SEARCH · Engineering Papers

Results for “evaluation datasets”

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 OSTI2020

Development of a Comprehensive Two-Phase Flow Database for the Validation of NEK-2P

Three-dimensional (3-D) two-phase Computational Fluid Dynamics (CFD) codes are emerging as a powerful and potentially practical tool for applications in which detailed local flow information is needed. However, two-phase flow models and associated closure relations are not well established for CFD applications, which is partly due to the lack of high-quality validation data. The main objective of this work is to develop a comprehensive database of two-phase flows that can be used to validate two-phase CFD codes such as NEK-2P. In this project, four advanced local measurement systems, including Particle Image Velocimetry and Planar Laser-Induced Fluorescence (PIV-PLIF), high-speed imaging, x-ray densitometry, and multi-sensor conductivity probe are employed to measure the local two-phase flow parameters of both gas and liquid phases. By combining these techniques, the local void fraction, bubble velocity, interfacial area concentration, bubble frequency, liquid velocity, turbulence intensity, etc., in various two-phase flow regimes can be obtained. These local measurement techniques are first used in a 25.4 mm circular pipe test section. Seven air-water two-phase flow conditions spanning the bubbly, slug, churn-turbulent, and annular flow regimes are measured in this facility. The obtained database contains the radial profiles of both gas- and liquid-phase parameters at three axial locations along the test section. The second facility used in this work contains a 30 mm × 10 mm rectangular test section, in which three two-phase flow and two single-phase flow conditions are measured. Two-dimensional distributions of local two-phase flow parameters in the cross-sectional plane are measured in this facility at three axial locations as well. A facility featuring a 3×3 electrically heated rod bundle is also designed and being constructed in this project. This facility is specially designed for optical measurements and is expected to provide high-quality boiling data in the future. Preliminary analyses have been performed for the data taken in the 25.4 mm circular pipe. Both center-peaked and wall-peaked void fraction profiles have been observed in the data depending on the two-phase flow conditions and/or developing lengths. The 1-D drift-flux model was evaluated with the newly obtained datasets, in which both gas- and liquid-phases data were directly measured. The distribution parameter model has been optimized based on a new void-profile classification method proposed in this study. The optimized drift-flux model shows a significant improvement in predicting both gas velocity and void fraction. The measured liquid-phase turbulence was used to benchmark Sato’s turbulence model considering the bubble-induced shear stress for the three tested bubbly flows. The benchmark results showed good agreement between the PIV measurements and model predictions. In the bubbly flows tested that have low void fractions less than 3%, the effect of the bubble-induced turbulence was found not significant. However, the bubble-induced shear stress becomes important with the increase of the void fraction. The Conjugate Heat Transfer (CHT) model was developed and implemented in NEK-2P. This model allows the coupled simulation of the solid domain and two-phase fluid domain, allowing the specification of realistic boundary conditions. The CHT implementation was verified first with non-boiling simulations. The predicted temperatures in the fluid domain were shown to be identical in simulations with or without the CHT model. The CHT model was then validated through simulations of three Becker benchmark CHF tests performed under both Dryout (DO) and Departure from Nucleate Boiling (DNB) conditions. Reasonably good agreement was obtained between calculated wall temperatures and corresponding experimental data.

42 ENGINEERING↗

DOE OSTI2021

Fuzzy Simplicial Networks: A Topology-Inspired Model to Improve Task Generalization in Few-shot Learning

Deep learning has shown great success in settings with massive amounts of data but has struggled when data is limited. Few-shot learning algorithms, which seek to address this limitation, are designed to generalize well to new tasks with limited data. Typically, models are evaluated on unseen classes and datasets that are defined by the same fundamental task as they are trained for (e.g. category membership). One can also ask how well a model can generalize to fundamentally different tasks within a fixed dataset (for example: moving from category membership to tasks that involve detecting object orientation or quantity). To formalize this kind of shift we define a notion of “independence of tasks” and identify three new sets of labels for established computer vision datasets that test a model's ability to generalize to tasks which draw on orthogonal attributes in the data. We use these datasets to investigate the failure modes of metric-based few-shot models. Based on our findings, we introduce a new few-shot model called Fuzzy Simplicial Networks (FSN) which leverages a construction from topology to more flexibly represent each class from limited data. In particular, FSN models can not only form multiple representations for a given class but can also begin to capture the low-dimensional structure which characterizes class manifolds in the encoded space of deep networks. We show that FSN outperforms state-of-the-art models on the challenging tasks we introduce in this paper while remaining competitive on standard few-shot benchmarks.

deep learning↗

DOE OSTI2024

A Bayesian Learning Approach to Wireless Outdoor Heatmap Construction using Deep Gaussian Process

We present a novel Bayesian learning approach to outdoor radio heatmap construction utilizing deep Gaussian process (GP). The proposed approach employs a two-layer hierarchy which consists of two cascaded Gaussian processes that are capable of modeling more complex input-output relations than standard single-layer Gaussian processes. Since deriving the exact model likelihood is challenging, a lower bound is optimized instead so that gradient descent-based methods can be performed to find out the optimal model parameters. Typically, inducing points are used in GPs to facilitate low-rank approximation of covariance (kernel) matrices for computation speedup. However, the inaccuracy induced by inducing points can accumulate when stacking multiple layers of GP which may hinder the performance of deep GP. Moreover, since inducing points need to be learned, having them at all layers of deep GP also incurs computational burden. To overcome the above challenges, in contrast to the canonical deep GP model, we use a modified architecture where a full standard GP resides in the first layer and inducing points are only introduced for the second layer. This modified architecture strikes a balance between model accuracy and training complexity. In the proposed model, the noise parameter of the first GP layer is also eliminated to improve the training efficiency as the noise parameter at the output of the second layer suffices to model the uncertainty in the output. The proposed approach is evaluated on real-world datasets, in the form of location-Received Signal Strength (RSS) pairs, collected from the Platform for Open Wireless Data-driven Experimental Research (POWDER) located at the campus of the University of Utah. Experiment results show that the proposed approach can achieve smaller prediction errors on various training and testing data configurations than DNN-based and GP-based methods.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

Neural Image Compression: Generalization, Robustness, and Spectral Biases

Recent advances in neural image compression (NIC) have resulted in models which are starting to outperform traditional codecs. While this has led to growing excitement about using these methods in real-world applications, the successful adoption of any machine learning system (including NIC) in the wild requires it to generalize (and be robust) to unseen distribution shifts at deployment time. Unfortunately, current research lacks comprehensive datasets and informative tools to evaluate and understand compression performance in real-world settings. To bridge this crucial gap, first, this paper presents a comprehensive benchmark suite to evaluate the out-of-distribution (OOD) performance of image compression methods. Specifically, we design CLIC-C and Kodak-C by introducing 15 common corruptions to popular CLIC and Kodak benchmarks. Next, we propose spectrally inspired introspection tools to gain a deeper understanding of errors introduced by image compression methods as well as their OOD performance. To this end, we carry out a detailed performance comparison of the classical codec with various variants of NIC (e.g., original, variable rate, pruned), revealing intriguing findings that challenge our current understanding of the strengths and limitations of NIC. Finally, we corroborate our empirical findings with theoretical analysis, providing an in-depth view of the OOD performance of NIC. Our benchmarks, spectral introspection tools, and findings provide a crucial bridge to the real-world adoption of NIC. We hope that our work will propel future efforts in designing more robust and generalizable NIC methods.

neural networks, Variational Autoencoder, robustne↗

DOE OSTI2025

AI Benchmark Democratization and Carpentry

Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.

von Laszewski, Gregor [Virginia U.]↗

DOE OSTI2025

Immersive Scientific Visualization of Molten-Salt Reactor Waste Characteristics Using Virtual Reality

Immersive visualization is changing how we explore, communicate, and understand complex scientific systems. In nuclear energy, an area in which data are often multidimensional, time-dependent, and difficult to interpret, virtual reality (VR) represents a powerful and intuitive informational medium. This work introduces a VR-based platform that visualizes the post-shutdown behavior and waste management lifecycle of molten-salt reactors (MSRs), a next-generation reactor type with unique operational and safety characteristics. The platform, built in Unity, is streamed on the Meta Quest 3 headset. It transforms high-fidelity simulation data into an interactive, immersive experience. Users can explore time-dependent reactor characteristics such as nuclide decay, which is a key factor for evaluating reactor waste strategies. The datasets were generated using the MOOSE (Multiphysics Object-Oriented Simulation Environment) framework and then processed through ParaView scripting for smooth integration into Unity. From a visualization standpoint, the platform emphasizes spatial storytelling, temporal exploration, and user-centered interaction. Users can navigate 3D reactor geometries, slice through volumetric data, and manipulate time to observe how physical phenomena evolve. Real-scale rendering and embodied interaction make the experience feel tangible. The interface is designed to be accessible, even to those without nuclear or simulation expertise. This lowers the barrier for stakeholders, policymakers, and the general public, while still supporting expert analysis and collaborative decision-making. This work shows how immersive visualization can function as both a scientific tool and a communication interface. By integrating simulation, processing, and visualization into a cohesive workflow, we offer a scalable framework for immersive scientific storytelling. The modular design supports future extensions to other reactor types and lifecycle stages, from shutdown to long-term storage, making the platform adaptable for both research and outreach.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2025

Performance Evaluation of Weather@home2 Simulations over West African Region

Weather and climate forecasting, using climate models, have become essential tools and life-savers in the West African region; in spite of the fact that climate models do not fully comply with attributes of forecast qualities—RASAP: reliability, association, skill, accuracy, and precision. The objective of this paper is to quantitatively evaluate, in comparison to CRU and ERA5 datasets, the RASAP compliance-level of the weather@home2 modeling system (w@h2). Findings from some statistical evaluations show that, to a moderately significant extent, w@h2 model provides useful information during the monsoon seasons; skills to capture the Little Dry Season over the Guinea zone; predictive skills for the onset season; ability to reproduce all the annual characteristics of the surface maximum air temperature over the region; as well as skill to detect heat waves that usually ravage West Africa during the boreal spring. The model displays traces of attributes that are needed for seasonal climate predictions and applications. Deficiencies in the quantitative reproducibility point to the facts that the model does provide a reliability akin to that of regional climate models. This paper further furnishes a prospective user with information on whether the model might be “useful or not” for a particular application.

West Africa↗

DOE OSTI2025

Next-Level Energy Management in Manufacturing: Facility-Level Energy Digital Twin Framework Based on Machine Learning and Automated Data Collection

This research introduces an energy prediction framework at the facility level supported by automated data collection and machine learning models. It investigates whether reducing the prediction time scale allows for applying more complex machine learning techniques and if those techniques improve the prediction accuracy. The primary advantages of this framework lie in its automation of the energy prediction process and its provision of real-time energy data suitable for use in energy dashboards or digital twins. A sitewide dataset was created by combining 15 min energy and daily production data of five shops—assembly, battery, body (electric), body (gas), and paint—from a globally recognized electric vehicle manufacturer. Various machine learning models were evaluated on daily, weekly, and monthly datasets, including, in increasingly complex order: naïve, simple linear regression, net regularized generalized linear regression, principal component regression, k-nearest neighbor, random forest, and Bayesian regularized neural network. Compared to the current state-of-the-art energy consumption prediction for the industrial facility level, this research investigates more complex models and smaller time intervals for higher accuracy. The findings revealed that the more complex monthly models require a minimum of a year and a half of data to operate, while weekly models demand a year of data to achieve improved accuracy. Daily models can operate with only six months of data but exhibit poor performance due to reduced prediction accuracy of production. Key challenges identified include access to reliable, high-quality energy and production data and the initial demand for human labor.

digital twin↗

DOE OSTI2025

An Evaluation of Representation Learning Methods in Particle Physics Foundation Models

We present a systematic evaluation of representation learning objectives for particle physics within a unified framework. Our study employs a shared transformer-based particle-cloud encoder with standardized preprocessing, matched sampling, and a consistent evaluation protocol on a jet classification dataset. We compare contrastive (supervised and self-supervised), masked particle modeling, and generative reconstruction objectives under a common training regimen. In addition, we introduce targeted supervised architectural modifications that achieve state-of-the-art performance on benchmark evaluations. This controlled comparison isolates the contributions of the learning objective, highlights their respective strengths and limitations, and provides reproducible baselines. We position this work as a reference point for the future development of foundation models in particle physics, enabling more transparent and robust progress across the community.

Chen, Michael [Caltech]↗

DOE OSTI2020

Assessing Current and Future Infrastructure Hazards

Project Objective Execute intelligent analytics via an advanced analytical framework, to assess the current state of offshore infrastructure, evaluate infrastructure life, and identify technologies to reduce infrastructure hazards, costs, and extend infrastructure life. Approach & Results thus Far• Build comprehensive dataset• Perform data-driven analytics to evaluate infrastructure integrity 1. Remaining lifespan 2. Likelihood of future risk• Apply data-driven advanced spatial, statistical, and Machine Learning (ML) models to quantify existing infrastructure integrity• Release data and models through a smart, online platform hosted by Energy Data eXchange (EDX)

Romeo, Lucy F.↗

DOE OSTI2020

Evaluation of historical CMIP6 model simulations of extreme precipitation over contiguous US regions

Simulated historical precipitation is evaluated for Coupled Model Intercomparison Project Phase 6 (CMIP6) models using precipitation indices defined by the Expert Team on Climate Change Detection and Indices. The model indices are evaluated against corresponding indices from the CPC unified gauge-based analyses of precipitation over seven geographical regions across the contiguous US (CONUS). The regions assessed match those in recent US National Climate Assessment Reports. To estimate observational uncertainty, precipitation indices for three other observational datasets (HadEx2, Livneh and PRISM) are evaluated against the CPC analyses. Both the moderate and extreme mean precipitation intensities are overestimated over the western CONUS and underestimated in the areas of the Central Great Plains (CGP) in most CMIP6 models tested. Most CMIP6 models overestimate the mean and variability of wet spell durations and underestimate the mean and variability of dry spell durations across the CONUS. Biases in interannual variability of most of the indices have similar patterns to those in corresponding mean biases. The median and interquartile model spreads in CMIP6 model biases are clearly smaller than those in CMIP5 model biases for wet spell durations. Multimodel medians of CMIP6 (CMIP6-MMM) and CMIP5 (CMIP5-MMM) have similar biases in climatology and variability but biases tend to be smaller in CMIP6-MMM. Depending on the index, extreme precipitation is slightly better in parts of the eastern half of the CONUS in CMIP6-MMM, otherwise, the biases in climatology and variability are similar to CMIP5-MMM. CMIP6-MMM performs better than individual models and even observational datasets in some cases. Differences between observational datasets for most indices are comparable to the CMIP6 interquartile model spread. The better-performing observational and model datasets are different in different parts of the CONUS.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2026

A Centralized AI Lakehouse Framework for Brain Tumor MRI Classification and Segmentation, University KPI Forecasting, and Water Potability Prediction

In many university and healthcare projects, models are built for very different data types such as tables, institutional time series, and medical images, but they are deployed as separate applications. In this work, that separation made testing and maintenance difficult because each module had its own pipeline and runtime requirements. This paper presents an integrated AI lakehouse-style implementation that runs three model pipelines inside one containerized backend. For medical imaging, we used MRI datasets from IEEE DataPort: a four-class classification set with 7012 images (5708 train/1304 test) and a segmentation set with 3063 image–mask pairs. The classification model (ResNet50 transfer learning) is evaluated using a proper train–validation–test protocol across multiple splits (80/10/10, 70/10/20, 60/10/30, and 10/30/60), achieving a test accuracy of 99.00% under the standard 80/10/10 split. Additionally, a patient-level evaluation is conducted using an external glioma dataset to provide a more realistic assessment without data leakage. The segmentation model (DeepLabV3-ResNet50) achieved 83.09% validation mIoU and 88.79% Dice score. For university KPI forecasting, we used annual IPEDS and NSF HERD data from 2010 to 2023 for three universities (BSU, EOU, and UAB). To examine the effect of preprocessing on forecasting performance, two case studies are conducted. In the first case, linear interpolation is applied to generate semester-level data. In the second case, the original annual data is used directly without interpolation. Random Forest regression and ARIMA models are evaluated using MAE, RMSE, MAPE, and R 2 . The results showed that interpolation improved apparent forecasting performance due to smoothing, while evaluation on the original annual data provided a more realistic assessment of model behavior. To further validate the framework on a larger dataset, an additional case study is conducted using a student dropout dataset. For water potability, we trained and compared multiple tabular classifiers on a large dataset (1,048,575 samples). A Random Forest model (100 trees, max depth 10) achieved 85.86% test accuracy and high recall for unsafe samples (0.8447). All modules are served via FastAPI and deployed together using Docker, with workflow automation routing requests to the correct endpoint. System-level benchmarking indicates that the backend maintains stable throughput and latency under concurrent requests.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Domain Shift Analysis in Chest Radiographs Classification in a Veterans Healthcare Administration Population

This study aims to assess the impact of domain shift on chest X-ray classification accuracy and to analyze the influence of ground truth label quality and demographic factors such as age group, sex, and study year. We used a DenseNet121 model pre-trained MIMIC-CXR dataset for deep learning-based multi-label classification using ground truth labels from radiology reports extracted using the CheXpert and CheXbert Labeler. We compared the performance of the 14 chest X-ray labels on the MIMIC-CXR and Veterans Healthcare Administration chest X-ray dataset (VA-CXR). The validation of ground truth and the assessment of multi-label classification performance across various NLP extraction tools revealed that the VA-CXR dataset exhibited lower disagreement rates than the MIMIC-CXR datasets. Additionally, there were notable differences in AUC scores between models utilizing CheXpert and CheXbert. When evaluating multi-label classification performance across different datasets, minimal domain shift was observed in the unseen VA dataset, except for the label “Enlarged Cardiomediastinum.” The subgroup with the most significant variations in multi-label classification performance was study year. These findings underscore the importance of considering domain shift in chest X-ray classification tasks, paying particular attention to the temporality of the exam. Our study reveals the significant impact of domain shift and demographic factors on chest X-ray classification, emphasizing the need for improved transfer learning and robust model development. Addressing these challenges is crucial for advancing medical imaging research and improving patient care.

chest X-ray image classification↗

DOE OSTI2023

Templates of expected measurement uncertainties for neutron-induced capture and charged-particle production cross section observables

This paper provides a template of expected uncertainties and correlations for measurements of neutron-induced capture and charged-particle production cross sections. Measurements performed in-beam include total absorption spectroscopy, total energy detection, γ -ray spectroscopy, and direct charged-particle detection. Offline measurements include activation analysis and accelerator mass spectrometry. The information needed for proper use of the datasets in resonance region and high energy region evaluations is described, and recommended uncertainties are provided when specific values are not available for a dataset.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Monthly hydropower generation data for Western Canada to support Western-US interconnect power system studies

Hydroelectric power generation in Western Canada significantly contributes to power grid operations of the North American Western Interconnection through substantial generation, some of which is exported to the United States (U.S.). However, the lack of publicly available hydropower generation datasets poses challenges for future market projections and resource adequacy evaluations. We present a simulation-based monthly power system model-ready hydropower generation dataset for 110 facilities in British Columbia and Alberta from 1981 to 2019. These monthly hydropower generation estimates are developed from integrated hydrologic model simulations of runoff and reservoir-operated streamflow, followed by scaling that considers diversion inflow constraints based on hydropower water license information. To address the lack of comparable hydropower generation records, we conduct step-by-step evaluations for simulated runoff, regulated streamflow, and hydropower generation using available observations or estimates. The presented hydropower dataset aims to enhance the representation of hydropower resources in Western Canada, supporting power grid system studies for the Western Interconnection of the U.S. and Canada.

13 HYDRO ENERGY↗

DOE OSTI2025

Structural Evolution of the Hogback Monocline and Its Tectonic Significance in the San Juan Basin

The San Juan Basin is recognized as a Laramide foreland basin. It is located within the Colorado Plateau, a broad tectonic province characterized by a thick sedimentary sequence that was segmented into smaller sub basins during the Late Cretaceous to Paleogene Laramide orogeny. The Hogback Monocline lies along the northwestern margin of the San Juan Basin and is considered a Laramide-age structure formed in response to compressional stress. In this study, we interpret surface and subsurface datasets to construct a structural geological model and evaluate its tectonic significance. Through seismic data, we identify key fault and fold geometries at depth. The seismic dataset used in this study was reprocessed in depth and constrained with well log velocity data to enhance seismic imaging quality. Additionally, we performed well log correlations to identify formation tops and assess variations in basin infill and thickness geometry. A series of structural cross-sections, constructed using seismic data and a high density of boreholes, are presented to evaluate geometric variations along the structure and its evolution during basin development. Furthermore, kinematic restoration and forward modeling analyses were conducted to validate our structural interpretation. This work suggests that the Hogback Monocline formed through fault-propagation folding and flexural slip affecting the pre-Laramide sedimentary sequence under compressional stresses associated with the Laramide orogeny. This structure is interpreted as a high-angle reverse fault that influenced the geometry of the late basin infill. Additionally, monocline bending along the structure may have been controlled by fault relay systems and, in some cases, influenced by strike-slip faulting.

Reyes, Martin [New Mexico Bureau o fGeology and Mi↗

DOE OSTI2024

Power System Operational Impacts of Electric Vehicle Dynamic Wireless Charging

The electrification of the transportation sector poses an opportunity for reducing greenhouse gas (GHG) emissions from passenger vehicles. Electric vehicle (EV) charging through dynamic wireless power transfer (DWPT), known as roadway electrification, could shift EV demand profiles to better coincide with renewable electricity generation. However, this would be a very large new load and few studies evaluate the regional impacts of DWPT charging in a power transmission system. This paper defines methods that address dataset generation for passenger vehicle trips and models to evaluate regional impacts for this emerging technology. Household vehicle miles traveled (VMT) data form localized EV demand profiles through discrete-event simulation. This data serves as exogenous inputs for a Production Cost Model (PCM) of a synthetic transmission system based on the Electric Reliability Council of Texas's (ERCOT) network. EV charging methods are compared for both a 2018 baseline generation mixture and a high-renewable generation case incorporating 20 GW of installed solar photovoltaic (PV) capacity. The PCM employs unit commitment and economic dispatch (UC&ED) models to compare financial, environmental, and grid reliability impacts from EV charging across passenger EV adoption levels. In-transit charging could reduce grid operational costs by as much as 1.49%, with up to $13.7B saved in annual vehicle operational costs for consumers compared to gas-powered vehicles. Health impacts analysis from power plant and vehicle tailpipe emissions from this study show net health benefits increase by 40% for in-transit charging coupled with high renewable generation. Renewable resources provide an avenue for cost-effective in-transit charging with reduced emissions. The combination of dataset generation and open-source power system modeling establish a foundation for the holistic evaluation of regional DWPT impacts.

dynamic wireless power transfer↗

DOE OSTI2021

Gap-filling eddy covariance methane fluxes: Comparison of machine learning model predictions and uncertainties at FLUXNET-CH4 wetlands

Time series of methane fluxes measured by eddy-covariance require gap-filling to estimate annual emissions. Gap-filling methane fluxes is challenging because of high variability and complex responses to multiple drivers. To date, there is no widely established gap-filling standard for methane, with regards both to the best model algorithms and predictors. In this study, we address the need for standardization by synthesizing results of gap-filling methods applied at 17 wetland sites spanning boreal to tropical regions including all major wetlands classes and two rice paddies. We introduce new procedures for: 1) creating realistic artificial gap scenarios, 2) training and evaluating gap-filling models without overstating performance, and 3) predicting half-hourly methane fluxes and annual emissions with robust uncertainty estimates. We tested a conventional method (marginal distribution sampling) and four machine learning algorithms - penalized linear regression, artificial neural networks, random forests, and boosted decision trees - and four predictor sets, including temporal, meteorological, ecosystem carbon and energy flux, and soil predictors. We find that the conventional method can achieve similar median performance to the machine learning models but is worse than the best machine learning models and relatively insensitive to predictor choices. Of the machine learning models, decision tree algorithms performed the best in cross-validation experiments, even with a baseline predictor set, and artificial neural networks showed comparable performance when using all predictors. Soil temperature was frequently the most important predictor whilst water table depth was important at sites with substantial water table fluctuations, highlighting the value of data on soil conditions. Raw gap-filling uncertainties from the machine learning models were underestimated and we propose a method to calibrate uncertainties to observations. Finally, we gap-fill and provide summary evaluation metrics for all 81 sites in the FLUXNET-CH4 community dataset and publicly release the python code for model development, evaluation, and uncertainty estimation.

42 ENGINEERING↗