Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Building Dataset”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 937 records · Page 52

A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers

Current Deep Learning models use highly optimized convolutional neural networks (CNN) trained on large graphical processing units (GPU)-based computers with a fairly simple layered network topology, i.e., highly connected layers, without intra-layer connections. Complex topologies have been proposed, but are intractable to train on current systems. Building the topologies of the deep learning network requires hand tuning, and implementing the network in hardware is expensive in both cost and power. In this paper, we evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing (HPC) to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation. Due to input size limitations of current quantum computers we use the MNIST dataset for our evaluation. The results show the possibility of using the three architectures in tandem to explore complex deep learning networks that are untrainable using a von Neumann architecture. We show that a quantum computer can find high quality values of intra-layer connections and weights, while yielding a tractable time result as the complexity of the network increases; a high performance computer can find optimal layer-based topologies; and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware. This represents a new capability that is not feasible with current von Neumann architecture. It potentially enables the ability to solve very complicated problems unsolvable with current computing technologies.

Potok, Thomas E↗

DNABERT-S: pioneering species differentiation with species-aware DNA embeddings

SUMMARY: We introduce DNABERT-S, a tailored genome model that develops species-aware embeddings to naturally cluster and segregate DNA sequences of different species in the embedding space. Differentiating species from genomic sequences (i.e. DNA and RNA) is vital yet challenging, since many real-world species remain uncharacterized, lacking known genomes for reference. Embedding-based methods are therefore used to differentiate species in an unsupervised manner. DNABERT-S builds upon a pre-trained genome foundation model named DNABERT-2. To encourage effective embeddings to error-prone long-read DNA sequences, we introduce Manifold Instance Mixup (MI-Mix), a contrastive objective that mixes the hidden representations of DNA sequences at randomly selected layers and trains the model to recognize and differentiate these mixed proportions at the output layer. We further enhance it with the proposed Curriculum Contrastive Learning (C2LR) strategy. Empirical results on 28 diverse datasets show DNABERT-S's effectiveness, especially in realistic label-scarce scenarios. For example, it identifies twice more species from a mixture of unlabeled genomic sequences, doubles the Adjusted Rand Index (ARI) in species clustering, and outperforms the top baseline's performance in 10-shot species classification with just a 2-shot training. AVAILABILITY AND IMPLEMENTATION: Model, codes, and data are publically available at https://github.com/MAGICS-LAB/DNABERT_S.

Zhou, Zhihan↗

An FPGA-Based Neuromorphic Processor with All-to-All Connectivity

Neuromorphic computing is a promising paradigm for future energy-efficient computing. At present, however, it is in its nascent stages—most hardware implementations are research-grade, commercial products are not available, and the software tools are not production-ready. The lack of hardware and software tools makes neuromorphic computing inaccessible to researchers around the globe. To this extent, we intend to build a low-cost, open-source, FPGA-based digital neuromorphic processor that can be used by researchers worldwide. In this paper, we present a preliminary implementation of the processor on a Xilinx Artix-7 FPGA using SystemVerilog. Our implementation supports the integrate-and-fire neuron with two parameters each for neurons and synapses. It also features all-to-all connectivity among neurons on the hardware. We test our implementation on four cases: bars and stripes datasets, shortest path algorithm, logic gates, and 8-3 encoder. We also perform a scalability study to understand the resource utilization of the FPGA as the number of all-to-all connected neurons increases. With our implementation, the Artix- 7 supports 65 neurons with all-to-all connectivity. Moreover, all the test cases mentioned above achieve 100% accuracy.

Maheshwari, Disha↗

IRNet: A General Purpose Deep Risidual Regression Framework for Materials Discovery

Materials discovery is crucial for making scientific advances in many domains. Collections of data from experiments and first-principle computations have spurred interest in applying machine learning methods to create predictive models capable of mapping from composition and crystal structures to materials properties. Generally, these are regression problems with the input being a 1D vector composed of numerical attributes representing the material composition and/or crystal structure. While neural networks consisting of fully connected layers have been applied to such problems, their performance often suffers from the vanishing gradient problem when network depth is increased. Hence, predictive modeling for such tasks has been mainly limited to traditional machine learning techniques such as Random Forest. In this paper, we study and propose design principles for building deep regression networks composed of fully connected layers with numerical vectors as input. We introduce a novel deep regression network with individual residual learning, IRNet, that places shortcut connections after each layer so that each layer learns the residual mapping between its output and input. We use the problem of learning properties of inorganic materials from numerical attributes derived from material composition and/or crystal structure to compare IRNet’s performance against that of other machine learning techniques. Using multiple datasets from the Open Quantum Materials Database (OQMD) and Materials Project for training and evaluation, we show that IRNet provides significantly better prediction performance than the state-of-the-art machine learning approaches currently used by domain scientists. We also show that IRNet’s use of individual residual learning leads to better convergence during the training phase than when shortcut connections are between multi-layer stacks while maintaining the same number of parameters.

deep learning↗

A Novel Segmentation Algorithm for the ARM User Facility All-Sky Imagers Using Machine Learning Applications

Cloud cover plays a pivotal role in modulating the Earth's energy budget through the reflection of incoming solar radiation and the trapping of outgoing longwave radiation. Ground-based all-sky imagers offer an objective assessment of cloud cover that can be used to estimate solar irradiance, classify cloud types, track cloud movement, and serve as a benchmark 10 for the evaluation of satellite and reanalysis data products. The Atmospheric Radiation Measurement (ARM) user facility has utilized all-sky imagers for more than 25 years to monitor cloud cover and augment its comprehensive suite of atmospheric measurements. Following the retirement of its Total Sky Imager (TSI), ARM recently deployed the TSI’s successor, the All Sky Imager (ASI-16 camera systems). To provide a smooth transition and continuity to the vast amount of knowledge gathered by the TSI over the years, while addressing typical deployment issues, we developed a novel pixel segmentation algorithm, 15 the ASI Sky Cover (ASISKYCOVER). ASISKYCOVER builds on the different strengths and properties of the TSI processing algorithm while integrating machine learning techniques, ensuring data validity and accuracy across diverse atmospheric conditions. It enhances cloud cover characterization with new features such as artifact detection and uncertainty quantification. ASISKYCOVER also includes cloud cover estimates for near-zenith (narrow field-of-view) and reduces susceptibility to false detections. This study introduces ASISKYCOVER, details its algorithm framework, and demonstrates its capabilities using a 20 year-long dataset from the ARM Southern Great Plains site. Comparisons with co-located TSI data and other ARM measurements, such as zenith-pointing radars and lidars, are presented, underscoring the ASISKYCOVER’s potential to improve cloud cover analyses and data evaluation efforts, as well as to be integrated into higher-level data products that synergize instrument suites to generate new and insightful information

Silber, Israel↗

Corrective Action Decision Document/Closure Report for Corrective Action Unit 560: Septic Systems, Nevada Test Site, Nevada, Revision 0 with ROTC 1

Corrective Action Unit 560 comprises seven corrective action sites (CASs): •03-51-01, Leach Pit •06-04-02, Septic Tank •06-05-03, Leach Pit •06-05-04, Leach Bed •06-59-03, Building CP-400 Septic System •06-59-04, Office Trailer Complex Sewage Pond •06-59-05, Control Point Septic System The purpose of this CADD/CR is to provide justification and documentation supporting the recommendation for closure of CAU 560 with no further corrective action. To achieve this, corrective action investigation (CAI) activities were performed from October 7, 2008, through February 24, 2010, as set forth in the Corrective Action Investigation Plan for Corrective Action Unit 560: Septic Systems, Nevada Test Site, Nevada, and Record of Technical Change No. 1. The purpose of the CAI was to fulfill the following data needs as defined during the data quality objective (DQO) process: •Determine whether contaminants of concern (COCs) are present. •If COCs are present, determine their nature and extent. •Provide sufficient information and data to complete appropriate corrective actions. The CAU 560 dataset from the investigation results was evaluated based on the data quality indicator parameters. This evaluation demonstrated the quality and acceptability of the dataset for use in fulfilling the DQO data needs. Analytes detected during the CAI were evaluated against final action levels (FALs) established in this document. The following contaminants were determined to be present at concentrations exceeding their corresponding FALs: •No contamination exceeding the FALs was identified at CASs 03-51-01, 06-04-02, and 06-59-04. •The soil at the base of the leach pit chamber at CAS 06-05-03 contains arsenic above the FAL of 23 milligrams per kilogram (mg/kg) and polychlorinated biphenyl (PCBs) above the FAL of 0.74 mg/kg, confined vertically from a depth of approximately 5 to 20 feet (ft) below ground surface. The contamination is confined laterally to the walls of the leach pit chamber and leach rock. The contamination present at CAS 06-05-03 within the leach pit was not feasible to remove. •The surface and subsurface soils within and surrounding the septic system at CAS 06-05-04 contained PCB concentrations above the FAL of 0.74 mg/kg. The lateral and vertical extent of COCs was determined for this CAS. Contaminated soils were removed up to within 18 ft of the building. The remaining contamination is confined to subsurface soils adjacent to and beneath Building CP-162 and was not feasible to remove. •The solid materials within the septic tank and soils immediately surrounding the inlet end of the tank at CAS 06-59-03 contained benzo(a)pyrene above the FAL of 0.21 mg/kg. The soils, tank contents, and tank were removed. Materials remaining at this CAS do not contain contamination exceeding FALs. •The solids contained within the septic tank and inlet pipe at CAS 06-59-05 contained the following contaminants above their respective FALs: PCBs, arsenic, lead, benzo(a)pyrene, and pesticides. The tank and inlet pipe contents were removed. Materials remaining at this CAS do not contain contamination exceeding FALs. Therefore, the U.S. Department of Energy, National Nuclear Security Administration Nevada Site Office (NNSA/NSO) provides the following corrective action recommendations: •No further action for CASs 03-51-01, 06-04-02, and 06-59-04, as no contaminants of potential concern were present that exceed FALs. •Closure in place for CAS 06-05-03 under a corrective action with a use restriction (UR) for remaining PCB- and arsenic-impacted potential source material (PSM). The UR form and map have been filed in the NNSA/NSO Facility Information Management System, the FFACO database, and NNSA/NSO CAU/CAS files. •Closure in place for CAS 06-05-04 under a corrective action with a UR for remaining PCBs in soil adjacent to and beneath Building CP-162. The UR form and map have been filed in the NNSA/NSO Facility Information Management System, the FFACO database, and NNSA/NSO CAU/CAS files. •No further action for CAS 06-59-03, as the COC of benzo(a)pyrene in soil and PSM have been removed. •No further action for CAS 06-59-05, as the COCs in PSM within the septic tank and inlet piping have been removed and the tank was filled with concrete.

54 ENVIRONMENTAL SCIENCES↗

Climatology of the Vertical Profiles of Polarimetric Radar Variables and Retrieved Microphysical Parameters in Continental/Tropical MCSs and Landfalling Hurricanes

Most existing cloud models tend to overestimate the size of cloud ice particles and underestimate their concentration. This emphasizes the need to provide a reliable observational reference to optimize cloud model performance, particularly in areas of high concentration of ice at high altitudes. The dual-polarization radars give the community a unique opportunity to quantify cloud ice with a good accuracy using polarimetric radar retrievals. Here, in this study, we utilize the network of operational WSR-88D radars to build a climatology of the vertical profiles of radar variables, such as radar reflectivity Z, differential reflectivity Z DR , and specific differential phase K DP as well as the radar-retrieved vertical profiles of ice water content (IWC) above the melting layer and liquid water content below it, mean volume diameter D m , and total number concentration N t of ice and liquid particles. Such climatology was created for continental/marine mesoscale convective systems (MCSs) and tropical cyclones including hurricanes. The dataset includes 13 continental MCSs, 10 marine MCSs, and 11 tropical cyclones. Separate statistics of the “background” vertical profiles and the ones associated with high IWC aloft have been obtained in the course of this study. It is shown that continental MCSs exhibit larger size of ice in lower concentration aloft compared to the marine MCSs and especially tropical cyclones/hurricanes. A combination of high KDP and low Z aloft signifies lower D m , higher N t , and often substantial IWC.

54 ENVIRONMENTAL SCIENCES↗

Long- and short-term temporal variability in cloud condensation nuclei spectra over a wide supersaturation range in the Southern Great Plains site

Abstract. When aerosol particles seed the formation of liquid water droplets in the atmosphere, they are called cloud condensation nuclei (CCN). Different aerosols will act as CCN under different degrees of water supersaturation (relative humidity above 100 %), depending on their size and composition. In this work, we build and analyze a best-estimate CCN spectrum product, tabulated at ∼ 45 min resolution, generated using high quality data from seven independent instruments at the U.S. Department of Energy Atmospheric Radiation Measurement (ARM) Southern Great Plains site. The data product spans a large supersaturation range, from 0.0001 % to ∼ 30 %, and time period of 5 years, from 2009–2013, and is available on the ARM data archive. We leverage this added statistical power to examine relationships that are unclear in smaller datasets. Our analysis is performed in three main areas. First, probability distributions of many aerosol and CCN metrics are found to exhibit skewed log-normal distribution shapes. Second, clustering analyses of CCN spectra reveal that the primary drivers of CCN differences are aerosol number size distributions, rather than hygroscopicity or composition, especially at supersaturations above 0.2 %, while also allowing for a simplified understanding of seasonal and diurnal variations in CCN behavior. The predictive ability of using limited hygroscopicity data with accurate number size distributions to estimate CCN spectra is investigated, and the uncertainties of this approach are estimated. Third, the dynamics of CCN spectral clusters and concentrations are examined with cross-correlation and autocorrelation analyses. We find that CCN concentrations change rapidly on the timescale of 1–3 h, with some conservation beyond that which is greatest for the lower supersaturation region of the spectrum.

54 ENVIRONMENTAL SCIENCES↗

Hyperspectral Fluorescence and Reflectance Imaging Instrument

The system is a single hyperspectral imaging instrument that has the unique capability to acquire both fluorescence and reflectance high-spatial-resolution data that is inherently spatially and spectrally registered. Potential uses of this instrument include plant stress monitoring, counterfeit document detection, biomedical imaging, forensic imaging, and general materials identification. Until now, reflectance and fluorescence spectral imaging have been performed by separate instruments. Neither a reflectance spectral image nor a fluorescence spectral image alone yields as much information about a target surface as does a combination of the two modalities. Before this system was developed, to benefit from this combination, analysts needed to perform time-consuming post-processing efforts to co-register the reflective and fluorescence information. With this instrument, the inherent spatial and spectral registration of the reflectance and fluorescence images minimizes the need for this post-processing step. The main challenge for this technology is to detect the fluorescence signal in the presence of a much stronger reflectance signal. To meet this challenge, the instrument modulates artificial light sources from ultraviolet through the visible to the near-infrared part of the spectrum; in this way, both the reflective and fluorescence signals can be measured through differencing processes to optimize fluorescence and reflectance spectra as needed. The main functional components of the instrument are a hyperspectral imager, an illumination system, and an image-plane scanner. The hyperspectral imager is a one-dimensional (line) imaging spectrometer that includes a spectrally dispersive element and a two-dimensional focal plane detector array. The spectral range of the current imaging spectrometer is between 400 to 1,000 nm, and the wavelength resolution is approximately 3 nm. The illumination system consists of narrowband blue, ultraviolet, and other discrete wavelength light-emitting-diode (LED) sources and white-light LED sources designed to produce consistently spatially stable light. White LEDs provide illumination for the measurement of reflectance spectra, while narrowband blue and UV LEDs are used to excite fluorescence. Each spectral type of LED can be turned on or off depending on the specific remote-sensing process being performed. Uniformity of illumination is achieved by using an array of LEDs and/or an integrating sphere or other diffusing surface. The image plane scanner uses a fore optic with a field of view large enough to provide an entire scan line on the image plane. It builds up a two-dimensional image in pushbroom fashion as the target is scanned across the image plane either by moving the object or moving the fore optic. For fluorescence detection, spectral filtering of a narrowband light illumination source is sometimes necessary to minimize the interference of the source spectrum wings with the fluorescence signal. Spectral filtering is achieved with optical interference filters and absorption glasses. This dual spectral imaging capability will enable the optimization of reflective, fluorescence, and fused datasets as well as a cost-effective design for multispectral imaging solutions. This system has been used in plant stress detection studies and in currency analysis.

Ryan, Robert E.↗

Building a Large Scale Climate Data System in Support of HPC Environment

The Earth System Grid Federation (ESG) is a large scale, multi-institutional, interdisciplinary project that aims to provide climate scientists and impact policy makers worldwide a web-based and client-based platform to publish, disseminate, compare and analyze ever increasing climate related data. This paper describes our practical experiences on the design, development and operation of such a system. In particular, we focus on the support of the data lifecycle from a high performance computing (HPC) perspective that is critical to the end-to-end scientific discovery process. We discuss three subjects that interconnect the consumer and producer of scientific datasets: (1) the motivations, complexities and solutions of deep storage access and sharing in a tightly controlled environment; (2) the importance of scalable and flexible data publication/population; and (3) high performance indexing and search of data with geospatial properties. These perceived corner issues collectively contributed to the overall user experience and proved to be as important as any other architectural design considerations. Although the requirements and challenges are rooted and discussed from a climate science domain context, we believe the architectural problems, ideas and solutions discussed in this paper are generally useful and applicable in a larger scope.

54 ENVIRONMENTAL SCIENCES↗

Inference of the linear matter power spectrum at z = 0 using DESI DR1 Full-Shape data

Measurements of galaxy distributions at large cosmic distances capture clustering from the past. In this study, we use a cosmological model to translate these observations into the present-day galaxy distribution. Specifically, we reconstruct the 3D linear matter power spectrum at redshift z = 0 using Dark Energy Spectroscopic Instrument (DESI) Year 1 (DR1) galaxy clustering data and Cosmic Microwave Background (CMB) observations, assuming the ΛCDM model, and compare it to the result assuming the w 0 w a CDM model. Building on previous state-of-the-art methods, we apply Effective Field Theory (EFT) modelling of the galaxy power spectrum to account for small-scale effects in the 2-point statistics of galaxy data. Implementation of the EFT approach improves the modelling of the galaxy power spectrum, providing a more robust consistency test of the assumed cosmological model. By casting both CMB and galaxy clustering observations, spanning distinct redshift regimes, into k-space, we can identify discrepancies between the datasets of different redshifts, which would indicate potential inaccuracies in the assumed expansion history. While previous studies have shown consistency with ΛCDM, this work extends the analysis with higher-quality data to further test the expansion histories of both ΛCDM and w 0 w a CDM. Our findings show that both ΛCDM and w 0 w a CDM provide consistent fits to the linear matter power spectrum recovered from DESI DR1 data.

cosmological parameters from LSS↗

DIGS: deep inference of galaxy spectra with neural posterior estimation

Abstract With the advent of billion-galaxy surveys with complex data, the need of the hour is to efficiently model galaxy spectral energy distributions (SEDs) with robust uncertainty quantification. The combination of simulation-based inference (SBI) and amortized neural posterior estimation (NPE) has been successfully used to analyse simulated and real galaxy photometry both precisely and efficiently. In this work, we utilise this combination and build on existing literature to analyse simulated noisy galaxy spectra. Here, we demonstrate a proof-of-concept study of spectra that is (a) an efficient analysis of galaxy SEDs and inference of galaxy parameters with physically interpretable uncertainties; and (b) amortized calculations of posterior distributions of said galaxy parameters at the modest cost of a few galaxy fits with Markov chain Monte Carlo (MCMC) methods. We utilise the SED generator and inference framework Prospector to generate simulated spectra, and train a dataset of 2 × 10 6 spectra (corresponding to a five-parameter SED model) with NPE. We show that SBI—with its combination of fast and amortized posterior estimations—is capable of inferring accurate galaxy stellar masses and metallicities. Our uncertainty constraints are comparable to or moderately weaker than traditional inverse-modelling with Bayesian MCMC methods (e.g. 0.17 and 0.26 dex in stellar mass and metallicity for a given galaxy, respectively). We also find that our inference framework conducts rapid SED inference (0.9–1.2 × 10 5 galaxy spectra via SBI/NPE at the cost of 1 MCMC-based fit). With this work, we set the stage for further work that focuses of SED fitting of galaxy spectra with SBI, in the era of JWST galaxy survey programs and the wide-field Roman Space Telescope spectroscopic surveys.

spectroscopy↗

Biomimetic, Soft-Material Synapse for Neuromorphic Computing: from Device to Network

Neuromorphic computing refers to a variety of brain-inspired computers, devices, and models inspired by the interconnectivity, performance, and energy efficiency of the human brain. Unlike the ubiquitous von Neumann computer architectures with complex processor cores and sequential computation, biological neurons and synapses operate by storing and processing information simultaneously with the capacity of flexible adaptation resulting in massive computational capability with much less power consumption. The search for a synaptic material which can closely imitate bio-synapse has led to an alamethicin-doped, synthetic biomembrane which can emulate key synaptic functions due to generic memristive property enabling learning and computation. This two-terminal, biomolecular memristor, in contrast to its solid-state counterparts, features similar structure, switching mechanism, and ionic transport modality as biological synapses while consuming considerably lower power. In this paper, we outline a methodology for using this biomolecular synapse to build neural networks capable of solving real-world problems. The physical mechanism underlying its volatile memristance is explored followed by the development of a model of this device for circuit simulation. We outline a circuit design technique to integrate this synapse with solid-state neuron circuit for hardware implementation. Based on these results, we develop a high level simulation framework and use a training scheme called Evolutionary Optimization for Neuromorphic System (EONS) to generate networks for solving two problems, namely iris dataset classification and EEG classification task. The small network size and comparable to state-of-the-art accuracy of these preliminary networks show its potential to enhance synaptic functionality in next generation neuromorphic hardware.

Sakib hasan, Md↗

Quantum Machine Learning Applications in High-Energy Physics

Some of the most significant achievements of the modern era of particle physics, such as the discovery of the Higgs boson, have been made possible by the tremendous effort in building and operating large-scale experiments like the Large Hadron Collider or the Tevatron. In these facilities, the ultimate theory to describe matter at the most fundamental level is constantly probed and verified. These experiments often produce large amounts of data that require storing, processing, and analysis techniques that continually push the limits of traditional information processing schemes. Thus, the High-Energy Physics (HEP) field has benefited from advancements in information processing and the development of algorithms and tools for large datasets. More recently, quantum computing applications have been investigated to understand how the community can benefit from the advantages of quantum information science. Nonetheless, to unleash the full potential of quantum computing, there is a need to understand the quantum behavior and, thus, scale up current algorithms beyond what can be simulated in classical processors. In this work, we explore potential applications of quantum machine learning to data analysis tasks in HEP and how to overcome the limitations of algorithms targeted for Noisy Intermediate-Scale Quantum (NISQ) devices.

Delgado, Andrea↗

Scalable Regression Tree Learning on Hadoop using OpenPlanet

As scientific and engineering domains attempt to effectively analyze the deluge of data arriving from sensors and instruments, machine learning is becoming a key data mining tool to build prediction models. Regression tree is a popular learning model that combines decision trees and linear regression to forecast numerical target variables based on a set of input features. Map Reduce is well suited for addressing such data intensive learning applications, and a proprietary regression tree algorithm, PLANET, using MapReduce has been proposed earlier. In this paper, we describe an open source implement of this algorithm, OpenPlanet, on the Hadoop framework using a hybrid approach. Further, we evaluate the performance of OpenPlanet using realworld datasets from the Smart Power Grid domain to perform energy use forecasting, and propose tuning strategies of Hadoop parameters to improve the performance of the default configuration by 75% for a training dataset of 17 million tuples on a 64-core Hadoop cluster on FutureGrid.

Yin, Wei↗

Computationally Accelerated Discovery and Experimental Demonstration of High-Performance Materials for Advanced Solar Thermochemical Hydrogen Production

This project achieved its overarching goal of accelerating the discovery and validation of solar thermochemical hydrogen (STCH) materials through a tightly integrated approach that combined high-throughput computational screening, advanced machine learning (ML), and experimental testing. Guided by the objectives outlined in the Statement of Project Objectives (SOPO), our work fulfilled all major milestones across four technical tasks and delivered scientific breakthroughs and practical tools that significantly exceeded the original scope of the project. We began by addressing the challenge of predicting material phase stability through machine learning. A novel Python module was developed to generate thousands of meaningful features from composition, structure, and electronic properties, enabling rapid and reproducible ML model development. Using these tools, we trained a model to predict temperature-dependent Gibbs energies (G(T)) for inorganic crystalline materials with near-chemical accuracy—roughly 40 meV/atom—marking the first such descriptor of its kind. We also introduced a new machine-learned tolerance factor, τ, that accurately predicted perovskite formability with over 90% success, outperforming traditional heuristic models, such as the Goldschmidt tolerance factor. These capabilities allowed for rapid and accurate predictions of phase stability across a vast oxide composition space, setting the stage for high-throughput thermodynamic screening. Building on this foundation, we conducted an extensive computational screening of candidate STCH oxide materials. Over 1.1 million perovskite compositions were evaluated using the τ descriptor, leading to the identification of more than 27,000 predicted stable structures. Using density functional theory (DFT), we refined over 68,000 multinary perovskite structures and computed oxygen vacancy formation energies for over 1,300 ternary and double perovskites. These calculations enabled us to isolate compounds with redox behavior consistent with STCH requirements and resulted in a public dataset now hosted on the Materials Project. Recognizing that thermodynamic screening alone is insufficient, we addressed kinetic limitations by developing a suite of tools to estimate transition state (TS) energies for key redox reactions. We implemented a novel bounding approach that provides lower and upper estimates of TS energies with dramatically reduced computational cost, requiring less than 10% of the CPU time of a full nudged elastic band (NEB) calculation while maintaining high accuracy. This enabled rapid evaluation of over 200 reaction pathways across 90 materials. To further accelerate screening, we developed a SISSO-based ML model to predict diffusion barriers with a 96.7% success rate in classifying fast vs. slow materials, supporting a robust, data-driven framework for assessing redox kinetics. Experimental validation was critical to confirming the predictive power of our models. We synthesized and tested a wide array of candidate materials, including Mn-doped hercynite and several Gd- and La-based perovskites. Notably, Sr 0.4 Gd 0.6 Mn 0.6 Al 0.4 O 3 (SGMA) and Gd 0.5 La 0.5 Co 0.5 Fe 0.5 O 3 (GLCF) emerged as leading STCH materials, exhibiting robust redox cycling and high hydrogen yields exceeding 150 µmol H 2 /g per cycle. These materials also retained over 50% of their hydrogen productivity under high-conversion conditions (H 2 O:H 2 = 1333:1), demonstrating strong thermodynamic favorability and promising performance under industrially relevant scenarios. Additional candidates, such as La 2 MnNiO 6 (L2MN), were found to produce even higher yields than ceria under standard STCH conditions. Our collaborators at Sandia National Laboratories confirmed these findings using high-temperature X-ray diffraction and thermogravimetric analysis, observing stable phase evolution and reversible redox activity. In several respects, the project went beyond the goals initially outlined in the SOPO. We published 17 peer-reviewed articles, including a large dataset of over 66,000 theoretical perovskites and a new structure prediction method (SPuDS-DFT) that accurately identifies ground-state structures at a fraction of the cost of traditional DFT. We demonstrated that our machine-learned G(T) model offers accuracy rivaling quasiharmonic calculations while being orders of magnitude faster. In partnership with the Materials Project, we made our datasets openly available, providing a powerful new resource for the broader materials science community. The combined computational and experimental advances of this project represent a significant advance in STCH materials discovery. By creating a robust, generalizable, and open workflow for thermodynamic and kinetic screening, and validating key findings through synthesis and reactor testing, we have provided a practical and scalable pathway for the rapid identification of new redox-active materials. The tools, data, and materials developed under this project are already supporting ongoing research and have laid the groundwork for the next generation of solar fuel technologies.

08 HYDROGEN↗

Maipo River Valley Agriculture: Determining Crop Coefficients Using Remote Sensing for the Maipo River Valley Basin in Chile

Agriculture is the primary use of water in the Maipo River basin of Central Chile, accounting for ~ 75% of the total demand. Assessment of irrigation needs for agricultural production has commonly relied on reference crop coefficients (Kc) derived from geographic and climatic conditions that differ from those of Chile. In partnership with the Centro de Información de Recursos Naturales (CIREN), this work focused on calculating site-specific crop coefficients tailored to crop production in the water-stressed Maipo River Valley. Two distinct approaches were implemented, each relying on remotely-sensed Earth observation datasets from NASA over consecutive growing seasons from 2019 to 2022. The first method estimated Kc values based on their linear relationship with the Normalized Difference Vegetation Index (NDVI) obtained from either Terra Moderate Resolution Imaging Spectroradiometer (MODIS) or Landsat 8 Operational Land Imager (OLI) surface reflectance. The second technique leveraged information from the ISS Ecosystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) by computing the ratio between actual crop evapotranspiration (ET) and potential evapotranspiration (PET). Both procedures showed promising results that can build on one another. The former approach best captured vegetation signals of annual crops while the latter appeared suited for perennials. Overall, this study provides a strong basis and novel way to accurately estimate Kc using remote sensing, with the potential for improved irrigation management and reduction in water consumption.

Benjamin Goffin↗

Milwaukee Urban Development: Assessing Climate Vulnerability through the InVEST Model on Urban Cooling in Milwaukee Using NASA Earth Observations

Milwaukee’s neighborhoods experience increased social, health, and ecological stress from the Urban Heat Island (UHI) effect due to changing land cover and climate. Extreme urban heat disproportionately affects Black and Latine communities due to systematic disinvestment in infrastructure and lack of resources to cope with heat. Our partner, Groundwork Milwaukee, supports Environmental Justice organizing around urban heat in historically redlined neighborhoods. This study explores heat mitigation in Metcalfe Park, one such neighborhood that is a focus area for our partner, as well as across the city and county of Milwaukee. Our team utilized the Natural Capital Project’s Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) Model for urban cooling to model heat mitigation scenarios for Milwaukee in support of Groundwork’s climate resilience and heat mitigation work. The data inputs used in our analysis were Land Use Land Cover (LULC), evapotranspiration, and surface temperature derived from NASA Earth observations, as well as a biophysical table containing land cover class attributes. We developed a heat vulnerability index from American Community Survey datasets and satellite imagery to identify areas in most need of heat mitigation intervention. We found that central and northwest city areas, home to a majority of Milwaukee’s Black and Latine population, have low heat mitigation and high vulnerability to heat impacts compared to the rest of the county. We also found that tree canopy increases across scales are effective in promoting heat mitigation. Our results and end products will bolster the efforts of our partners to make decisions on heat mitigation strategies, build community resilience, and reverse legacies of environmental racism in the face of extreme heat and climate change.

urban heat island↗