SEARCH · Engineering Papers

Results for “Spreading models”

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 703 records · Page 39

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Abrupt emissions reductions during COVID-19 contributed to record summer rainfall in China

Record rainfall and severe flooding struck eastern China in the summer of 2020. The extreme summer rainfall occurred during the COVID-19 pandemic, which started in China in early 2020 and spread rapidly across the globe. By disrupting human activities, substantial reductions in anthropogenic emissions of greenhouse gases and aerosols might have affected regional precipitation in many ways. Here, we investigate such connections and show that the abrupt emissions reductions during the pandemic strengthened the summer atmospheric convection over eastern China, resulting in a positive sea level pressure anomaly over northwestern Pacific Ocean. The latter enhanced moisture convergence to eastern China and further intensified rainfall in that region. Modeling experiments show that the reduction in aerosols had a stronger impact on precipitation than the decrease of greenhouse gases did. We conclude that through abrupt emissions reductions, the COVID-19 pandemic contributed importantly to the 2020 extreme summer rainfall in eastern China.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Coordinated time variability of multi-phase ultra-fast outflows in J132216.25 + 052446.3

ABSTRACT We present a time variability analysis of broad absorption lines (BAL; spread over the velocity range of 5800–29 000 km s−1) seen in the spectrum of J132216.25 + 052446.3 (zem = 2.04806) at ten different epochs spanning over 19 yr. The strongest absorption component (BAL-A; spread over 5800–9900 km s−1) is made up of several narrow components having velocity separations close to C iv doublet splitting. The C iv, N v, and Si iv absorption from BAL-A show correlated optical depth variability without major changes in the velocity structure. A very broad and shallow absorption (BAL-C; spread over the velocity range 15 000–29 000 km s−1) emerged during our monitoring period coinciding with a dimming episode of J1322 + 0524. All the identified absorption lines show correlated variability with the equivalent widths increasing with decreasing flux. This together with the C iv emission line variability is consistent with ionization being the main driver of the correlated variability. The observed UV-continuum variations are weaker than what is required by the photoionization models. This together with a scatter in the C iv equivalent width at a given continuum flux can be understood if variations of the C iv ionizing photons are much larger than that of the UV continuum, the variations in the ionizing photon and UV fluxes are not correlated and/or the covering factor of the flow varies continuously. We suggest BAL-A is produced by a stable clumpy outflow located beyond the broad emission line region and BAL-C is a newly formed wind component located near the accretion disc and both respond to changes in the ionizing continuum.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2022

Neural Architecture and Feature Search for Predicting the Ridership of Public Transportation Routes

Accurately predicting the ridership of public-transit routes provides substantial benefits to both transit agencies, who can dispatch additional vehicles proactively before the vehicles that serve a route become crowded, and to passengers, who can avoid crowded vehicles based on publicly available predictions. The spread of the coronavirus disease has further elevated the importance of ridership prediction as crowded vehicles now present not only an inconvenience but also a public-health risk. At the same time, accurately predicting ridership has become more challenging due to evolving ridership patterns, which may make all data except for the most recent records stale. One promising approach for improving prediction accuracy is to fine-tune the hyper-parameters of machine-learning models for each transit route based on the characteristics of the particular route, such as the number of records. However, manually designing a machine-learning model for each route is a labor-intensive process, which may require experts to spend a significant amount of their valuable time. To help experts with designing machine-learning models, we propose a neural-architecture and feature search approach, which optimizes the architecture and features of a deep neural network for predicting the ridership of a public-transit route. Our approach is based on a randomized local hyper-parameter search, which minimizes both prediction error as well as the complexity of the model. We evaluate our approach on real-world ridership data provided by the public transit agency of Chattanooga, TN, and we demonstrate that training neural networks whose architectures and features are optimized for each route provides significantly better performance than training neural networks whose architectures and features are generic.

Ayman, Afiya↗

DOE OSTI2022

Measuring the electron Yukawa coupling via resonant s-channel Higgs production at FCC-ee

The Future Circular Collider (FCC-ee) offers the unique opportunity of studying the Higgs Yukawa coupling to the electron, $y_\mathrm {e}$ , via resonant s -channel production, $\mathrm {e^+e^-}\rightarrow \mathrm {H}$ , in a dedicated run at $\sqrt{s} = m_\mathrm {H}$ . The signature for direct Higgs production is a small rise in the cross sections for particular final states, consistent with Higgs decays, over the expectations for their occurrence due to Standard Model (SM) background processes involving $\mathrm {Z}^*$ , $\gamma ^*$ , or t -channel exchanges alone. Performing such a measurement is remarkably challenging for four main reasons. First, the low value of the e $^\pm $ mass leads to a tiny $y_\mathrm {e}$ coupling and correspondingly small cross section: $\sigma _\mathrm {ee\rightarrow H} \propto m_\mathrm {e}^2 = 0.57$ fb accounting for initial-state $\gamma $ radiation. Second, the $\mathrm {e^+e^-}$ beams must be monochromatized such that the spread of their centre-of-mass (c.m.) energy is commensurate with the narrow width of the SM Higgs boson, $\varGamma _\mathrm {H} = 4.1$ MeV, while keeping large beam luminosities. Third, the Higgs mass must also be known beforehand with a few-MeV accuracy in order to operate the collider at the resonance peak, $\sqrt{s} = m_\mathrm {H}$ . Last but not least, the cross sections of the background processes are many orders-of-magnitude larger than those of the Higgs decay signals. A preliminary generator-level study of 11 Higgs decay channels using a multivariate analysis, which exploits boosted decision trees to discriminate signal and background events, identifies two final states as the most promising ones in terms of statistical significance: $\mathrm {H}\rightarrow gg$ and $\mathrm {H}\rightarrow \mathrm {W}\mathrm {W}^*\!\rightarrow \ell \nu $ + 2 jets. For a benchmark monochromatization with 4.1-MeV c.m. energy spread (leading to $\sigma _\mathrm {ee\rightarrow H} = 0.28$ fb) and 10 ab $^{-1}$ of integrated luminosity, a $1.3\sigma $ signal significance can be reached, corresponding to an upper limit on the e $^\pm $ Yukawa coupling at 1.6 times the SM value: |y e | < 1.6|y e SM | at 95% confidence level, per FCC-ee interaction point per year. Directions for future improvements of the study are outlined.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Uncovering heterogeneous intercommunity disease transmission from neutral allele frequency time series

The COVID-19 pandemic has underscored the need for accurate epidemic forecasting to predict pathogen spread, evolution, and evaluate intervention strategies. Forecast reliability hinges on detailed knowledge of disease transmission across population segments, which may be inferred from contact surveys or mobility data. However, these indirect approaches make it difficult to estimate rare transmissions between socially or geographically distant communities. We show that the steep ramp-up of genome sequencing surveillance during the pandemic can be leveraged to directly identify transmission patterns between geographically defined communities. Our approach uses a hidden Markov model to infer the fraction of infections a community imports from others based on how rapidly allele frequencies in the focal community converge to those in the donor communities. Applying this method to SARS-CoV-2 sequencing data from England and the United States, we uncover networks of intercommunity transmission that reflect geographical relationships while exposing significant long-range interactions. The scaling of importation rate with distance is consistent across both countries, yet weaker than expected based on mobility data, highlighting limitations of indirect inference. We show that transmission patterns can change between waves of variants of concern and analyze how the inferred heterogeneity in intercommunity transmission impacts evolutionary forecasts. While applied here to geographically defined communities, our approach could be applied to those defined by other traits (e.g., age, socioeconomic status), provided time-series data can be stratified accordingly. Overall, our study highlights population genomic time series data as a crucial record of epidemiological interactions, which can be deciphered using tree-free inference methods.

Okada, Takashi [Department of Physics; University ↗

DOE OSTI2021

Mycobacterium tuberculosis Phe-tRNA synthetase: structural insights into tRNA recognition and aminoacylation

Abstract Tuberculosis, caused by Mycobacterium tuberculosis, responsible for ∼1.5 million fatalities in 2018, is the deadliest infectious disease. Global spread of multidrug resistant strains is a public health threat, requiring new treatments. Aminoacyl-tRNA synthetases are plausible candidates as potential drug targets, because they play an essential role in translating the DNA code into protein sequence by attaching a specific amino acid to their cognate tRNAs. We report structures of M. tuberculosis Phe-tRNA synthetase complexed with an unmodified tRNAPhe transcript and either L-Phe or a nonhydrolyzable phenylalanine adenylate analog. High-resolution models reveal details of two modes of tRNA interaction with the enzyme: an initial recognition via indirect readout of anticodon stem-loop and aminoacylation ready state involving interactions of the 3′ end of tRNAPhe with the adenylate site. For the first time, we observe the protein gate controlling access to the active site and detailed geometry of the acyl donor and tRNA acceptor consistent with accepted mechanism. We biochemically validated the inhibitory potency of the adenylate analog and provide the most complete view of the Phe-tRNA synthetase/tRNAPhe system to date. The presented topography of amino adenylate-binding and editing sites at different stages of tRNA binding to the enzyme provide insights for the rational design of anti-tuberculosis drugs.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Scrambling and quantum chaos indicators from long-time properties of operator distributions

Scrambling is a key concept in the analysis of nonequilibrium properties of quantum many-body systems. Most studies focus on its characterization via out-of-time-ordered correlation (OTOC) functions, particularly through the early-time decay of the OTOC. However, scrambling is a complex process which involves operator spreading and operator entanglement, and a full characterization requires one to access more refined information on the operator dynamics at several timescales. In this work we analyze operator scrambling by expanding the target operator in a complete basis and studying the structure of the expansion coefficients treated as a coarse-grained probability distribution in the space of operators. Here we study different features of this distribution, such as its mean, variance, and participation ratio, for the Ising model with longitudinal and transverse fields, kicked collective spin models, and random circuit models. We show that the long-time properties of the operator distribution display common features across these cases and discuss how these properties can be used as a proxy for the onset of quantum chaos. Finally, we discuss the connection with OTOCs and analyze the cost of probing the operator distribution experimentally using these correlation functions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2025

Three-dimensional analysis of microbunched electron cooling

Microbunched electron cooling (MBEC, Ratner, ), a special category of coherent electron cooling, has been under consideration as a candidate method for the strong hadron cooling component of the Electron Ion Collider (EIC) at BNL. Most of the theoretical treatments of this scheme so far have been based on simplified models for the key space charge effect in which electrons and/or hadrons are approximated by charged disks. Although such disk-based models provide a flexible tool that captures a good part of the underlying physics, they lack the fidelity and robustness of a rigorous analysis that explicitly treats the hadron and cooler electron beams as collections of point charges. In this paper, we present such a rigorous, three-dimensional (3D) model, both from a theory-based and a simulation perspective. Specifically, we calculate the generalized wakefield (or Green’s function) of the cooling system—a crucial quantity that determines the overall performance—paying special attention to the dependence of the wake on the hadron transverse positions. Additionally, our treatment takes into account previously neglected 3D effects that can cause damping of the plasma oscillations in the amplification section of the cooler, such as focusing and the angular spread of the electron beam. Published by the American Physical Society 2025

43 PARTICLE ACCELERATORS↗

DOE OSTI2022

Nature-based solutions in mountain catchments reduce impact of anthropogenic climate change on drought streamflow

Quantifying how well Nature-based Solutions can offset anthropogenic climate change impacts is important for adaptation planning, but has rarely been done. Here we show that a widely-applied Nature-based Solution in South Africa – invasive alien tree clearing – reduces the impact of anthropogenic climate change on drought streamflow. Using a multi-model joint-attribution of climate and landscape-vegetation states during the 2015–2017 Cape Town “Day Zero” drought, we find that anthropogenic climate change reduced streamflow by 12–29% relative to a counterfactual world with anthropogenic emissions removed. This impact on streamflow was larger than corresponding reductions in rainfall (7–15%) and reference evapotranspiration (1.7–2%). Clearing invasive alien trees could have ameliorated streamflow reductions by 3–16% points for moderate invasions levels. Preventing further invasive alien tree spread avoided potential additional reductions of 10–27% points. Total clearing could not have offset the anthropogenic climate change impact completely. Invasive alien tree clearing is an important form of catchment restoration for managing changing hydroclimatic risk, but will need to be combined with other adaptation options as climate change accelerates.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Identifying Genomic Islands with Deep Neural Networks

Background Horizontal gene transfer is the main source of adaptability for bacteria, through which genes are obtained from different sources including bacteria, archaea, viruses, and eukaryotes. This process promotes the rapid spread of genetic information across lineages, typically in the form of clusters of genes referred to as genomic islands (GIs). Different types of GIs exist, and are often classified by the content of their cargo genes or their means of integration and mobility. While various computational methods have been devised to detect different types of GIs, no single method is capable of detecting all types. Results We propose a method, which we call Shutter Island, that uses a deep learning model (Inception V3, widely used in computer vision) to detect genomic islands. The intrinsic value of deep learning methods lies in their ability to generalize. Via a technique called transfer learning, the model is pre-trained on a large generic dataset and then re-trained on images that we generate to represent genomic fragments. We demonstrate that this image-based approach generalizes better than the existing tools. Conclusions We used a deep neural network and an image-based approach to detect the most out of the correct GI predictions made by other tools, in addition to making novel GI predictions. The fact that the deep neural network was re-trained on only a limited number of GI datasets and then successfully generalized indicates that this approach could be applied to other problems in the field where data is still lacking or hard to curate.

Computer Vision↗

DOE OSTI2025

Influence of background sources and topographic resolution in the Weather Research and Forecasting Model on xenon plume characteristics at monitoring stations

For many atmospheric monitoring applications, networks of measurement sites—such as the radionuclide stations of the International Monitoring System—can be sparse. With measurement locations potentially hundreds to thousands of kilometers from a release it is important to quantify the effects of physical processes on transport and dispersion of plumes between source and measurement locations. This study addresses the effects of background sources and topography resolution near the release location of radionuclides. We use the Weather Research and Forecasting (WRF) model with inline chemistry to investigate (1) how an additional, time-varying source of 133 Xe, such as an operational medical isotope production facility, contributes to activity concentration measurements at monitoring sites, and (2) how complex topography influences on atmospheric conditions near emission sources impact plume concentrations at varying distances from the source. Two 133 Xe emission sources, including (1) a high flux rate of short duration representing an explosive event, and (2) a variable and continuous background source, are simulated. The continuous background source contributes significantly to total 133 Xe concentrations at several monitoring stations. Further, a WRF simulation at 9 km horizontal resolution is compared with a nested grid simulation, where the innermost domain has a resolution of 1 km. Increased topographic resolution leads to an improved representation of plume responses to local winds, with topographic influences greatest at locations closest to the sources. Differences between the two domain resolutions decrease at greater distances from the sources, as plumes have time to spread and mix and are influenced by synoptic scale circulation patterns that are represented similarly in both simulations.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Beyond the 3rd moment: a practical study of using lensing convergence CDFs for cosmology with DES Y3

ABSTRACT Widefield surveys probe clustered scalar fields – such as galaxy counts, lensing potential, etc. – which are sensitive to different cosmological and astrophysical processes. Constraining such processes depends on the statistics that summarize the field. We explore the cumulative distribution function (CDF) as a summary of the galaxy lensing convergence field. Using a suite of N-body light-cone simulations, we show the CDFs’ constraining power is modestly better than the second and third moments, as CDFs approximately capture information from all moments. We study the practical aspects of applying CDFs to data, using the Dark Energy Survey (DES Y3) data as an example, and compute the impact of different systematics on the CDFs. The contributions from the point spread function and reduced shear approximation are $\lesssim 1~{{\ \rm per\ cent}}$ of the total signal. Source clustering effects and baryon imprints contribute 1–10 per cent. Enforcing scale cuts to limit systematics-driven biases in parameter constraints degrade these constraints a noticeable amount, and this degradation is similar for the CDFs and the moments. We detect correlations between the observed convergence field and the shape noise field at 13σ. The non-Gaussian correlations in the noise field must be modelled accurately to use the CDFs, or other statistics sensitive to all moments, as a rigorous cosmology tool.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2026

Measuring X-Ray Emission Line Shapes in Neutral Species for XRISM Calibration

Space X-ray spectrometers such as the Resolve instrument on XRISM require precise calibration in order to interpret the spectra of astrophysical objects. Key components of the calibration are the energy scale and the core line spread function, both of which vary with photon energy. A major issue in the calibration of high-resolution spectrometers is locating good calibrators with well-known and stable intrinsic line shapes. Neutral fluorescence is widely used, but inner-shell transitions in neutral atoms often exhibit complex, poorly documented line shapes that vary with excitation conditions. Here, in this study, we present empirical measurements of K-shell transitions in neutral O and F below 1 keV using an engineering model XRISM calorimeter array, an electron bombardment modulated X-ray source, and an electron beam ion trap (EBIT) to provide a precise energy reference. In addition, we report measurements of the Mo Lα complex with the transition-edge microcalorimeter spectrometer (TEMS), which reveal strong satellite structure and sensitivity of the line shape to the incident exciting spectrum. Together, these results demonstrate the need for empirical line-shape models, highlight the nonstationary nature of neutral fluorescence features, and define a path toward developing transfer standards for XRISM and future precision instruments such as Athena/X-IFU.

Astronomy and AstroPhysics↗

DOE OSTI2024

Accelerating FEM-Based Corrosion Predictions Using Machine Learning

Atmospheric corrosion of metallic parts is a widespread materials degradation phenomena that is challenging to predict given its dependence on many factors (e.g. environmental, physiochemical, and part geometry). For materials with long expected service lives, accurately predicting the degree to which corrosion will degrade part performance is especially difficult due to the stochastic nature of corrosion damage spread across years or decades of service. The Finite Element Method (FEM) is a computational technique capable of providing accurate estimates of corrosion rate by numerically solving complex differential Eqs. characterizing this phenomena. Nevertheless, given the iterative nature of FEM and the computational expense required to solve these complex equations, FEM is ill-equipped for an efficient exploration of the design space to identify factors that accelerate or deter corrosion, despite its accuracy. In this work, a machine learning based surrogate model capable of providing accurate predictions of corrosion with significant computational savings is introduced. Specifically, this work leverages AdaBoosted Decision trees to provide an accurate estimate of corrosion current per width given different values of temperature, water layer thickness, molarity of the solution, and the length of the cathode for a galvanic couple of aluminum and stainless steel.

36 MATERIALS SCIENCE↗

DOE OSTI2021

Transient cycling of nitrogen, organic carbon and oxygen within the free-flowing Columbia River corridor: Linking exposure time dependent biogeochemical reactions to river stage fluctuations (Final project report)

The objective of this project was to develop next-generation techniques for representing the transformations of complex reaction networks like those impacting transient river-corridors, then apply them to develop upscaling tools. The project contributed significantly to the development of Lagrangian “mass transfer particle tracking” (MTPT) tools that are the first numerical framework that explicitly separates mixing and spreading processes; a crucial distinction for accurate representations of reaction rates across scales. The MTPT approach uses an operator splitting scheme where physical transport processes are simulated using classical random walk methods and each particle is treated as a “container” that carries with it any number of chemical species. Mixing between particles is simulated using a colocation-probability based mass transfer kernel and reactions are evaluated on each particle after mixing. The method is stable, accurate, and also allows the explicit modeling of residence time distributions of the individual containers of mass. The MTPT scheme allowed us to explore several research questions related to reactive transport in river corridors and complex reaction networks. We found that it is possible to quantify the exposure time distributions (ETDs) of reactants and that this does suggest some pathways to upscaling. However, comparable approximations can be obtained using simplified (reduced dimensionality) MTPT simulations in less time with similar uncertainty, so development of extensive ETD-based methods was not productive. The major findings are that i) the MTPT schemes are robust and highly accurate across a range of arbitrarily complex reactions, ii) the methods can be efficiently parallelized and the parallel performance characteristics of MTPT are predictable, and iii) the MTPT tools allow simultaneous tracking of residence time. The development of these research tools into efficient software packages continues. The methods are already available to the community because open-source, working examples have been included with all publications.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Artificial Intelligence/Machine Learning Technologies for Advanced Reactors (Workshop Summary Report)

A workshop on artificial intelligence and machine learning (AI/ML) for advanced reactors (AR) was held October 5-6, 2021. The workshop was to be attended in-person at ANL but COVID restrictions forced the workshop to go virtual. The objectives of the workshop were to identify the most promising AI/ML opportunities for improving advanced reactor design, optimizing plant performance, and enhancing economic competitiveness and to develop an understanding of the scientific, engineering and licensing challenges facing their application. The workshop planning committee included GAIN, EPRI and NEI and members of three national laboratories (ANL, INL, and ORNL). The workshop was attended by more than 200 individuals representing academic and scientific institutions and the nuclear power industry. The definition put forth for an AI/ML system was one that perceives its environment and takes actions that maximize its chance of achieving its goals. In this report AI/ML refers to next generation algorithms that include deep learning, statistical analysis and data analytics and associated scientific computing and their potential application to the design, licensing, operation and maintenance of ARs. These methods typically incorporate models built from process data and may also include data generated by simulations that represent the behavior of a system. The workshop was organized in response to the growing interest in application of AI/ML for improving the economic competitiveness of nuclear energy. Increasingly more resources are being allocated to investigating the benefits of AI/ML methods. The DOE created the Artificial Intelligence & Technology Office to promote their development. And within the Office of Nuclear Energy, resources have been allocated to explore and understand the potential benefits of AI/ML. Additionally, the national laboratories are strategically positioned with DOE computing facilities such as Summit, Perlmutter, Aurora and Frontier that support large-scale simulations, hybrid HPC models with AI surrogates, and the exploration of new types of generative models emerging from multi-model data streams and sources. The workshop was organized with members of the AR community to understand the effort and to identify the level of interest and progress in this emerging technology. The workshop discussions focused on identifying opportunities for AI/ML across diverse areas of the nuclear industry and identifying current scientific and engineering challenges for advanced reactors that might be addressed through transformational uses of AI/ML. Discussion panels focused on four high-interest technical domains for advanced reactors: design, maintenance and operations, energy storage, and materials. The results of those discussions are summarized in this report. This includes opportunities that were identified for exploiting AI techniques and methods to improve the efficacy and efficiency of reactor analysis and to improve the operation and optimization of advanced reactors. Advanced reactor developers expressed an interest in learning more about AI/ML methods and their application. This included understanding whether ML methods can provide an advantage over existing nonlinear data regression methods for collapsing high-fidelity simulation results into faster running models. A consensus emerged that AR advances planned for the next decade will benefit from the use of AI/ML tools. The need exists to understand and model complex systems across length scales and modalities. AI/ML is a tool for discovery that can yield a set of engineering principles for use by nuclear engineers, licensing bodies, and operators to solve problems in plant design, safety analyses, autonomous operation, and predictive maintenance. While AI/ML represents a new set of tools, an awareness by the nuclear community of the full potential is still in the early stages so there is a need to increase awareness. It appears that the wide-spread adoption of AI/ML tools for ARs would be facilitated by future educational workshops that describe foundational methods and capabilities and describe successful applications.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2020

A cancellation problem in hybrid particle-in-cell schemes due to finite particle size

The quasi-neutral hybrid particle-in-cell algorithm with kinetic ions and fluid electrons is a popular model to study multi-scale problems in laboratory, space, and astrophysical plasmas. Here, it is shown that the different spatial discretizations of ions as finite-spatial-size particles and electrons as a grid-based fluid can lead to significant numerical wave dispersion errors in the long wavelength limit (kd i «1, where k is the wavenumber and di is the ion skin-depth). The problem occurs when high-order particle-grid interpolations, or grid-based smoothing, spreads the electric field experienced by the ions across multiple spatial cells and leads to inexact cancellation of electric field terms in the total (ion + electron) momentum equation. Practical requirements on the mesh spacing Δx/d i are suggested to bound these errors from above. The accuracy impact of not respecting these resolution constraints is shown for a non-linear shock problem.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2024

Data Imbalance, Uncertainty Quantification, and Transfer Learning in Data‐Driven Parameterizations: Lessons From the Emulation of Gravity Wave Momentum Transport in WACCM

Abstract Neural networks (NNs) are increasingly used for data‐driven subgrid‐scale parameterizations in weather and climate models. While NNs are powerful tools for learning complex non‐linear relationships from data, there are several challenges in using them for parameterizations. Three of these challenges are (a) data imbalance related to learning rare, often large‐amplitude, samples; (b) uncertainty quantification (UQ) of the predictions to provide an accuracy indicator; and (c) generalization to other climates, for example, those with different radiative forcings. Here, we examine the performance of methods for addressing these challenges using NN‐based emulators of the Whole Atmosphere Community Climate Model (WACCM) physics‐based gravity wave (GW) parameterizations as a test case. WACCM has complex, state‐of‐the‐art parameterizations for orography‐, convection‐, and front‐driven GWs. Convection‐ and orography‐driven GWs have significant data imbalance due to the absence of convection or orography in most grid points. We address data imbalance using resampling and/or weighted loss functions, enabling the successful emulation of parameterizations for all three sources. We demonstrate that three UQ methods (Bayesian NNs, variational auto‐encoders, and dropouts) provide ensemble spreads that correspond to accuracy during testing, offering criteria for identifying when an NN gives inaccurate predictions. Finally, we show that the accuracy of these NNs decreases for a warmer climate (4 × CO 2 ). However, their performance is significantly improved by applying transfer learning, for example, re‐training only one layer using ∼1% new data from the warmer climate. The findings of this study offer insights for developing reliable and generalizable data‐driven parameterizations for various processes, including (but not limited to) GWs.

54 ENVIRONMENTAL SCIENCES↗