Search NASA⌕ Search

SEARCH · Search NASA

Results for “evaluation datasets”

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

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

At least 1,279 records · Page 71

Machine Learning Approaches for Nuclear Material Accounting Data from Irradiation and Reprocessing

We are currently exploring data analysis methods for their ability to strengthen the synthesis and evaluation of information generated within domestic and international safeguard regimes. Safeguard data typically includes rich heterogenous datasets amenable to advanced data analytics methods, such as machine learning. We have converted transactional data containing both numerical and categorical attributes from a domestic nuclear material control and accountability (NMC&A) system into a low-dimensional numerical structure via linear principal component analysis. This data representation allows for global structure discovery via cluster analysis, which can characterize the typical behavior of each of the primary types of transaction events. Furthermore, the structure of the top principal components captures the “typical behavior” of the data and is thus amenable to anomaly detection through statistical hypothesis testing. We explored this capability by generating erroneous permutations of the data and computing the Q-residual quantity increase associated with the information loss when these data are projected into the low-dimensional principal component analysis space representative of typical transactions.Future work will focus on identifying data transformations (e.g., graph networks) that more closely align with the inherent structure of these transactions to explain more salient information and disambiguate the underlying nuclear process—in this case, irradiation and reprocessing—from artifacts of NMC&A system transactional record keeping data.

Drescher, Adam↗

Selective Amnesia using Contrastive Subnet Erasure for Class Level Unlearning in Vision Models

We study concept-level forgetting in pretrained vision models: removing an entire semantic category so the system no longer recognizes that object in unseen images and contexts, rather than merely forgetting specific training examples. Prior work either applies blunt global projections or fine-tunes parameters, which can introduce collateral damage to unrelated features, add compute, and become unstable as forgetting strength increases. We introduce Contrastive Subnet Erasure (CSE), a training-free, encoder-centric edit that targets a compact set of channels most responsible for the class and attenuates them in a calibrated manner. The modification is algebraically folded into the subsequent layer, yielding no inference-time overhead and leaving task heads unchanged. To evaluate whether forgetting generalizes beyond the data used to specify the class, we introduce a cross dataset protocol in which the class is defined on a source dataset and performance is measured on a disjoint target dataset drawn from a different distribution with no shared images. This setup tests whether the model still fails to recognize the object when it looks different or appears in new scenes, and it helps avoid overfitting to patterns in the source dataset. Across CIFAR 10, CIFAR 100, and ImageNet under this protocol, CSE achieves stronger forgetting of the target class while better preserving non target utility than existing baselines in both single class and multi class settings. Overall, CSE provides a simple, stable, and deployment-ready mechanism for class-level unlearning in vision.

Kotevska, Olivera [ORNL] (ORCID:0000000316772243)↗

Risk Informed Timelines (RIT) v.1.0

Risk Informed Timelines (RIT) is a more modern and holistic method for understanding how access delay timelines are developed and utilized in physical security system evaluations. The software utilizes Bayesian methods to combine subject matter expert judgement and small performance test datasets in a consistent and defensible way. It allows a more holistic view of delay performance that provides distributions of task times and task success probabilities to account for tasks that, if failed, would result in failure of the attack. The RIT methodology is meant to enable security risk managers to begin the transition into more comprehensive delay timeline analysis. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. SAND2021-8294 O

Noel, ToddG.↗

Methodology for physics-informed generation of synthetic neutron time-of-flight measurement data

Accurate neutron cross section data are a vital input to the simulation of nuclear systems for a wide range of applications from energy production to national security. The evaluation of experimental data is a key step in producing accurate cross sections. There is a widely recognized lack of reproducibility in the evaluation process due to its artisanal nature and therefore there is a call for improvement within the nuclear data community. This can be realized by automating/standardizing viable parts of the process, namely, parameter estimation by fitting theoretical models to experimental data. This automation effort could greatly benefit from a synthetic data resource. This work leverages problem-specific physics, Monte Carlo sampling, and a general methodology for data synthesis to generate unlimited, labelled experimental cross-section data that is statistically indistinguishable to the observed data. Heuristic and, where applicable, rigorous statistical comparisons to observed data support this claim. The demonstration is based on/limited to transmission measurements at Rensselaer Polytechnic Institute (RPI) and energy-differential cross sections in the resolved resonance region (RRR). An open-source software is published alongside this article that executes the complete methodology to produce high-utility synthetic datasets. The goal of this work is to provide an approach and corresponding tool that will allow the evaluation community to begin exploring more data-driven, ML-based solutions to long-standing challenges in the field.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Integrating Data From In Vitro New Approach Methodologies for Developmental Neurotoxicity

Abstract In vivo developmental neurotoxicity (DNT) testing is resource intensive and lacks information on cellular processes affected by chemicals. To address this, DNT new approach methodologies (NAMs) are being evaluated, including: the microelectrode array neuronal network formation assay; and high-content imaging to evaluate proliferation, apoptosis, neurite outgrowth, and synaptogenesis. This work addresses 3 hypotheses: (1) a broad screening battery provides a sensitive marker of DNT bioactivity; (2) selective bioactivity (occurring at noncytotoxic concentrations) may indicate functional processes disrupted; and, (3) a subset of endpoints may optimally classify chemicals with in vivo evidence for DNT. The dataset was comprised of 92 chemicals screened in all 57 assay endpoints sourced from publicly available data, including a set of DNT NAM evaluation chemicals with putative positives (53) and negatives (13). The DNT NAM battery provides a sensitive marker of DNT bioactivity, particularly in cytotoxicity and network connectivity parameters. Hierarchical clustering suggested potency (including cytotoxicity) was important for classifying positive chemicals with high sensitivity (93%) but failed to distinguish patterns of disrupted functional processes. In contrast, clustering of selective values revealed informative patterns of differential activity but demonstrated lower sensitivity (74%). The false negatives were associated with several limitations, such as the maximal concentration tested or gaps in the biology captured by the current battery. This work demonstrates that this multi-dimensional assay suite provides a sensitive biomarker for DNT bioactivity, with selective activity providing possible insight into specific functional processes affected by chemical exposure and a basis for further research.

Carstens, Kelly E.↗

MO-G-17A-04: Internal Dosimetric Calculations for Pediatric Nuclear Imaging Applications, Using Monte Carlo Simulations and High-Resolution Pediatric Computational Models

Purpose: Our purpose is to evaluate the administered absorbed dose in pediatric, nuclear imaging studies. Monte Carlo simulations with the incorporation of pediatric computational models can serve as reference for the accurate determination of absorbed dose. The procedure of the calculated dosimetric factors is described, while a dataset of reference doses is created. Methods: Realistic simulations were executed using the GATE toolkit and a series of pediatric computational models, developed by the “IT'IS Foundation”. The series of the phantoms used in our work includes 6 models in the range of 5–14 years old (3 boys and 3 girls). Pre-processing techniques were applied to the images, to incorporate the phantoms in GATE simulations. The resolution of the phantoms was set to 2 mm3. The most important organ densities were simulated according to the GATE “Materials Database”. Several used radiopharmaceuticals in SPECT and PET applications are being tested, following the EANM pediatric dosage protocol. The biodistributions of the several isotopes used as activity maps in the simulations, were derived by the literature. Results: Initial results of absorbed dose per organ (mGy) are presented in a 5 years old girl from the whole body exposure to 99mTc - SestaMIBI, 30 minutes after administration. Heart, kidney, liver, ovary, pancreas and brain are the most critical organs, in which the S-factors are calculated. The statistical uncertainty in the simulation procedure was kept lower than 5%. The Sfactors for each target organ are calculated in Gy/(MBq*sec) with highest dose being absorbed in kidneys and pancreas (9.29*10{sup 10} and 0.15*10{sup 10} respectively). Conclusion: An approach for the accurate dosimetry on pediatric models is presented, creating a reference dosage dataset for several radionuclides in children computational models with the advantages of MC techniques. Our study is ongoing, extending our investigation to other reference models and evaluating the results with clinical estimated doses.

60 APPLIED LIFE SCIENCES↗

Predicting Biomass Yields of Advanced Switchgrass Cultivars for Bioenergy and Ecosystem Services Using Machine Learning

The production of advanced perennial bioenergy crops within marginal areas of the agricultural landscape is gaining interest due to its potential to sustainably produce feedstocks for biofuels and bioproducts while also improving the sustainability and resilience of commodity crop production. However, predicting the biomass yields of this production system is challenging because marginal areas are often relatively small and spread around agricultural fields and are typically associated with various abiotic conditions that limit crop production. Machine learning (ML) offers a viable solution as a biomass yield prediction tool because it is suited to predicting relationships with complex functional associations. The objectives of this study were to (1) evaluate the accuracy of commonly applied ML algorithms in agricultural applications for predicting the biomass yields of advanced switchgrass cultivars for bioenergy and ecosystem services and (2) determine the most important biomass yield predictors. Datasets on biomass yield, weather, land marginality, soil properties, and agronomic management were generated from three field study sites in two U.S. Midwest states (Illinois and Iowa) over three growing seasons. The ML algorithms evaluated in the study included random forests (RFs), gradient boosting machines (GBMs), artificial neural networks (ANNs), K-neighbors regressor (KNR), AdaBoost regressor (ABR), and partial least squares regression (PLSR). Coefficient of determination (R 2 ) and mean absolute error (MAE) were used to evaluate the predictive accuracy of the tested algorithms. Results showed that the ensemble methods, RF (R 2 = 0.86, MAE = 0.62 Mg/ha), GBM (R 2 = 0.88, MAE = 0.57 Mg/ha), and GBM (R 2 = 0.78, MAE = 0.66 Mg/ha), were the most accurate in predicting biomass yields of the Independence, Liberty, and Shawnee switchgrass cultivars, respectively. This is in agreement with similar studies that apply ML to multi-feature problems where traditional statistical methods are less applicable and datasets used were considered to be relatively small for ANNs. Consistent with previous studies on switchgrass, the most important predictors of biomass yield included average annual temperature, average growing season temperature, sum of the growing season precipitation, field slope, and elevation. This study helps pave the way for applying ML as a management tool for alternative bioenergy landscapes where understanding agronomic and environmental performance of a multifunctional cropping system seasonally and interannually at the sub-field scale is critical.

09 BIOMASS FUELS↗

Development of a Framework and Methodology for an Advanced Reactor Materials Environmental Effects Design Guide

Advanced non-light-water reactor components may operate at elevated temperature while experiencing cyclic loading, significant neutron irradiation, and exposure to reactor coolant. ASME Boiler and Pressure Vessel Code, Section III, Division 5, provides design rules for elevated-temperature service but does not include specific procedures to account for environmental effects on material properties. This report develops an initial framework and methodology for an Environmental Effects Design Guide (EEDG) focused on neutron irradiation; coolant-environment effects are reserved for future work. The proposed approach treats irradiation as a property-based overlay on the existing Division 5 design process, with two routes: a sparse-data route applying two reduction factors — FCR on creep-rupture strength and FF on fatigue life — for the creep-fatigue evaluations that typically control the design of advanced high-temperature reactor components, and a fuller framework developing the property-to-rule chain across the four Division 5 checks (primary load, strain limits and ratcheting, creep-fatigue, and buckling), together with swelling and weldments as scope items. Both routes are scoped by an in-pile qualification that restricts the use of post-irradiation-examination-derived properties in regimes where an in-pile mechanism could control the design outcome. Illustrative outputs derived on a compiled annealed Type 316 database — FCR ≈ 0.78–0.86 and FF ≈ 0.4 — demonstrate the calculation method within that specific dataset. The framework is an initial, testable design-rule concept; it identifies a practical path for preliminary design evaluations under sparse data and the material data and testing needed to develop the framework further.

Barua, Bipul (ORCID:0000000247184113)↗

Improved representation of agricultural land use and crop management for large-scale hydrological impact simulation in Africa using SWAT+

To date, most regional and global hydrological models either ignore the representation of cropland or consider crop cultivation in a simplistic way or in abstract terms without any management practices. Yet, the water balance of cultivated areas is strongly influenced by applied management practices (e.g. planting, irrigation, fertilization, and harvesting). The SWAT+ (Soil and Water Assessment Tool) model represents agricultural land by default in a generic way, where the start of the cropping season is driven by accumulated heat units. However, this approach does not work for tropical and subtropical regions such as sub-Saharan Africa, where crop growth dynamics are mainly controlled by rainfall rather than temperature. In this study, we present an approach on how to incorporate crop phenology using decision tables and global datasets of rainfed and irrigated croplands with the associated cropping calendar and fertilizer applications in a regional SWAT+ model for northeastern Africa. We evaluate the influence of the crop phenology representation on simulations of leaf area index (LAI) and evapotranspiration (ET) using LAI remote sensing data from Copernicus Global Land Service (CGLS) and WaPOR (Water Productivity through Open access of Remotely sensed derived data) ET data, respectively. Results show that a representation of crop phenology using global datasets leads to improved temporal patterns of LAI and ET simulations, especially for regions with a single cropping cycle. However, for regions with multiple cropping seasons, global phenology datasets need to be complemented with local data or remote sensing data to capture additional cropping seasons. In addition, the improvement of the cropping season also helps to improve soil erosion estimates, as the timing of crop cover controls erosion rates in the model. With more realistic growing seasons, soil erosion is largely reduced for most agricultural hydrologic response units (HRUs), which can be considered as a move towards substantial improvements over previous estimates. We conclude that regional and global hydrological models can benefit from improved representations of crop phenology and the associated management practices. Future work regarding the incorporation of multiple cropping seasons in global phenology data is needed to better represent cropping cycles in areas where they occur using regional to global hydrological models.

crop phenology↗

I Know I'm Right, But Does My Phone?

Transportation is the largest source of green-house gas emissions in the United States. Reducing transportation emissions depends on human travel behavior, which relies on local land use and planning. Travel diaries, consisting of sequences of trips between places for a particular individual, are typically used to instrument human travel behavior. However, these diaries are only as accurate as the underlying methods used to construct them. Travel diary algorithms have been a popular research topic since the advent of GPS tracking surveys. Mode inference algorithms in particular have been well represented in literature. However, these algorithms have typically been validated using prompted recall of pre-segmented trips, which doesn't account for segmentation error, thus disregarding the continuity of mode inference. Furthermore, phone operating systems and applications have adopted battery-conserving techniques, but we are not aware of prior work that has characterized the resulting data collection errors or evaluated procedures to mitigate them. We introduce a framework to evaluate accuracy of trip length computations and mode inference. We develop a temporal alignment procedure in analyzing continuous mode-segmented trajectories for groups of trips. We then apply our framework to evaluate an example set of travel diary algorithms from the open-source OpenPATH travel diary platform against MobilityNet, a public dataset containing information from three artificial timelines that cover 15 different travel modes. Our results show that inference based on an integration with map features results in weighted F_1 scores of 0.60 (iOS) and 0.74 (android). We also show that OpenPATH tends to under count trip length, with mean of signed relative error of -0.0438 on android and -0.0704 on iOS. We hope that other travel diary algorithms will be evaluated using this standardized process, and that the results used to understand and improve the state-of-the-art in this field.

ADVANCED PROPULSION SYSTEMS,ENERGY PLANNING, POLIC↗

Entropy-Bayesian Inversion of Time-Lapse Tomographic GPR data for Monitoring Dielectric Permittivity and Soil Moisture Variations

In this study, we evaluate the possibility of monitoring soil moisture variation using tomographic ground penetrating radar travel time data through Bayesian inversion, which is integrated with entropy memory function and pilot point concepts, as well as efficient sampling approaches. It is critical to accurately estimate soil moisture content and variations in vadose zone studies. Many studies have illustrated the promise and value of GPR tomographic data for estimating soil moisture and associated changes, however, challenges still exist in the inversion of GPR tomographic data in a manner that quantifies input and predictive uncertainty, incorporates multiple data types, handles non-uniqueness and nonlinearity, and honors time-lapse tomograms collected in a series. To address these challenges, we develop a minimum relative entropy (MRE)-Bayesian based inverse modeling framework that non-subjectively defines prior probabilities, incorporates information from multiple sources, and quantifies uncertainty. The framework enables us to estimate dielectric permittivity at pilot point locations distributed within the tomogram, as well as the spatial correlation range. In the inversion framework, MRE is first used to derive prior probability distribution functions (pdfs) of dielectric permittivity based on prior information obtained from a straight-ray GPR inversion. The probability distributions are then sampled using a Quasi-Monte Carlo (QMC) approach, and the sample sets provide inputs to a sequential Gaussian simulation (SGSim) algorithm that constructs a highly resolved permittivity/velocity field for evaluation with a curved-ray GPR forward model. The likelihood functions are computed as a function of misfits, and posterior pdfs are constructed using a Gaussian kernel. Inversion of subsequent time-lapse datasets combines the Bayesian estimates from the previous inversion (as a memory function) with new data. The memory function and pilot point design takes advantage of the spatial-temporal correlation of the state variables. We first apply the inversion framework to a static synthetic example and then to a time-lapse GPR tomographic dataset collected during a dynamic experiment conducted at the Hanford Site in Richland, WA. We demonstrate that the MRE-Bayesian inversion enables us to merge various data types, quantify uncertainty, evaluate nonlinear models, and produce more detailed and better resolved estimates than straight-ray based inversion; therefore, it has the potential to improve estimates of inter-wellbore dielectric permittivity and soil moisture content and to monitor their temporal dynamics more accurately.

58 GEOSCIENCES↗

Entropy-Bayesian Inversion of Time-Lapse Tomographic GPR data for Monitoring Dielectric Permittivity and Soil Moisture Variations

In this study, we evaluate the possibility of monitoring soil moisture variation using tomographic ground penetrating radar travel time data through Bayesian inversion, which is integrated with entropy memory function and pilot point concepts, as well as efficient sampling approaches. It is critical to accurately estimate soil moisture content and variations in vadose zone studies. Many studies have illustrated the promise and value of GPR tomographic data for estimating soil moisture and associated changes, however, challenges still exist in the inversion of GPR tomographic data in a manner that quantifies input and predictive uncertainty, incorporates multiple data types, handles non-uniqueness and nonlinearity, and honors time-lapse tomograms collected in a series. To address these challenges, we develop a minimum relative entropy (MRE)-Bayesian based inverse modeling framework that non-subjectively defines prior probabilities, incorporates information from multiple sources, and quantifies uncertainty. The framework enables us to estimate dielectric permittivity at pilot point locations distributed within the tomogram, as well as the spatial correlation range. In the inversion framework, MRE is first used to derive prior probability density functions (pdfs) of dielectric permittivity based on prior information obtained from a straight-ray GPR inversion. The probability distributions are then sampled using a Quasi-Monte Carlo (QMC) approach, and the sample sets provide inputs to a sequential Gaussian simulation (SGSIM) algorithm that constructs a highly resolved permittivity/velocity field for evaluation with a curved-ray GPR forward model. The likelihood functions are computed as a function of misfits, and posterior pdfs are constructed using a Gaussian kernel. Inversion of subsequent time-lapse datasets combines the Bayesian estimates from the previous inversion (as a memory function) with new data. The memory function and pilot point design takes advantage of the spatial-temporal correlation of the state variables. We first apply the inversion framework to a static synthetic example and then to a time-lapse GPR tomographic dataset collected during a dynamic experiment conducted at the Hanford Site in Richland, WA. We demonstrate that the MRE-Bayesian inversion enables us to merge various data types, quantify uncertainty, evaluate nonlinear models, and produce more detailed and better resolved estimates than straight-ray based inversion; therefore, it has the potential to improve estimates of inter-wellbore dielectric permittivity and soil moisture content and to monitor their temporal dynamics more accurately.

Bayesian Inversion↗

Strategies to improve reference databases for soil microbiomes

A database of curated genomes is needed to better assess soil microbial communities and their processes associated with differing land management and environmental impacts. Interpreting soil metagenomic datasets with existing sequence databases is challenging because these datasets are biased towards medical and biotechnology research and can result in misleading annotations. We have curated a database of 928 genomes of soil-associated organisms (888 bacteria, 34 archaea, and 6 fungi). Using this database as a representation of the current state of knowledge of soil microbes that are well-characterized, we evaluated its composition and compared it to broader microbial databases, specifically NCBI’s RefSeq, as well as 3,035 publicly available soil amplicon datasets. These comparisons identified phyla and functions that are enriched in soils as well as those that may be underrepresented in RefSoil. For example, RefSoil was observed to have increased representation of Firmicutes despite its low abundance in soil environments and also lacked representation of Acidobacteria and Verrucomicrobia, which are abundant in soils. Our comparison of RefSoil to soil amplicon datasets allowed us to identify targets that if cultured or sequenced would significantly increase the biodiversity represented within RefSoil. To demonstrate the opportunities to access these underrepresented targets, we employed single cell genomics in a pilot experiment to recover 14 genomes from the "most wanted" list, which improved RefSoil's representation of EMP sequences by 7% by abundance. This effort demonstrates the value of RefSoil in the guidance of future research efforts and the capability of single cell genomics as a practical means to fill the existing genomic data gaps.

59 BASIC BIOLOGICAL SCIENCES↗

Meteorological and hydrological parameters for 17 locations of meteorological stations of the East River Watershed

The Data Package includes a set of csv files of input meteorological parameters for locations of 17 meteorological stations within the East River watershed. These meteorological datasets were downloaded from the (1) PRISM database--monthly precipitation, air temperature (minimum, mean, and maximum), vapor pressure deficit (minimum and maximum), and dewpoint temperature, and (2) NCEP/NCAR Reanalysis database – wind database. The datasets were used for calculations of the Potential Evapotranspiration (ETo), Actual Evapotranspiration (ET), Standard Precipitation Index (SPI) , Standard Evapotranspiration-Precipitation Index (SPEI) for the period from 1966 to 2021. The main research questions addressed are: the evaluation of the long-term temporal trends of climatic parameters, hierarchical clustering, and areal mapping/zonation of the East River watershed. Calculations were conducted in the Rstudio environment. The dataset additionally includes a file-level metadata (flmd.csv) file that lists each file contained in the dataset with associated metadata; and a data dictionary (dd.csv) file that contains column/row headers used throughout the files along with a definition, units, and data type.The input datasets were downloaded from (a) PRISM database (the Northwest Alliance for Computational Science and Engineering at the Oregon State University), and (b) NCEP/NCAR Reanalysis database.

54 ENVIRONMENTAL SCIENCES↗

Expanding NeMO-Net Machine Learning Capabilities for Citizen Science

NASA NeMO-Net, the neural multi-modal observation and training network for global coral reef assessment, is an open-source deep convolutional neural network and interactive active learning training software aiming to accurately assess the present and past dynamics of coral reef ecosystems through determination of percent living cover and morphology as well as mapping of spatial distribution. We present an interactive citizen science video game, released this April, for desktop and iOS devices where users interactively label morphology classifications over mm-scale 3D coral reef imagery captured using diver photomosaic imagery, the UAV enabled NASA FluidCam instrument, and satellite datasets. To date, the application has had over 40,000 downloads and over 60,000 unique coral reef classifications, each filtered through a user-based rating and expert evaluation system. We also present results from NeMO-Net’s convolutional neural network (CNN) models used to semantically segment 2D satellite imagery as well as projections of 3D coral reconstructions using user input data as training datasets. Fusing datasets using machine learning from multiple remote sensing platforms presents novel methodologies for assessing the health of coral ecosystems, which are critically endangered by a changing climate. In partnering with Mission Blue, the National Oceanic and Atmospheric Administration (NOAA), and the Living Oceans Foundation (LOF), NeMO-Net leverages an international consortium of subject matter experts to provide both proper training for citizen scientists and the generation of a labeled datasets to ingest into machine learning algorithms for global coral reef identification.

NeMO-Net↗

Understanding the Physics Representation of Deep Learning Models in Environmental Applications

Deep learning (DL) models have been popular in earth and environmental modeling and analysis, which exhibit huge potential in capturing and reconstructing the non-linearity of relevant environmental processes. They are extensively used as analytical tools or emulators for multiple domains (atmosphere, land surface, ocean, and biogeochemistry). Despite their success, their internal working mechanism remains largely unknown. Such a lack of knowledge hinders the identification of physically consistent models that are fully adaptive to non-stationary climate, as well as the development of physics-informed machine learning such as physics-informed neural network (PINN). To establish preliminary knowledge and framework of such physics representation evaluation, this project focuses on an improved understanding of DL models in the environmental applications. DL models are increasingly applied to environmental modeling and prediction. However, they have been evaluated mostly from a performance perspective, and there is a gap in understanding how they represent the known physics internally. Such knowledge is especially critical when applying DL models under climate change conditions, where new inputs are likely outside the ranges of the training datasets. In this project, we reveal how the known physical processes are represented within DL models from both statistical and mechanistic perspectives. Leveraging the traditional model evaluations that focus more on the accuracies of predictions, we establish a framework that examines both the accuracy and physics representation of DL models. This analysis framework can identify DL models that make the correct predictions based on correct physics, thus enhancing the existing explainable artificial intelligence (explainable-AI) portfolio. It lays a foundation for developing novel metrics to evaluate the emerging DL models in environmental applications. This knowledge also informs the development of physics-informed DL models by revealing the direct connections between the known physical processes and specific model components or structures.

54 ENVIRONMENTAL SCIENCES↗

Aerial and Processed Model Data Representing As-built Conditions in Coastal Port Arthur, Texas in May 2025

This dataset was collected by the Co-Design Team of the Southeast Texas Urban Integrated Field Lab, a research initiative led by the University of Texas at Austin and funded by the U.S. Department of Energy. The broader project focuses on developing climate-resilient design solutions for the Beaumont–Port Arthur region, with more information available at www.setx-uifl.org. Our team conducted aerial surveys of the Port Arthur coastal neighborhood in May 2025, before the start of construction scheduled for Summer 2026. These pre-construction datasets are designed to facilitate comparative analyses, including pre- and post-construction assessments and simulated inundation scenario evaluations. Aerial images were captured using DroneDeploy autonomous flight systems, with imagery processed through the DroneDeploy engine. All original aerial photographs are provided in JPG format and organized in zipped folders by area. The processed data package includes: 3D surface models Orthomosaics Geospatial and topographic mappings Point clouds For guidance on file contents, structure, and recommended usage, please refer to the included README file.

2D mapping↗

Non-Cooperative Facial Recognition Video Dataset Collection Plan

The Pacific Northwest National Laboratory (PNNL) will produce a non-cooperative (i.e. not posing for the camera) facial recognition video data set for research purposes to evaluate and enhance facial recognition systems technology. The aggregate data set consists of 1) videos capturing PNNL role players and public volunteers in three key operational settings, 2) photographs of the role players for enrolling in an evaluation database, and 3) ground truth data that documents when the role player is within various camera fields of view. PNNL will deliver the aggregate data set to DHS who may then choose to make it available to other government agencies interested in evaluating and enhancing facial recognition systems. The three operational settings that will be the focus of the video collection effort include: 1) unidirectional crowd flow 2) bi-directional crowd flow, and 3) linear and/or serpentine queues.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗