SEARCH · Engineering Papers

Results for “Big data processing”

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 307 records · Page 17

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Automatic Search of Cataclysmic Variables Based on LightGBM in LAMOST-DR7

The search for special and rare celestial objects has always played an important role in astronomy. Cataclysmic Variables (CVs) are special and rare binary systems with accretion disks. Most CVs are in the quiescent period, and their spectra have the emission lines of Balmer series, HeI, and HeII. A few CVs in the outburst period have the absorption lines of Balmer series. Owing to the scarcity of numbers, expanding the spectral data of CVs is of positive significance for studying the formation of accretion disks and the evolution of binary star system models. At present, the research for astronomical spectra has entered the era of Big Data. The Large Sky Area Multi-Object Fiber Spectroscopy Telescope (LAMOST) has produced more than tens of millions of spectral data. the latest released LAMOST-DR7 includes 10.6 million low-resolution spectral data in 4926 sky regions, providing ideal data support for searching CV candidates. To process and analyze the massive amounts of spectral data, this study employed the Light Gradient Boosting Machine (LightGBM) algorithm, which is based on the ensemble tree model to automatically conduct the search in LAMOST-DR7. Finally, 225 CV candidates were found and four new CV candidates were verified by SIMBAD and published catalogs. This study also built the Gradient Boosting Decision Tree (GBDT), Adaptive Boosting (AdaBoost), and eXtreme Gradient Boosting (XGBoost) models and used Accuracy, Precision, Recall, the F1-score, and the ROC curve to compare the four models comprehensively. Experimental results showed that LightGBM is more efficient. The search for CVs based on LightGBM not only enriches the existing CV spectral library, but also provides a reference for the data mining of other rare celestial objects in massive spectral data.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2015

Technical Challenges and Opportunities of Centralizing Space Science Mission Operations (SSMO) at NASA Goddard Space Flight Center

The NASA Goddard Space Science Mission Operations project (SSMO) is performing a technical cost-benefit analysis for centralizing and consolidating operations of a diverse set of missions into a unified and integrated technical infrastructure. The presentation will focus on the notion of normalizing spacecraft operations processes, workflows, and tools. It will also show the processes of creating a standardized open architecture, creating common security models and implementations, interfaces, services, automations, notifications, alerts, logging, publish, subscribe and middleware capabilities. The presentation will also discuss how to leverage traditional capabilities, along with virtualization, cloud computing services, control groups and containers, and possibly Big Data concepts.

Science↗

DOE OSTI2017

Open Source Release of NJOY2016 and NJOY21

NJOY has been the standard for nuclear data processing for many decades. NJOY is a comprehensive computer code used to produce point-wise and multigroup nuclear data application libraries from cross sections and related quantities obtained from evaluated nuclear data in the Evaluated Nuclear Data Format (ENDF). NJOY was first publicly released in 1977 and distributed by what is now the Radiation Safety Information Computational Center (RSICC) at Oak Ridge National Laboratory and the National Energy Software Center (NESC) at Argonne National Laboratory. From that time until the release of NJOY2012, NJOY has been distributed by RSICC and made available to researchers outside the United States by agreements between RSICC and the NEA Databank. NJOY2012 was distributed by the Feynman Center at Los Alamos National Laboratory which did not have agreements with the NEA Databank, thus making it more difficult for non US citizens to obtain copies of NJOY2012. Furthermore, NJOY has had export control restrictions preventing some researchers from obtaining NJOY at all. Two new versions of NJOY have been released, NJOY2016 and NJOY21, in January of 2017. Both NJOY2016 and NJOY21 have been released with no export control restrictions and are distributed with an open-source license. This is a big departure from previous releases and impacts not only who has access to NJOY, but also how updates are made available to users and how collaboration is facilitated among interested developers. In addition to distributing NJOY2016 and NJOY21 with an open-source license, these two new versions provide new capabilities for NJOY which will maintain NJOY as the standard for nuclear data processing into the foreseeable future. We report in this summary about the changes that come with these two new versions of NJOY and how this impacts the users and long-time contributors to NJOY.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

NASA NTRS2024

Digital Lunar Exploration Sites (DLES) Terrain Crafting

Humans will soon be returning to the surface of the Moon with NASA’s Artemis program. The Artemis program is an international collaboration that will consist of a complex series of space systems and missions to explore the lunar surface and pave the way for the future exploration of Mars. NASA and its partners rely heavily on simulation for lighting and navigation studies as well as training astronauts, flight controllers, and mission support staff. The NASA Exploration Systems Simulations (NExSyS) team in the Simulation and Graphics Branch (ER7) in the Engineering Directorate at NASA’s Johnson Space Center has built up many simulation products to support this effort, one of which is the Digital Lunar Exploration Sites (DLES). DLES is a collection of products used to simulate and render the lunar surface in a digital environment. We discussed and presented an overview of the DLES products at the 2022 IEEE Aerospace Conference in Big Sky, MT with a paper titled "Digital Lunar Exploration Sites". This “DLES Terrain Crafting” paper will expand on the information previously provided in “DLES” paper and dive deeper into the details of the terrain crafting process and the toolsets used to support this task. The best digital data currently available of the lunar surface is provided by the Lunar Reconnaissance Orbiter (LRO). Its Lunar Orbiter Laser Altimeter (LOLA) achieves an impressive resolution of 5m per pixel at the Lunar South Pole (LSP) and can generate datasets covering a large continuous region near the LSP. There are a few additional methods, such as Shape from Shading which can infer higher resolution data (up to 1m per pixel) from the LRO Narrow Angle Camera (NAC) images. However, surface-based simulations require higher-resolution data, and this paper will discuss the process of enhancing the terrain to meet that need. The process begins with capturing statistical data of craters in the regions of interest using images provided by the LRO NAC. This data is then used to scatter artificial features which are not captured in the truth data, resulting in an enhanced DEM with a much higher resolution of 20cm per pixel. Many tools were built up to assist in the creation of these artificial Digital Elevation Models (DEM), which this paper will discuss in detail. DEMs themselves are a very powerful representation of a planetary surface, and many operations and tools can utilize the data they contain. This paper includes a description of the rendering of the lunar surface in a graphics engine, generation of contact patches to simulate tire to ground interaction, and ray tracing utilities to model Line of Sight (LOS) interactions with the terrain. This paper will also explore some new tool sets currently under development which aim to utilize Machine Learning (ML) to assist in the identification of craters from LRO NAC imagery. While this is not a novel idea, the NExSyS team is developing a unique approach which may result in more robust identification of crater characteristics.

Artemis↗

DOE OSTI2020

Next-generation geospatial-temporal information technologies for disaster management

Traditional geographic information systems (GIS) have been disrupted by the emergence of Big Data in the form of geo-coded raster, vector, and time-series Internet-of-Things data. This article discusses the application of new scalable technologies that go far beyond relational databases and file-based storage on spinning disk or tape to incorporate both storage and processing data in the same platform. The roles of the Apache Hadoop Distributed File Systems and NoSQL key-value stores such as the Apache Hbase are discussed, along with indexing schemes that optimally support geospatial-temporal use. Here, we highlight how this new approach can rapidly search multiple GIS data layers to obtain insights in the context of early warning, impact evaluation, response, and recovery to earthquake and wildfire disasters.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2019

Realized ecological forecast through an interactive Ecological Platform for Assimilating Data (EcoPAD, v1.0) into models

Abstract. Predicting future changes in ecosystem services is not only highly desirable but is also becoming feasible as several forces (e.g., available big data, developed data assimilation (DA) techniques, and advanced cyber-infrastructure) are converging to transform ecological research into quantitative forecasting. To realize ecological forecasting, we have developed an Ecological Platform for Assimilating Data (EcoPAD, v1.0) into models. EcoPAD (v1.0) is a web-based software system that automates data transfer and processing from sensor networks to ecological forecasting through data management, model simulation, data assimilation, forecasting, and visualization. It facilitates interactive data–model integration from which the model is recursively improved through updated data while data are systematically refined under the guidance of model. EcoPAD (v1.0) relies on data from observations, process-oriented models, DA techniques, and the web-based workflow. We applied EcoPAD (v1.0) to the Spruce and Peatland Responses Under Climatic and Environmental change (SPRUCE) experiment in northern Minnesota. The EcoPAD-SPRUCE realizes fully automated data transfer, feeds meteorological data to drive model simulations, assimilates both manually measured and automated sensor data into the Terrestrial ECOsystem (TECO) model, and recursively forecasts the responses of various biophysical and biogeochemical processes to five temperature and two CO 2 treatments in near-real time (weekly). Forecasting with EcoPAD-SPRUCE has revealed that mismatches in forecasting carbon pool dynamics are more related to model (e.g., model structure, parameter, and initial value) than forcing variables, opposite to forecasting flux variables. EcoPAD-SPRUCE quantified acclimations of methane production in response to warming treatments through shifted posterior distributions of the CH 4 :CO 2 ratio and the temperature sensitivity ( Q 10 ) of methane production towards lower values. Different case studies indicated that realistic forecasting of carbon dynamics relies on appropriate model structure, correct parameterization, and accurate external forcing. Moreover, EcoPAD-SPRUCE stimulated active feedbacks between experimenters and modelers to identify model components to be improved and additional measurements to be taken. It has become an interactive model–experiment (ModEx) system and opens a novel avenue for interactive dialogue between modelers and experimenters. Altogether, EcoPAD (v1.0) acts to integrate multiple sources of information and knowledge to best inform ecological forecasting.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRSDate not supplied

Modern Scientific Data Governance Framework

Science has entered the era of Big Data with new challenges related to data governance, stewardship, and management. The existing data governance practices must catch up to ensure proper data management. Existing data governance policies and stewardship best practices tend to be disconnected from operational data management practices and enforcement and mainly exist in well-meaning documents or reports. These governance policies are, at best, partially implemented and rarely monitored or audited. In addition, existing governance policies keep adding additional data management steps that require a human, ‘a data steward’, in the loop, and the cost of data management can no longer scale proportionately with the current and future increased data volume and complexity. The goal for developing an updated data governance framework is to modernize scientific data governance to the reality of Big data and align it with the current technology trends such as cloud computing and AI. The goals of this framework are two folds. One is to ensure thoroughness that the governance adequately covers the entire data life cycle. Two, provide a practical approach that offers a consistent and repeatable process for different projects. Three core principles ground this framework. First, focus on just enough governance and prevent data governance from becoming a roadblock toward the scientific process. Remove any unnecessary processes and steps. Second, automate data management steps where possible. Actively remove steps that require ‘human in the loop’ within the management process to be efficient and scale with increasing data. Third, all the processes should continually be optimized using quantified metrics to streamline the monitoring and auditing workflows.

Rahul Ramachandran↗

DOE OSTI2018

PySFD: comprehensive molecular insights from significant feature differences detected among many simulated ensembles

Motivation Many modeling analyses of molecular dynamics (MD) simulations are based on a definition of states that can be (groups of) clusters of simulation frames in a feature space composed of molecular coordinates. With increasing dimension of this feature space (due to the increasing size or complexity of a simulated molecule), it becomes very difficult to cluster the underlying MD data and estimate a statistically robust model. To mitigate this “curse of dimensionality”, one can reduce the feature space, e.g., with principal component or time-lagged independent component analysis transformations, focusing the analysis on the most important modes of transitions. In practice, however, all these reduction strategies may neglect important molecular details that are susceptible to experimental verification. Results To recover such molecular details, I have developed PySFD (Significant Feature Differences analyzer for Python), a multi-processing software package that efficiently selects significantly different features of any user-defined feature type among potentially many different simulated state ensembles, such as meta-stable states of a Markov State Model (MSM). Applying PySFD on MSMs of an aggregate of 300 microseconds MD simulations recently performed on the major histocompatibility complex class II (MHCII) protein, I demonstrate how this toolkit can extract and visualize valuable mechanistic information from big MD simulation data, e.g., in form of networks of dynamic interaction changes connecting functionally relevant sites of a protein complex.

Biochemistry & Molecular Biology↗

DOE OSTI2016

Integration Of PanDA Workload Management System With Supercomputers for ATLAS and Data Intensive Science

The.LHC, operating at CERN, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 150 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more than 250,000 cores with a peak performance of 0.3 petaFLOPS, LHC data taking runs require more resources than grid can possibly provide. To alleviate these challenges, LHC experiments are engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with supercomputers in United States, in particular with Titan supercomputer at Oak Ridge Leadership Computing Facility. Current approach utilizes modified PanDA pilot framework for job submission to the supercomputers batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on LCFs multi-core worker nodes. This implementation was tested with a variety of Monte-Carlo workloads on several supercomputing platforms for ALICE and ATLAS experiments and it is in full pro duction for the ATLAS since September 2015. We will present our current accomplishments with running PanDA at supercomputers and demonstrate our ability to use PanDA as a portal independent of the computing facilities infrastructure for High Energy and Nuclear Physics as well as other data-intensive science applications, such as bioinformatics and astro-particle physics.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

A network of soil moisture, soil temperature, air temperature, net radiation, ground heat flux and ground water for Chicago, Illinois

This dataset contains environmental monitoring data collected using solar-powered Multi-Function Research (MFR) Long Range Wide Area (LoRaWAN)-enabled nodes at 11 sites in Chicago, Illinois, as part of the DOE Urban Integrated Field Lab CROCUS project. The MFR node system consists of an Input/Output Digital Input Module (IB8) interface box (ICT International) providing wired connections for environmental sensors and an MFR-Node-L data logger that manages power, data processing, and LoRaWAN communication. The wireless data are ingested via Sage network (https://sagecontinuum.org/) nodes that contain LoRaWAN antennae. Measurements were collected from 11 MFR nodes deployed across Chicago State University (CSU), Northeastern Illinois University (NEIU), Northwestern University (NU), University of Illinois Chicago (UIC), West Woodlawn "Blacks in Green" (BIG), and Indian Boundary Prairies (IBP). Each MFR node supports a consistent suite of sensors measuring atmospheric, soil, and hydrological variables. Atmospheric measurements include 2m air temperature (°C), 2m vapor pressure deficit (kPa), and 2m shortwave/longwave radiation (incoming and outgoing, W/m²) measured using ATH-VPD and Apogee SN500 sensors. Soil measurements include volumetric water content (VWC, %) and temperature (°C) at four depths (15, 30, 45, and 60 cm below surface) using Meter Teros54 sensors, and heat flux (W/m²) at 10 cm depth using Huske HFP01-05 sensors. At selected locations, Meter Hydros21 sensors measure groundwater depth (mm), specific conductivity (dS/m), and temperature (°C). The dataset includes timestamps, site identifiers with location names, device IDs, Global Positioning System (GPS) coordinates, variable names with units, measurement depths, values, sensor names, and Sage node identifiers. All timestamps are in local Chicago time (CDT/CST). Quality control flags are provided using a 6-bit binary system indicating physical range violations, step spikes, 24-hour flat-line conditions, 6-hour jitter, 7-day ultra-low variance, and persistent high offset. Data is provided in CSV and CF-compliant NetCDF formats. This dataset is part of a larger collection of CROCUS environmental monitoring data, including linked datasets from Air Quality Transmitter (AQT) sensors, Weather Transmitter (WXT) sensors, and Sap Flow Meter (SFM1x) sensors.

Chicago↗

NASA NTRS1974

The origin of the diffuse background gamma radiation

Recent observations provided evidence for diffuse background gamma radiation extending to energies beyond 100 MeV, and evidence of isotropy and implied cosmological origin. Significant features in the spectrum of this background radiation were observed which provide evidence for its origin in nuclear processes in the early stages of big-bang cosmology, and connect these processes with the galaxy formation theory. A test of the theory is in future observations of the background radiation in the 100 MeK to 100 GeV energy range which are made with large orbiting spark-chamber satellite detectors. The theoretical interpretations of present data, their connection with baryon-symmetric cosmology and galaxy formation theory, and the need for future observations are discussed.

Stecker, F. W.↗

DOE OSTI2017

A New Method for Generating Probability Tables in the Unresolved Resonance Region

One new method for constructing probability tables in the unresolved resonance region (URR) has been developed. This new methodology is an extensive modification of the single-level Breit-Wigner (SLBW) pseudo-resonance pair sequence method commonly used to generate probability tables in the URR. The new method uses a Monte Carlo process to generate many pseudo-resonance sequences by first sampling the average resonance parameter data in the URR and then converting the sampled resonance parameters to the more robust R-matrix limited (RML) format. Furthermore, for each sampled set of pseudo-resonance sequences, the temperature-dependent cross sections are reconstructed on a small grid around the energy of reference using the Reich-Moore formalism and the Leal-Hwang Doppler broadening methodology. We then use the effective cross sections calculated at the energies of reference to construct probability tables in the URR. The RML cross-section reconstruction algorithm has been rigorously tested for a variety of isotopes, including 16 O, 19 F, 35 Cl, 56 Fe, 63 Cu, and 65 Cu. The new URR method also produced normalized cross-section factor probability tables for 238 U that were found to be in agreement with current standards. The modified 238 U probability tables were shown to produce results in excellent agreement with several standard benchmarks, including the IEU-MET-FAST-007 (BIG TEN), IEU-MET-FAST-003, and IEU-COMP-FAST-004 benchmarks.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2017

Enabling Diverse Software Stacks on Supercomputers using High Performance Virtual Clusters.

While large-scale simulations have been the hallmark of the High Performance Computing (HPC) community for decades, Large Scale Data Analytics (LSDA) workloads are gaining attention within the scientific community not only as a processing component to large HPC simulations, but also as standalone scientific tools for knowledge discovery. With the path towards Exascale, new HPC runtime systems are also emerging in a way that differs from classical distributed com- puting models. However, system software for such capabilities on the latest extreme-scale DOE supercomputing needs to be enhanced to more appropriately support these types of emerging soft- ware ecosystems. In this paper, we propose the use of Virtual Clusters on advanced supercomputing resources to enable systems to support not only HPC workloads, but also emerging big data stacks. Specifi- cally, we have deployed the KVM hypervisor within Cray's Compute Node Linux on a XC-series supercomputer testbed. We also use libvirt and QEMU to manage and provision VMs directly on compute nodes, leveraging Ethernet-over-Aries network emulation. To our knowledge, this is the first known use of KVM on a true MPP supercomputer. We investigate the overhead our solution using HPC benchmarks, both evaluating single-node performance as well as weak scaling of a 32-node virtual cluster. Overall, we find single node performance of our solution using KVM on a Cray is very efficient with near-native performance. However overhead increases by up to 20% as virtual cluster size increases, due to limitations of the Ethernet-over-Aries bridged network. Furthermore, we deploy Apache Spark with large data analysis workloads in a Virtual Cluster, ef- fectively demonstrating how diverse software ecosystems can be supported by High Performance Virtual Clusters.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

A COLD FLARE WITH DELAYED HEATING

Recently, a number of peculiar flares have been reported that demonstrate significant nonthermal particle signatures with low, if any, thermal emission, which implies a close association of the observed emission with the primary energy release/electron acceleration region. This paper presents a flare that appears “cold” at the impulsive phase, while displaying delayed heating later on. Using hard X-ray data from Konus- Wind , microwave observations by SSRT, RSTN, NoRH, and NoRP, context observations, and three-dimensional modeling, we study the energy release, particle acceleration, and transport, and the relationships between the nonthermal and thermal signatures. The flaring process is found to involve the interaction between a small loop and a big loop with the accelerated particles divided roughly equally between them. Precipitation of the electrons from the small loop produced only a weak thermal response because the loop volume was small, while the electrons trapped in the big loop lost most of their energy in the coronal part of the loop, which resulted in coronal plasma heating but no or only weak chromospheric evaporation, and thus unusually weak soft X-ray emission. The energy losses of the fast electrons in the big tenuous loop were slow, which resulted in the observed delay of the plasma heating. We determined that the impulsively accelerated electron population had a beamed angular distribution in the direction of the electric force along the magnetic field of the small loop. The accelerated particle transport in the big loop was primarily mediated by turbulent waves, which is similar to other reported cold flares.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2014

Final Report - Phylogenomic tools and web resources for the Systems Biology Knowledgebase

The major advance during this last reporting period (8/15/12 to present) is our release of data on the PhyloFacts website: phylogenetic trees, multiple sequence alignments and other data for protein families are now available for download from http://phylogenomics.berkeley.edu/data/. This project as a whole aimed to develop high-throughput functional annotation systems that exploit information from protein 3D structure and evolution to provide highly precise inferences of various aspects of gene function, including molecular function, biological process, pathway association, Pfam domains, cellular localization and so on. We accomplished these aims by developing and testing different systems on a database of protein family trees: the PhyloFacts Phylogenomic Encyclopedia (at http://phylogenomics.berkeley.edu/phylofacts/ ).

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

Interactive Visualization of Multi-Group Cross Sections on High-Fidelity Spatial Meshes

Core simulation tools utilizing Monte Carlo transport methods represent the 'Holy Grail' for nuclear reactor analysis. However, deterministic transport methods pose a number of advantages to Monte Carlo from a computational perspective, which seems likely to make these methods more viable in the near to intermediate term. Deterministic methods almost universally discretize the energy domain in a few to hundreds of energy groups. One of the key challenges for deterministic transport-based methods to be viable is accurate, reactor agnostic multi-group cross section generation. This paper describes tools for data processing and visualization of multigroup cross sections generated on high-fidelity spatial meshes from the OpenMC Monte Carlo code. Next generation deterministic transport reactor core modeling tools will require multi-group cross section (MGXS) generation on high-fidelity spatial meshes. High-fidelity MGXS are necessary for the accurate simulation of neutron physics in nuclear reactors with a high degree of spectral heterogeneity - e.g., large localized neutron flux gradients throughout a reactor core. Although many local (e.g., pin cell) approximations have been developed to model the scalar flux to compute spectral-averaged MGXS, these techniques cannot account for complex intra-pin and intra-assembly spatial self-shielding effects. Monte Carlo presents the most accurate method for reactor agnostic multi-group cross section generation since it does not require the use of local approximations to the flux. Monte Carlo has been proposed as a highly desirable framework for reactor agnostic cross section generation for many years. Monte Carlo methods can effectively embed all of the relevant spectral effects into the MGXS through the random particle tracking and interaction sampling processes. Previous work has applied MCNP and Serpent to compute few-group constants for diffusion calculations, and OpenMC to compute multi-group cross sections for transport-based methods. The primary difficulty with using Monte Carlo for MGXS generation is the tremendous computational resources required to adequately converge multi-group cross sections in each spatial region and energy group. Recent development of the OpenMC Monte Carlo code for high-performance computing platforms has made multigroup cross section generation from Monte Carlo more tractable in the foreseeable future. The recent release of OpenMC v0.7.1 includes a new Python-based module for MGXS generation. The MGXS generation module leverages many of the algorithms and features for 'big data' analysis of large tally datasets on complex spatial meshes. This paper describes the data management workflow and visualization tools which have been developed to explore the plethora of MGXS data generated with OpenMC. It is the authors' hope that the development of the visualization tool described in this paper will ultimately enable the design of machine learning algorithms to accelerate the convergence rate of high-fidelity multi-group cross section generated with Monte Carlo. (authors)

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2017

Plant Uptake of Atmospheric Carbonyl Sulfide in Coast Redwood Forests: Carbonyl Sulfide Sink in Coast Redwoods

The future resilience of coast redwoods (Sequoia sempervirens) is now of critical concern due to the detection of a 33% decline in California coastal fog over the 20th century. However, ecosystem-scale measurements of photosynthesis and stomatal conductance are challenging in coast redwood forests, making it difficult to anticipate the impacts of future changes in fog. To address this methodological problem, we explore coastal variations in atmospheric carbonyl sulfide (COS or OCS), which could potentially be used as a tracer of these ecosystem processes. We conducted atmospheric flask campaigns in coast redwood sites, sampling at surface heights and in the canopy (~70 m), at the University of California Landels-Hill Big Creek Reserve and Big Basin State Park. We simulated COS atmosphere-biosphere exchange with a high-resolution 3-D model to interpret these data. Flask measurements indicated a persistent daytime drawdown between the coast and the downwind forest (45 ± 6 ppt COS) that is consistent with the expected relationship between COS plant uptake, stomatal conductance, and gross primary production. Other sources and sinks of COS that could introduce noise to the COS tracer technique (soils, anthropogenic activity, nocturnal plant uptake, and surface hydrolysis on leaves) are likely to be small relative to daytime COS plant uptake. These results suggest that COS measurements may be useful for making ecosystem-scale estimates of carbon, water, and energy exchange in coast redwood forests.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2022

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space–time domain. Such PIML integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

42 ENGINEERING↗