only reconnection by deriving relations applicable to electron-scale current sheets. Using scaling estimates, the electron-only reconnection suppression relation is derived as Δβ$^*_e$ ≳ 2(L/d e ) tan (θ/2), while a natural extension of the ion-coupled reconnection suppression criterion is Δβ e ≳ 2(L/d e ) tan (θ/2). At low Δβ e and Δβ$^*_e$ values, reconnection is detected across a large range of magnetic shear (from low to high), whereas high Δβ e and Δβ$^*_e$ conditions restrict reconnection to large shear configurations: shear angles as small as θ ≈ 5° occur when Δβ e (or Δβ$^*_e$) ≲0.1. However, only events exceeding θ ≳ 70° are found when Δβ e ≳ 1.0; this apparent restriction weakens when employing Δβ$^*_e$, as the corresponding data points shift toward smaller values of the modified electron beta jump. These findings quantitatively support theoretical predictions of reconnection suppression in high-β e plasmas at low shears, attributed to X-line drift induced by pressure gradients across electron-scale current sheets.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Microbiome Aggregated Traits and Assembly Are More Sensitive to Soil Management than Diversity

How soil is managed, particularly for agriculture, exerts stresses upon soil microbiomes, resulting in altered community structures and functional states. Understanding how soil microbiomes respond to combined stresses is important for predicting system performance under different land use scenarios, aids in identification of the most environmentally benign managements, and provides insight into how system function can be recovered in degraded soils. We use a long-established field experiment to study the effects of combined chronic (press) disturbance of the magnitude of organic carbon inputs with acute (pulse) effects of physical disturbance by tillage and chemical disturbance due to inorganic fertilization and pesticide application. We show that because of the variety of ways it can be assessed, biodiversity—here based on microbial small subunit rRNA gene phylotypes—does not provide a consistent view of community change. In contrast, aggregated traits associated with soil microbiomes indicate general loss of function, measured as a reduction of average genome lengths, associated with chronic reduction of organic inputs in arable or bare fallow soils and altered growth strategies associated with rRNA operon copy number in prokaryotes, as well as a switch to pathogenicity in fungal communities. In addition, pulse disturbance by soil tillage is associated with an increased influence of stochastic processes upon prokaryote community assembly, but fungicide used in arable soils results in niche assembly of fungal communities compared to untilled grassland. Overall, bacteria, archaea, and fungi do not share a common response to land management change, and estimates of biodiversity do not capture important facets of community adaptation to stresses adequately.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Evaluating Probabilistic Deep Learning Methods for Uncertainty Quantification of Precipitation Bias Correction

Climate models often exhibit biases in their precipitation predictions, particularly underestimating high-intensity events and overestimating low precipitation. Deep learning approaches offer promising solutions, but their epistemic uncertainty associated with a deep learning–based bias correction method has not previously been quantified for reliable downstream climate impact studies. While methods for capturing the epistemic uncertainty in deep learning frameworks exist, there is currently no consensus on the best method. In this work, we compare three uncertainty quantification (UQ) methods—Deep Ensembles (DEns), Monte Carlo Dropout (MCD), and Flipout—by assessing the reliability of their uncertainty estimates using standard measures such as sharpness and calibration. These UQ methods are applied to an existing deep learning precipitation bias correction model known as UFNet: a coupled U-Net and fully connected neural network. The methods utilized to assess the models’ uncertainties are 1) calibration, which ensures that the expected probabilities of the model align with reality and 2) sharpness, which is a measure of the precision of the model’s probabilistic predictions. Of the three UQ methods evaluated, the DEns and MCD methods demonstrated the best-calibrated performance (expected calibration error of 0.36 and 0.35, respectively), compared to Flipout (0.58). In contrast, Flipout had the sharpest predictions and the highest metric performance in bias correcting precipitation—especially for higher-order moments such as kurtosis with a spatial correlation of 72% compared to 32% and 55% spatial correlation for DEns and MCD, respectively. Of the three UQ methods, MCD was found to be the most suitable method for UQ purposes based on its calibration, sharpness, and computational requirements.

Bayesian methods↗
DOE OSTI2025

Fine-scale vegetation composition and structure shape spatiotemporal variation in surface albedo across a low Arctic tundra landscape

The unprecedented rate of warming in the Arctic is driving changes in the structure and composition of tundra vegetation. Increases in deciduous tall shrub cover, height, and density are of particular concern, as these changes alter local surface albedo in ways that could amplify effects on the regional surface energy budget (SEB). Despite this importance, significant uncertainties remain in understanding the interplay between fine-scale vegetation patterns and emergent albedo dynamics across space and time. Here, we address these uncertainties by (1) quantifying spatiotemporal variation in surface shortwave albedo and (2) determining the relative influence of fine-scale vegetation composition, structure, and environmental conditions on albedo across a representative low-Arctic tundra landscape on Alaska’s Seward Peninsula. To do this, we synthesized multi-scale, multi-platform remote sensing observations, including a novel Landsat-derived albedo time series, a fine-scale map of Arctic plant functional type (PFT) fractional cover, and airborne LiDAR estimates of canopy height and topography. We show that there are substantial reductions in winter albedo for pixels dominated by tall, woody PFTs (28.13%) relative to pixels dominated by non-woody vegetation, but almost no change in summer albedo (3% increase). Further, we identified a unimodal trend in the relationship between canopy height and the timing of the springtime transition from high (snowy) to low (leafy) albedo (peak at 5.5 m), possibly because of competing ‘snow-fence’ and ‘protrusion’ snow-shrub interactions. To explore the primary drivers of albedo, we constructed a random forest model and found that canopy height and the fractional cover of woody PFTs were as- or more important predictors of winter albedo than topographic features. These findings provide strong evidence for the impacts of local vegetation characteristics on regional surface albedo, highlighting the need for better quantification of snow-shrub interactions to accurately predict the Arctic’s SEB under future environmental change.

Arctic↗
DOE OSTI2025

Validation of 3D MHD simulations of Ne/D 2 mixed shattered pellet injection in JET

Nonlinear 3D magnetohydrodynamic (MHD) modeling of shattered pellet injection (SPI) in JET tokamak plasmas is performed with the JOREK code. The study focuses on the validation of simulation result with respect to experimental observations, addressing in particular figures of merit for the efficiency of the SPI, as a technique to mitigate thermal loads by injecting radiative impurities like neon during the thermal quench (TQ) phase of disruptions. A set of JET pulses with neon/deuterium atomic mixture ratio in the shattered pellet varying from 10% to 100% provides the experimental data to be compared with. Simulations using different models for the ablation of solid fragments and the radiation of neon impurities are considered. Synthetic diagnostics are employed for a direct quantitative comparison with key JET diagnostic systems providing in particular radiation, electron temperature and density, and magnetic measurements. 3D radiation structures from MHD simulations are analyzed to estimate the toroidal asymmetry of radiation, which is difficult to measure in JET due to the presence of only two toroidally displaced multi-channel bolometry systems. JOREK simulations show that, before the TQ, the bulk of the radiation is concentrated in a toroidal region enclosing the radiating fragments that is missed by both multi-channel bolometers, suggesting a possible underestimation of the total radiation by such diagnostic systems in the pre-TQ phase. On the other hand, JOREK predicts that during the TQ the radiation becomes more toroidally symmetric, with positive implications for the heat load to plasma facing components.

JOREK↗
DOE OSTI2025

A Review of Intense Electromagnetic Fields in Heavy-Ion Collisions: Theoretical Predictions and Experimental Results

In heavy-ion collisions at relativistic energies, the incident nuclei travel at nearly the speed of light. These collisions deposit kinetic energy into the overlap region and create a high-temperature environment where hadrons “melt” into deconfined quarks and gluons. The spectator nucleons, which do not undergo scatterings, generate an ultraintense electromagnetic field—on the order of 10 18 G at the Relativistic Heavy Ion Collider and 10 19 G at the Large Hadron Collider. These powerful electromagnetic fields have a substantial impact on the produced particles, not only complicating the study of particle interactions but also inducing novel physical phenomena. To explore the nature of these fields and their interactions with deconfined quarks, we provide a detailed overview, encompassing theoretical estimations of their generation and evolution, as well as experimental efforts to detect them. We also provide physical interpretations of the discovered results and discuss potential directions for future investigations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Accelerating Hamiltonian Monte Carlo for Bayesian inference in neural networks and neural operators

Hamiltonian Monte Carlo (HMC) is a powerful and accurate method to sample from the posterior distribution in Bayesian inference. However, HMC techniques are computationally demanding for Bayesian neural networks due to the high dimensionality of the network’s parameter space and the non-convexity of their posterior distributions. Therefore, various approximation techniques, such as variational inference (VI) or stochastic gradient MCMC, are often employed to infer the posterior distribution of the network parameters. Such approximations introduce inaccuracies in the inferred distributions, resulting in unreliable uncertainty estimates. In this work, we propose a hybrid approach that combines inexpensive VI and accurate HMC methods to efficiently and accurately quantify uncertainties in neural networks and neural operators. The proposed approach leverages an initial VI training on the full network. We examine the influence of individual parameters on the prediction uncertainty, which shows that a large proportion of the parameters do not contribute substantially to uncertainty in the network predictions. This information is then used to significantly reduce the dimension of the parameter space, and HMC is performed only for the subset of network parameters that strongly influence prediction uncertainties. This yields a framework for accelerating the full batch HMC for posterior inference in neural networks. We demonstrate the efficiency and accuracy of the proposed framework on deep neural networks and operator networks, showing that inference can be performed for large networks with tens to hundreds of thousands of parameters. Finally, we show that this method can effectively learn surrogates for complex physical systems by modeling the operator that maps from upstream conditions to wall-pressure data on a cone in hypersonic flow.

Bayesian inference↗
DOE OSTI2026

Validation of an integrated modeling framework for investigating 3D plasma responses in tokamak plasmas

As contemporary experimental tokamaks are pushed toward reactor-relevant operation, they provide essential testbeds for demonstrating and exploring ELM control strategies for deployment in future fusion pilot plants. Accurate predictions of full plasma responses are essential to guide and optimize these demonstrations. This paper introduces and validates an integrated modeling framework over a historical range of DIII-D operational space. The integrated modeling framework uses only scalar plasma parameters and optional reference boundary to self-consistently and flexibly scan through tokamak operational space and estimate the corresponding ELM-suppression relevant plasma response over 3D coil phase space. The framework generates tightly converged equilibria that satisfy a target set of plasma parameters (I p , β N , l i ), with kinetic profiles constrained by an EPED(NN)-computed pedestal and empirical core model. The plasma response of these modeled equilibria is calculated with GPEC. Validated against 55 distinct DIII-D equilibria, the framework consistently reproduces experimentally constrained equilibrium pressure, q, and other representative profiles using the time-varying information from only evolving scalar plasma parameters; plasma response validation is performed for the n=3 perturbation. In particular, the plasma response of a DIII-D discharge scanning q 95 in search of ELM suppression windows is quantitatively reproduced. By accurately reproducing experimental equilibria and plasma responses across wide parameter variation, the framework supports potential for synthetic parameter scans in key operational, stability, and plasma response dimensions to investigate RMP ELM-suppression experiments and inform predictive RMP scenario optimization.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Modeling Multi-View Impedance-Based Cross-Geometry SOH Estimator for Li-ion Batteries

Abstract: Accurately estimating battery’s State of Health (SOH) remains challenging when models must generalize across cell designs and operating conditions. Most Electrochemical Impedance Spectroscopy (EIS)-based approaches either (i) hand-engineer a few Nyquist-plot features for shallow models—fast but does not generalize across geometries—or (ii) learn directly from Nyquist plots with deep networks, which removes manual feature extraction, yet still limited to a single plot type. As a result, cross-geometry robustness and deployability on constrained Internet of Things (IoT) devices remain open problems. We propose a compact Convolutional Neural Network (CNN) (∼ 10k parameters) that takes multi-representation EIS inputs—Nyquist (real/imaginary) and phase–magnitude (|Z|/ϕ) stacked as four channels, so the model can learn complementary degradation signatures while remaining small enough for fast inference. We build a dataset from cyclic aging of two geometries (LG INR18650MJ1 cylindrical cells and LIR2032 coin cells), acquire EIS every ten cycles from 10 kHz to 10 mHz (10 points/decade), and evaluate with leave-one-cell-out testing strategy. We further study fusion vs. single-representation inputs and assess feasibility for on-device deployment (e.g., NVIDIA Jetson device). The results show that training on multiple EIS representations improves SOH estimation accuracy and cross-geometry generalization compared to single-representation models, which uses only Nyquist or phase–magnitude plots. This design targets accurate, generalizable SOH prediction without manual feature engineering while enabling practical real-time use.

Bakr, Ahmed [The University of Alabama (UA)]↗
DOE OSTI2020

High-Performance Integrated Window and Façade Solutions for California

The researchers developed a new generation of high-performance façade systems and supporting design and management tools to support industry in meeting California’s greenhouse gas reduction targets, reduce energy consumption, and enable an adaptable response to minimize real-time demands on the electricity grid. The project resulted in five outcomes: (1) The research team developed an R-5, 1-inch thick, triplepane, insulating glass unit with a novel low-conductance aluminum frame. This technology can help significantly reduce residential cooling and heating loads, particularly during the evening. (2) The team developed a prototype of a windowintegrated local ventilation and energy recovery device that provides clean, dry fresh air through the façade with minimal energy requirements. (3) A daylight-redirecting louver system was prototyped to redirect sunlight 15–40 feet from the window. Simulations estimated that lighting energy use could be reduced by 35–54 percent without glare. (4) A control system incorporating physics-based equations and a mathematical solver was prototyped and field tested to demonstrate feasibility. Simulations estimated that total electricity costs could be reduced by 9-28 percent on sunny summer days through adaptive control of operable shading and daylighting components and the thermostat compared to state-of-the-art automatic façade controls in commercial building perimeter zones. (5) Supporting models and tools needed by industry for technology R&D and market transformation activities were validated. Attaining California’s clean energy goals require making a fundamental shift from today’s ad-hoc assemblages of static components to turnkey, intelligent, responsive, integrated building façade systems. These systems offered significant reductions in energy use, peak demand, and operating cost in California.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

CADIS and FW-CADIS Variance Reduction in Gamma Transport for Predicting Prompt Forensics Signatures

The goal of prompt nuclear forensics is to determine the characteristics of a nuclear detonation based on the signatures available almost immediately after the explosion. An important characteristic is the reaction time history (RTH), a measure of the device’s rate of neutron multiplication. The RTH can be estimated by observation of the gamma radiation emitted from the detonation, which can be detected directly or observed indirectly as Teller light. Gamma transport simulations used to predict these radiation fields are often modeled stochastically using the Monte Carlo N-Particle (MCNP) code, which can be a computationally demanding task due to the number of particle histories needed to achieve statistical convergence. In an attempt to improve the efficiency of these calculations, we evaluate two variance reduction techniques: Consistent Adjoint-Driven Importance Sampling (CADIS) and Forward-Weighted Consistent Adjoint-Driven Importance Sampling (FW-CADIS). These methods use a deterministically calculated adjoint flux to create weight windows and source biasing that guide MCNP sampling. We study the utility of CADIS and FW-CADIS for their use in MCNP gamma transport for nuclear forensics prediction simulations. Furthermore, the results demonstrate that both CADIS and FW-CADIS improve the accuracy for forensics-focused simulations, with CADIS being most beneficial in direct detection and FW-CADIS being ideal for computing a global Teller light source.

CADIS↗
DOE OSTI2025

Towards a Robust Adaptive Digital Twin for Fusion Applications

The development of a digital twin system for fusion applications is essential for enhancing the prediction, analysis, and optimization of complex plasma processes. Machine learning (ML), particularly deep learning has demonstrated strong capabilities in modeling such highly nonlinear and intricate systems. However, two critical challenges limit the deployment of deep learning-based digital twins: Uncertainty Quantification (UQ) and data drift. UQ is vital for ensuring trustworthy predictions, especially in decision-support scenarios. Additionally, data-driven models are often sensitive to changes in the underlying data distribution, such as shot-to-shot variations in fusion experiments, which can lead to performance degradation over time. To address these challenges, we are developing an uncertainty-aware, adaptive digital twin framework. Our approach incorporates deep learning models enhanced with Gaussian Process approximations for predictive uncertainty estimation, coupled with an online learning mechanism that enables continuous model adaptation to new experimental data. This adaptive capability allows the data driven models to respond effectively to evolving plasma behaviors and equipment conditions. Specifically, to mitigate the effects of shot-to-shot drift, our system updates itself incrementally as new data becomes available, improving both robustness and fidelity. Our vision is to evolve this data driven model into a self-sustaining digital twin system that leverages UQ based feedback to continuously refine itself and potentially support real-time decision making. This presentation will cover a brief background on uncertainty quantification for ML, our ongoing effort on development of UQ capabilities for ML, our data science pipeline from data collection to model development and analysis and online learning framework for modeling coil deflection at DIII-D. I will also briefly touch upon opportunities and challenges in development of digital twin framework.

Sammuli, Brian [General Atomics]↗
DOE OSTI2026

Self-supervised and multi-fidelity learning for extended predictive soil spectroscopy

Infrared spectroscopy is a cost-effective, non-destructive, and environmentally benign technology that is increasingly recognized as an important solution for meeting the global demand for soil data. While both near-infrared (NIR) and mid-infrared (MIR) diffuse reflectance spectroscopy enable rapid estimation of soil properties, they present a significant trade-off: NIR offers superior scalability and lower operational costs, whereas MIR provides higher analytical fidelity by capturing fundamental molecular vibrations. In this study, we propose a self-supervised, multi-fidelity learning framework designed to bridge this gap. Our approach leverages large-scale MIR spectral libraries to learn a compact, transferable latent representation, into which NIR spectra are subsequently aligned for downstream prediction. The workflow consists of pretraining a latent model on a large MIR library, adapting the representation using a smaller paired NIR–MIR dataset, and evaluating generalization on an independent external test set. Across a range of chemical and physical soil properties, we found that MIR-derived embeddings improved prediction accuracy relative to baseline models that used raw MIR inputs. Predictions derived from the spectrum conversion (NIR to MIR) task did not match the performance of the original MIR spectra but were similar or superior to predictive performance of NIR-only models, suggesting the unified spectral latent space can effectively leverage the larger and more diverse MIR dataset for prediction of soil properties not well represented in current NIR libraries.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Lake-Effect Snowstorm Events and Associated Snowfall Totals Integrated from NOAA Storm Reports, ERA5, and HRRR for the Laurentian Great Lakes (1997–2024)

Lake-effect snowstorms are localized, impactful winter weather phenomena that can generate substantial snowfall totals and pose significant challenges for forecasting, transportation, and regional infrastructure. To support the analysis and modeling of these events, this dataset compiles observational reports of lake-effect snowstorms alongside corresponding snowfall estimates derived from gridded atmospheric datasets. The observational component of the data originates from the National Weather Service (NWS) winter storm report, subset to lake-effect snow event type, covering 1997–2024. For each lake-effect snow event, this data provides the impacted county, event start and end datetimes at an hourly resolution, as well as relevant storm narratives. The complementary reanalysis-derived data is sourced from European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA5) and High-Resolution Rapid Refresh (HRRR) gridded data. For both gridded datasets, the maximum total snowfall (in units mm) was extracted, constrained by the county and datetimes specified by the observational report. ERA5 data covers the entire observational period (1997–2024), whereas HRRR data is only available from November 2016 – December 2024. Three CSV files are provided here: (1) the observational lake-effect snow event report, (2) ERA5 maximum snowfall detections for each event, and (3) HRRR maximum snowfall detections for each event. Relevant data from the observational files, such as impacted state and county, event datetimes, and event IDs, were included for convenience. Users can inspect and visualize the data using tools such as Microsoft Excel and Python pandas/matplotlib packages. This dataset may support a variety of applications, including climatological analyses of lake-effect snowfall, evaluation of snowfall representation in atmospheric datasets and numerical weather prediction models, and the development of machine learning approaches for detecting or predicting lake-effect snowfall events.

EARTH SCIENCE > ATMOSPHERE > PRECIPITATION > SOLID↗