Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “evaluation datasets”

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

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

At least 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

A seamless approach for evaluating climate models across spatial scales

In regions of the world where topography varies significantly with distance, most global climate models (GCMs) have spatial resolutions that are too coarse to accurately simulate key meteorological variables that are influenced by topography, such as clouds, precipitation, and surface temperatures. One approach to tackle this challenge is to run climate models of sufficiently high resolution in those topographically complex regions such as the North American Regionally Refined Model (NARRM) subset of the Department of Energy’s (DOE) Energy Exascale Earth System Model version 2 (E3SM v2). Although high-resolution simulations are expected to provide unprecedented details of atmospheric processes, running models at such high resolutions remains computationally expensive compared to lower-resolution models such as the E3SM Low Resolution (LR). Moreover, because regionally refined and high-resolution GCMs are relatively new, there are a limited number of observational datasets and frameworks available for evaluating climate models with regionally varying spatial resolutions. As such, we developed a new framework to quantify the added value of high spatial resolution in simulating precipitation over the contiguous United States (CONUS). To determine its viability, we applied the framework to two model simulations and an observational dataset. We first remapped all the data into Hierarchical Equal-Area Iso-Latitude Pixelization (HEALPix) pixels. HEALPix offers several mathematical properties that enable seamless evaluation of climate models across different spatial resolutions including its equal-area and partitioning properties. The remapped HEALPix-based data are used to show how the spatial variability of both observed and simulated precipitation changes with resolution increases. This study provides valuable insights into the requirements for achieving accurate simulations of precipitation patterns over the CONUS. It highlights the importance of allocating sufficient computational resources to run climate models at higher temporal and spatial resolutions to capture spatial patterns effectively. Furthermore, the study demonstrates the effectiveness of the HEALPix framework in evaluating precipitation simulations across different spatial resolutions. This framework offers a viable approach for comparing observed and simulated data when dealing with datasets of varying spatial resolutions. By employing this framework, researchers can extend its usage to other climate variables, datasets, and disciplines that require comparing datasets with different spatial resolutions.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2022

VESIcal: A Critical Approach to Volatile Solubility Modelling Using the Open-Source Engine Vesical

Accurate models of H(2)O and CO(2) solubility in silicate melts are vital for understanding volcanic plumbing systems. These models are used to estimate the depths of magma storage regions from melt inclusion volatile contents, investigate the role of volatile exsolution as a driver of volcanic eruptions, and track the degassing path followed by a magma ascending to the surface. However, despite the large increase in the number of experimental constraints over the last two decades, many recent studies still utilize an earlier generation of models which were calibrated on experimental datasets with restricted compositional ranges. This may be because many of the available tools for more recent models require large numbers of input parameters to be hand-typed (e.g., temperature, concentrations of H(2)O, CO(2), and 8–14 oxides), making them difficult to implement on large datasets. Here, we use a new open-source Python3 tool, VESIcal, to critically evaluate the behaviors and sensitivities of different solubility models for a range of melt compositions. Using literature datasets of andesitic-dacitic experimental products and melt inclusions as case studies, we illustrate the importance of evaluating the calibration dataset of each model. Finally, we highlight the limitations of particular data presentation methods, such as isobar diagrams, and provide suggestions for alternatives, and best practices regarding the presentation and archiving of data. This review will aid the selection of the most applicable solubility model for different melt compositions, and identifies areas where additional experimental constraints on volatile solubility are required.

magma↗
DOE OSTI2021

Evaluating the Benefits of Bayesian Hierarchical Methods for Analyzing Heterogeneous Environmental Datasets: A Case Study of Marine Organic Carbon Fluxes

Large compilations of heterogeneous environmental observations are increasingly available as public databases, allowing researchers to test hypotheses across datasets. Statistical complexities arise when analyzing compiled data due to unbalanced spatial sampling, variable environmental context, mixed measurement techniques, and other reasons. Hierarchical Bayesian modeling is increasingly used in environmental science to describe these complexities, however few studies explicitly compare the utility of hierarchical Bayesian models to simpler and more commonly applied methods. Here we demonstrate the utility of the hierarchical Bayesian approach with application to a large compiled environmental dataset consisting of 5,741 marine vertical organic carbon flux observations from 407 sampling locations spanning eight biomes across the global ocean. We fit a global scale Bayesian hierarchical model that describes the vertical profile of organic carbon flux with depth. Profile parameters within a particular biome are assumed to share a common deviation from the global mean profile. Individual station-level parameters are then modeled as deviations from the common biome-level profile. The hierarchical approach is shown to have several benefits over simpler and more common data aggregation methods. First, the hierarchical approach avoids statistical complexities introduced due to unbalanced sampling and allows for flexible incorporation of spatial heterogeneitites in model parameters. Second, the hierarchical approach uses the whole dataset simultaneously to fit the model parameters which shares information across datasets and reduces the uncertainty up to 95% in individual profiles. Third, the Bayesian approach incorporates prior scientific information about model parameters; for example, the non-negativity of chemical concentrations or mass-balance, which we apply here. We explicitly quantify each of these properties in turn. We emphasize the generality of the hierarchical Bayesian approach for diverse environmental applications and its increasing feasibility for large datasets due to recent developments in Markov Chain Monte Carlo algorithms and easy-to-use high-level software implementations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

CLM5 CAMELS Basins Ensemble

Land surface models such as Community Land Model Version 5 (CLM5) are essential tools for simulating the behaviors of the terrestrial system. Despite the extensive application of CLM5, limited attention has been paid to the underlying uncertainties associated with its hydrologic parameters and the implications that these uncertainties have on water resources applications. To address this long-standing issue, we conduct a comprehensive hydrologic parameter uncertainty characterization (UC) of CLM5 over the hydroclimatic gradients of the Conterminous United States using five meteorological datasets. Key datasets produced from the UC experiment include a benchmark dataset of CLM5 default hydrological performance, parameter sensitivity identified for 28 hydrological metrics, and large ensemble outputs for hydrological predictions. The presented datasets can assist CLM5 calibration and to support broad applications such as evaluating vulnerabilities to droughts and floods. The dataset can be used to identify under what hydroclimate conditions parametric uncertainties demonstrate substantial effects on hydrological predictions and clarify where further investigations are needed to understand how land runoff uncertainties interact with other Earth system processes. Please refer to the included README.pdf for a description of the included files. Note that raw CLM5 model outputs for each forcing dataset are hosted on a Globus endpoint: https://app.globus.org/file-manager?destination_id=d22ef858-27b0-11ed-a910-fd3165076336.

Yan, Hongxiang↗
DOE OSTI2023

CLM5 CAMELS Basins Ensemble

Land surface models such as Community Land Model Version 5 (CLM5) are essential tools for simulating the behaviors of the terrestrial system. Despite the extensive application of CLM5, limited attention has been paid to the underlying uncertainties associated with its hydrologic parameters and the implications that these uncertainties have on water resources applications. To address this long-standing issue, we conduct a comprehensive hydrologic parameter uncertainty characterization (UC) of CLM5 over the hydroclimatic gradients of the Conterminous United States using five meteorological datasets. Key datasets produced from the UC experiment include a benchmark dataset of CLM5 default hydrological performance, parameter sensitivity identified for 28 hydrological metrics, and large ensemble outputs for hydrological predictions. The presented datasets can assist CLM5 calibration and to support broad applications such as evaluating vulnerabilities to droughts and floods. The dataset can be used to identify under what hydroclimate conditions parametric uncertainties demonstrate substantial effects on hydrological predictions and clarify where further investigations are needed to understand how land runoff uncertainties interact with other Earth system processes. Please refer to the README.pdf for a description of the included files. Note that raw CLM5 model outputs for each forcing dataset are hosted on a Globus endpoint: https://app.globus.org/file-manager?destination_id=d22ef858-27b0-11ed-a910-fd3165076336.

Yan, Hongxiang↗
NASA NTRS2017

Evaluating a Priori Ozone Profile Information Used in TEMPO (Tropospheric Emissions: Monitoring of Pollution) Tropospheric Ozone Retrievals

A primary objective for TOLNet is the evaluation and validation of space-based tropospheric O3 retrievals from future systems such as the Tropospheric Emissions: Monitoring of Pollution (TEMPO) satellite. This study is designed to evaluate the tropopause-based O3 climatology (TB-Clim) dataset which will be used as the a priori profile information in TEMPO O3 retrievals. This study also evaluates model simulated O3 profiles, which could potentially serve as a priori O3 profile information in TEMPO retrievals, from near-real-time (NRT) data assimilation model products (NASA Global Modeling and Assimilation Office (GMAO) Goddard Earth Observing System (GEOS-5) Forward Processing (FP) and Modern-Era Retrospective analysis for Research and Applications version 2 (MERRA2)) and full chemical transport model (CTM), GEOS-Chem, simulations. The TB-Clim dataset and model products are evaluated with surface (0-2 km) and tropospheric (0-10 km) TOLNet observations to demonstrate the accuracy of the suggested a priori dataset and information which could potentially be used in TEMPO O3 algorithms. This study also presents the impact of individual a priori profile sources on the accuracy of theoretical TEMPO O3 retrievals in the troposphere and at the surface. Preliminary results indicate that while the TB-Clim climatological dataset can replicate seasonally-averaged tropospheric O3 profiles observed by TOLNet, model-simulated profiles from a full CTM (GEOS-Chem is used as a proxy for CTM O3 predictions) resulted in more accurate tropospheric and surface-level O3 retrievals from TEMPO when compared to hourly (diurnal cycle evaluation) and daily-averaged (daily variability evaluation) TOLNet observations. Furthermore, it was determined that when large daily-averaged surface O3 mixing ratios are observed (65 ppb), which are important for air quality purposes, TEMPO retrieval values at the surface display higher correlations and less bias when applying CTM a priori profile information compared to all other data products. The primary reason for this is that CTM predictions better capture the spatio-temporal variability of the vertical profiles of observed tropospheric O3 compared to the TB-Clim dataset and other NRT data assimilation models evaluated during this study.

Pollution↗
DOE OSTI2026

A Comprehensive Comparison of Building Height Products at a Global Level

The built environment continues to expand due to increasing human populations with some projections indicating additional growth of 2.5 billion persons by 2050. Buildings are a key component of the built environment and information on building height and urban morphology in general can help to better understand and address a host of development and urban planning issues including energy consumption, urbanization, population dynamics, urban heat islands, environmental degradation, and others. To date, several approaches have been developed to address the need for mapping building height at scale. These approaches can be grouped into two main categories: 1) directly measure height from remote sensing products or 2) indirectly estimate height from footprint morphology features at the building-by-building level. Due to a number of these global height datasets now being available for consumption and use across domains, there is a critical need to understand the differences displayed between these products and how the use of these may influence the outcome when applied. In this analysis, we compare and evaluate 5 global building height datasets: Global Human Settlement, World Settlement Footprint 3D, Global Building Morphology dataset, three-dimensional Global Building Footprints, and an approach developed at Oak Ridge National Laboratory (ORNL). Our analysis shows that, across 5 cities at a global scale with high-quality building height reference data, the approach developed by ORNL of inferring height at the building-by-building level is more accurate and consistent. This was evidenced by the ORNL product improving upon the baseline in all 5 cities, the only product to do so. These findings will aid researchers in choosing a global building height product that best meets their needs and provides global availability, consistency, and accuracy.

42 ENGINEERING↗
DOE OSTI2021

Expanding standards in viromics: in silico evaluation of dsDNA viral genome identification, classification, and auxiliary metabolic gene curation

Viruses influence global patterns of microbial diversity and nutrient cycles. Though viral metagenomics (viromics), specifically targeting dsDNA viruses, has been critical for revealing viral roles across diverse ecosystems, its analyses differ in many ways from those used for microbes. To date, viromics benchmarking has covered read pre-processing, assembly, relative abundance, read mapping thresholds and diversity estimation, but other steps would benefit from benchmarking and standardization. Here we use in silico-generated datasets and an extensive literature survey to evaluate and highlight how dataset composition (i.e., viromes vs bulk metagenomes) and assembly fragmentation impact (i) viral contig identification tool, (ii) virus taxonomic classification, and (iii) identification and curation of auxiliary metabolic genes (AMGs). The in silico benchmarking of five commonly used virus identification tools show that gene-content-based tools consistently performed well for long (≥3 kbp) contigs, while k -mer- and blast-based tools were uniquely able to detect viruses from short (≤3 kbp) contigs. Notably, however, the performance increase of k -mer- and blast-based tools for short contigs was obtained at the cost of increased false positives (sometimes up to ~5% for virome and ~75% bulk samples), particularly when eukaryotic or mobile genetic element sequences were included in the test datasets. Furthermore, for viral classification, variously sized genome fragments were assessed using gene-sharing network analytics to quantify drop-offs in taxonomic assignments, which revealed correct assignations ranging from ~95% (whole genomes) down to ~80% (3 kbp sized genome fragments). A similar trend was also observed for other viral classification tools such as VPF-class, ViPTree and VIRIDIC, suggesting that caution is warranted when classifying short genome fragments and not full genomes. Finally, we highlight how fragmented assemblies can lead to erroneous identification of AMGs and outline a best-practices workflow to curate candidate AMGs in viral genomes assembled from metagenomes. Together, these benchmarking experiments and annotation guidelines should aid researchers seeking to best detect, classify, and characterize the myriad viruses ‘hidden’ in diverse sequence datasets.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2013

Multi-Model Mean Nitrogen and Sulfur Deposition from the Atmospheric Chemistry and Climate Model Intercomparison Project (ACCMIP): Evaluation of Historical and Projected Future Changes

We present multi-model global datasets of nitrogen and sulfate deposition covering time periods from 1850 to 2100, calculated within the Atmospheric Chemistry and Climate Model Intercomparison Project (ACCMIP). The computed deposition fluxes are compared to surface wet deposition and ice core measurements.We use a new dataset of wet deposition for 2000-2002 based on critical assessment of the quality of existing regional network data. We show that for present day (year 2000 ACCMIP time slice), the ACCMIP results perform similarly to previously published multi-model assessments. For this time slice, we find a multimodel mean deposition of approximately 50 Tg(N) yr−1 from nitrogen oxide emissions, 60 Tg(N) yr−1 from ammonia emissions, and 83 Tg(S) yr−1 from sulfur emissions. The analysis of changes between 1980 and 2000 indicates significant differences between model and measurements over the United States but less so over Europe. This difference points towards a potential misrepresentation of 1980 NH3 emissions over North America. Based on ice core records, the 1850 deposition fluxes agree well with Greenland ice cores, but the change between 1850 and 2000 seems to be overestimated in the Northern Hemisphere for both nitrogen and sulfur species. Using the Representative Concentration Pathways (RCPs) to define the projected climate and atmospheric chemistry related emissions and concentrations, we find large regional nitrogen deposition increases in 2100 in Latin America, Africa and parts of Asia under some of the scenarios considered. Increases in South Asia are especially large, and are seen in all scenarios, with 2100 values more than double their 2000 counterpart in some scenarios and reaching >1300 mg(N)m−2 yr−1 averaged over regional to continental-scale regions in RCP 2.6 and 8.5, 30-50% larger than the values in any region currently (circa 2000). However, sulfur deposition rates in 2100 are in all regions lower than in 2000 in all the RCPs. The new ACCMIP multimodel deposition dataset provides state-of-the-science, consistent and evaluated time slice (spanning 1850-2100) global gridded deposition fields for use in a wide range of climate and ecological studies.

Lamarque, J.-F.↗
NASA NTRS2022

Flood Monitoring Using Enhanced Resolution Passive Microwave Data: A Test Case Over Bangladesh

Monitoring floods is a major issue in water resources management and risk mitigation, especially in the Global South. Optical and radar observations, even providing a fine spatial resolution, are still limited by cloud cover interaction or insufficient temporal resolution. On the other hand, passive microwave (PMW) sensors collect information on a daily frequency with minor cloud cover interaction, but they have been historically limited in terms of spatial resolution. Here, we evaluate the capability of an enhanced spatial resolution PMW dataset (3.125 km) in monitoring spatio-temporal evolution of flood events, focusing on a major flood event that occurred in October 2005 in Bangladesh. We apply an algorithm aimed to remove the seasonal variability of surface temperature from the PMW timeseries, exploiting the difference in emissivity between dry and water-covered pixels. We assess the capability of the algorithm in capturing flood evolution and extension through the comparison with quantities obtained from optical data collected by the Moderate Resolution Imaging Spectroradiometer (MODIS) and water level measurements. We also compare the enhanced product with the historical coarser resolution dataset by means of a variogram-based analysis to evaluate the improvements in terms of spatial representation. Finally, we evaluate the possibility to extract the water fraction within a single pixel by using an Advanced Microwave Scanning Radiometer—Earth Observing System (AMSR-E) emissivity dataset and compare the estimates with MODIS-derived water fractions. Our results show that the enhanced PMW product outperforms the coarser one when compared to flood mapped from optical data based on information content, indicating that it is possible to integrate such a product into the mapping of floods at a global scale on a daily basis.

flood detection↗
DOE OSTI2022

LBNL Fault Detection and Diagnostics Datasets

These datasets can be used to evaluate and benchmark the performance accuracy of Fault Detection and Diagnostics (FDD) algorithms or tools. It contains operational data from simulation, laboratory experiments, and field measurements from real buildings for seven HVAC systems/equipment (rooftop unit, single-duct air handler unit, dual-duct air handler unit, variable air volume box, fan coil unit, chiller plant, and boiler plant). Each dataset includes a .pdf file to document key information necessary to understand the content and scope, multiple csv files containing all the time-series data for faults at different severity levels and one fault-free case, and a ttl file to visualize the data according to BRICK schema. The dataset was created by LBNL, PNNL, NREL, ORNL and Drexel University.

AC↗
DOE OSTI2025

Knowledge graph-aided Bayesian active learning for top- K genetic interaction discovery

In silico methods for predicting the effects of multi-gene perturbations hold great promise for advancing functional genomics, computational drug discovery, and disease modeling. However, the development of these predictive algorithms for mammalian systems has been hampered by limited datasets and high experimental costs. In this study, we present a Bayesian active learning framework designed to discover pairwise host gene knockdowns that effectively inhibit viral proliferation in an in vitro HIV-1 infection model. Our method leverages a biological knowledge graph as side information and employs a computationally efficient batch diversification approach. We evaluated this framework using a dataset of viral load measurements obtained from multi-day dual-gene depletion experiments, encompassing all possible pairwise knockdowns of over 350 host genes associated with HIV infection. We demonstrate that our framework rapidly identifies the most effective gene knockdown pairs for reducing viral load. Furthermore, we show that incorporating side information enhances performance during the early stages of active learning (low data regime), while our batch diversification strategy significantly boosts performance in later stages (high data regime). This framework is general and can be adapted to explore gene interactions in other contexts, such as synthetic lethality prediction and mapping epistatic effects across quantitative trait loci.

Computational biology and bioinformatics↗
NASA NTRS2017

Importance of a Priori Vertical Ozone Profiles for TEMPO Air Quality Retrievals

Ozone (O3) is a toxic pollutant which plays a major role in air quality. Typically, monitoring of surface air quality and O3 mixing ratios is conducted using in situ measurement networks. This is partially due to high-quality information related to air quality being limited from space-borne platforms due to coarse spatial resolution, limited temporal frequency, and minimal sensitivity to lower tropospheric and surface-level O3. The Tropospheric Emissions: Monitoring of Pollution (TEMPO) satellite is designed to address the limitations of current space-based platforms and to improve our ability to monitor North American air quality. TEMPO will provide hourly data of total column and vertical profiles of O3 with high spatial resolution to be used as a near-real-time air quality product. TEMPO O3 retrievals will apply the Smithsonian Astrophysical Observatory profile algorithm developed based on work from GOME (Global Ozone Monitoring Experiment), GOME-2, and OMI (Ozone Monitoring Instrument). This algorithm is suggested to use a priori O3 profile information from a climatological data-base developed from long-term ozone-sonde measurements (tropopause-based (TB-Clim) O3 climatology). This study evaluates the TB-Clim dataset and model simulated O3 profiles, which could potentially serve as a priori O3 profile information in TEMPO retrievals, from near-real-time data assimilation model products (NASA GMAO's (Global Modeling and Assimilation Office) operational GEOS-5 (Goddard Earth Observing System, Version 5) FP (Forecast Products) model and reanalysis data from MERRA2 (Modern-Era Retrospective analysis for Research and Applications, Version 2)) and a full chemical transport model (CTM), GEOS-Chem. In this study, vertical profile products are evaluated with surface (0-2 kilometers) and tropospheric (0-10 kilometers) TOLNet (Tropospheric Ozone Lidar Network) observations and the theoretical impact of individual a priori profile sources on the accuracy of TEMPO O3 retrievals in the troposphere and at the surface are presented. Results indicate that while the TB-Clim climatological dataset can replicate seasonally-averaged tropospheric O3 profiles, model-simulated profiles from a full CTM resulted in more accurate tropospheric and surface-level O3 retrievals from TEMPO when compared to hourly and daily-averaged TOLNet observations. Furthermore, it is shown that when large surface O3 mixing ratios are observed, TEMPO retrieval values at the surface are most accurate when applying CTM a priori profile information compared to all other data products.

Priori↗
DOE OSTI2022

Proxy-Based Bayesian Inversion Of Poroelastic Simulations To Interpret Strain Tensor Data Measured During Well Testing

The long runtimes of 3D poroelastic numerical simulators makes it impractical to interpret deformation datasets using many inversion schemes. Recent advances in instrumentation have made it possible to measure the strain tensor during well testing, but the lack of robust inversion methods is limiting the ability to interpret these data. We have developed an inversion workflow that reduces the number of computations required to complete a Bayesian inversion using DREAMzs. The workflow trains a KNN model using output from the poroelastic simulator, and then uses the KNN model as a proxy for the simulator during inversion. The workflow also includes a strategy for ensuring the results from the proxy model converge to the results from the simulator, ensuring the accuracy of the final results. An idealized example configured to represent a well test in a deep aquifer is used to verify that the workflow correctly identifies parameters and characterizes noise. Field data measured using strainmeters during an injection test at an oil reservoir in Oklahoma are used to evaluate performance with a real dataset. The workflow identified 265 history matching solutions out of 1240 total simulation runs (21% acceptance ratio), and the results are used to characterize posterior parameter distribution and evaluate the prediction uncertainty. This approach makes it feasible to invert strain data measured during well testing and this has the potential to improve the characterization of aquifers and reservoirs.

Roudini, Soheil↗
DOE OSTI2024

Self-Supervised Cloud Classification

Abstract Low-level marine clouds play a pivotal role in Earth’s weather and climate through their interactions with radiation, heat and moisture transport, and the hydrological cycle. These interactions depend on a range of dynamical and microphysical processes that result in a broad diversity of cloud types and spatial structures, and a comprehensive understanding of cloud morphology is critical for continued improvement of our atmospheric modeling and prediction capabilities moving forward. Deep learning has recently accelerated our ability to study clouds using satellite remote sensing, and machine learning classifiers have enabled detailed studies of cloud morphology. A major limitation of deep learning approaches to this problem, however, is the large number of hand-labeled samples that are required for training. This work applies a recently developed self-supervised learning scheme to train a deep convolutional neural network (CNN) to map marine cloud imagery to vector embeddings that capture information about mesoscale cloud morphology and can be used for satellite image classification. The model is evaluated against existing cloud classification datasets and several use cases are demonstrated, including training cloud classifiers with very few labeled samples, interrogation of the CNN’s learned internal feature representations, cross-instrument application, and resilience against sensor calibration drift and changing scene brightness. The self-supervised approach learns meaningful internal representations of cloud structures and achieves comparable classification accuracy to supervised deep learning methods without the expense of creating large hand-annotated training datasets. Significance Statement Marine clouds heavily influence Earth’s weather and climate, and improved understanding of marine clouds is required to improve our atmospheric modeling capabilities and physical understanding of the atmosphere. Recently, deep learning has emerged as a powerful research tool that can be used to identify and study specific marine cloud types in the vast number of images collected by Earth-observing satellites. While powerful, these approaches require hand-labeling of training data, which is prohibitively time intensive. This study evaluates a recently developed self-supervised deep learning method that does not require human-labeled training data for processing images of clouds. We show that the trained algorithm performs competitively with algorithms trained on hand-labeled data for image classification tasks. We also discuss potential downstream uses and demonstrate some exciting features of the approach including application to multiple satellite instruments, resilience against changing image brightness, and its learned internal representations of cloud types. The self-supervised technique removes one of the major hurdles for applying deep learning to very large atmospheric datasets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

The value of human data annotation for machine learning based anomaly detection in environmental systems

Anomaly detection is the process of identifying unexpected data samples in datasets. Automated anomaly detection is either performed using supervised machine learning models, which require a labelled dataset for their calibration, or unsupervised models, which do not require labels. While academic research has produced a vast array of tools and machine learning models for automated anomaly detection, the research community focused on environmental systems still lacks a comparative analysis that is simultaneously comprehensive, objective, and systematic. This knowledge gap is addressed for the first time in this study, where 15 different supervised and unsupervised anomaly detection models are evaluated on 5 different environmental datasets from engineered and natural aquatic systems. To this end, anomaly detection performance, labelling efforts, as well as the impact of model and algorithm tuning are taken into account. As a result, our analysis reveals the relative strengths and weaknesses of the different approaches in an objective manner without bias for any particular paradigm in machine learning. Most importantly, our results show that expert-based data annotation is extremely valuable for anomaly detection based on machine learning.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

The Impact of Time-Aware Design Choices in ICS Anomaly Detection

Industrial control systems (ICS) remain vulnerable to increasingly sophisticated cyberattacks, yet evaluating anomaly detection models in these environments is challenging due to temporal dependencies, missing-not-at-random patterns, and extremely imbalanced datasets. These factors make common practices—especially random data splits and na¨ıve imputation— prone to severe temporal leakage, which can inflate reported performance and obscure real-world limitations. In this work, we systematically examine classical machine learning models, temporal deep learning architecture, and tensordecomposition– based methods on a gas-pipeline dataset using a fully temporally separated evaluation pipeline designed to mimic realistic deployment conditions. Our findings show that proper temporal handling and MNAR-aware preprocessing significantly alter the relative performance of popular anomaly-detection methods, providing practical guidance for designing reliable, leakage-resistant ICS intrusion-detection systems.

97 MATHEMATICS AND COMPUTING↗