Search NASA⌕ Search

SEARCH · Search NASA

Results for “dataset shift”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS2026

Timing the Flames: Geostationary Satellite Detection of Diurnally Shifting Stubble Burning in Northwestern India

Post-monsoon open-field stubble burning in northwestern (NW) India—a key agricultural region known as the “breadbasket”—is a longstanding practice used to clear fields. Satellite observations spanning over two decades have revealed significant upward trends in crop production, vegetative greenness, and the frequency of post-harvest fires, with this last contributing to hazardous air quality during the peak burning season (mid-October to mid-November). Since 2022, thermal anomaly data from Aqua-MODIS and SNPP-VIIRS sensors have shown a sharp decline in reported fire events—an observation that contrasts starkly with the concurrent rise in regional aerosol loading detected from space. This apparent discrepancy became particularly pronounced in 2024–2025, prompting a closer examination using high-temporal-resolution imagery from the Advanced Meteorological Imager (AMI) on the geostationary satellite GEO-KOMPSAT-2A. These observations revealed a clear spike in fire-related signals occurring around and after 4:00 p.m. local time, i.e., outside the typical noon to 2:00 p.m. detection window of the MODIS and VIIRS. A fire detection algorithm exploiting the fire-sensitive shortwave-infrared 3.8 μm signal and its contrast to 11.2 μm infrared observations is designed to adopt AMI observations and applied to its multi-year observations (2019–2025). The resulting fire dataset unambiguously shows a gradual shift in stubble burning activity toward the late afternoon hours beginning in 2022 which is underreported by polar-orbiting satellites. The orbital drift of NASA’s MODIS sensor on the Aqua platform allows detection of some of the gradually shifting fires during afternoon hours, but the MODIS still misses a large number of fires occurring around and after 4 p.m. The AMI’s relatively coarse spatial resolution (~4 km), a consequence of its slant viewing geometry over NW India, imposes inherent limitations on quantifying the full extent of fire occurrences. The operational air quality forecasting models currently assimilate satellite fire detections predominantly captured during early afternoon overpasses of the MODIS and VIIRS. The temporal shift in fire activity complicates such forecast, leading to a substantial underestimation of emissions. Intense stubble burning and the resulting air pollution highlight the need for effective crop residue management practices for mitigating the frequency of open biomass burning and thereby reducing episodic degradation of air quality and its associated public health and economic impacts.

post-monsoon stubble burning; northwestern India;↗
DOE OSTI2022

Benchmarking Demand Flexibility in Commercial Buildings and Flattening the Duck – Addressing Baseline and Commissioning Challenges

With the transition from our traditional electric grid to a cleaner grid with renewable power generation, there is a need to enable building loads to be flexible. Load shedding and shifting will be essential for flattening the “Duck” for decarbonization. This paper explored the trend in the timing of DR events as a reflection of the grid’s needs using recent four years of event data from 203 retail stores in 11 states. The events are becoming significantly shorter with 2-hour duration being the most popular; shifting to late afternoon and early evening is another trend beyond California. Benchmarking will be essential for accounting DF as a reliable grid resource. This paper addresses a challenging aspect of benchmarking – inaccuracies in counterfactual baseline methods can introduce significant DF metrics variations in addition to weather and building characteristics related factors. The conventional “10/10” with adjustment baseline method has inherent limitation by design for load shifting applications. Therefore, it is imperative to identify alternative methods. This study compared three hourly regression baseline methods with “10/10” methods using two groups of commercial buildings that participated in DR programs: (1) 121 big-box retail stores, and (2) 11 office buildings in CA. The 14-day hourly outdoor temperature regression method was found to produce least error in the tested datasets and is promising for load shifting. The paper also pointed out that commissioning issues can also be a significant barrier for achieving consistent DF performance, which building managers and utilities should be aware of.

Liu, Jingjing↗
DOE OSTI2026

Toward an AI-Powered Software Pipeline for Real-Time Tracking and Analysis of Wildfire and Smoke

Real-time tracking of wildfires and smoke is crucial for effective response, minimizing damage, protecting lives, and efficiently managing resources during fire emergencies. We develop a web-based AI-powered pipeline that detects wildfires in aerial video and estimates deployment-relevant behavior metrics, including cumulative burned area, burned-area growth rate, fire spread direction, and smoke dispersion. The system combines a YOLO-based detector with YCbCr-based fire segmentation, HSV-based smoke segmentation, Farneback optical flow, and centroid-based spatiotemporal tracking. Using ground sampling distance (GSD), pixel-level fire masks are converted to physical burned-area measurements by correlating fire pixel counts with camera altitude and tilt angle. We benchmark YOLO variants and non-YOLO baselines (GoogLeNet, CNN, DBN, Autoencoder, U-Net, and AlexNet) on the IEEE FLAME dataset and a newly created aerial frame dataset, Wildfire-DB. Cross-dataset evaluation uses a strict threshold-transfer protocol: decision thresholds are selected on FLAME validation and transferred unchanged to Wildfire-DB to quantify generalization under domain shift. YOLOv6 achieves the strongest cross-dataset frame-level fire detection on Wildfire-DB (ROC-AUC 0.8200, PR-AUC 0.8044, and transferred-threshold F1 0.7596). For tracking-oriented deployment requiring oriented localization, YOLO11-OBB provides the most reliable cross-dataset behavior among OBB-capable models while remaining computationally feasible. To analyze the feasibility of UAV deployment, we further measure inference efficiency using synchronized GPU and CPU power logs on a fixed workload of 1569 frames. YOLO-family models process the video in 5.73–12.47 seconds with net energy of 1247.28–1775.39 J, substantially lower latency and energy than heavier classification and reconstruction baselines. Overall, model optimality depends on operational objectives: YOLOv6 is best for cross-dataset detection robustness, whereas YOL...

Color segmentation↗
DOE OSTI2025

Leveraging Pre-Built Catalogs and Object-Level Scheduling to Eliminate I/O Bottlenecks in HPC Environments

Modern High-Performance Computing (HPC) environments face mounting challenges due to the shift from large to small file datasets, along with an increasing number of users and parallelized applications. As HPC systems rely on Parallel File Systems (PFS), such as Lustre for data processing, performance bottlenecks stemming from Object Storage Target (OST) contention have become a significant concern. Existing solutions, such as LADS with its object-level scheduling approach, fall short in large-scale HPC environments due to their inability to effectively address metadata I/O bottlenecks and the growing number of I/O processes. This study highlights the pressing need for a comprehensive solution that tackles both OST contention and metadata I/O challenges in diverse HPC workloads. To address these challenges, we propose SwiftLoad, an object-level I/O scheduling framework that leverages a metadata catalog to enhance the performance and efficiency of parallel HPC utilities. The adoption of the metadata catalog mitigates the metadata I/O bottlenecks that commonly occur in HPC utilities, a challenge that is particularly pronounced in object-level I/O scheduling. SwiftLoad addresses OST contention and the uneven distribution of I/O processes across different OSTs through mathematical modeling and incorporates a Loader Configuration Module to regulate the number of I/O processes. Evaluated with two representative utilities—data deduplication profiling and data augmentation—SwiftLoad achieved performance improvements of up to 5.63x and 11.0x, respectively, on a production supercomputer.

HPC↗
DOE OSTI2025

Deciphering the small-angle scattering of polydisperse hard spheres using deep learning

We introduce a deep learning approach for analyzing the scattering function of the polydisperse hard sphere system. We use a variational autoencoder-based neural network to learn the bidirectional mapping between the scattering function and the system parameters, including the volume fraction and polydispersity. Such that the trained model serves both as a generator that produces a scattering function from the system parameters and an inferrer that extracts system parameters from the scattering function. We first generate a scattering dataset by carrying out molecular dynamics simulations of the polydisperse hard spheres modeled by the truncated-shifted Lennard-Jones model, then analyze the scattering function dataset using singular value decomposition to confirm the feasibility of dimensional compression. Then, we split the dataset into training and testing sets and train our neural network on the training set only. Our generator model produces a scattering function with significantly higher accuracy compared to the traditional Percus–Yevick approximation and β correction, and the inferrer model can extract the volume fraction and polydispersity with much higher accuracy than traditional model functions.

Ding, Lijie [ORNL] (ORCID:0000000227454606)↗
DOE OSTI2024

Utah FORGE: Well 16B(78)-32 DTS, RFS DSS Strain, and Absolute Strain Circulation Test Fiber Optic Data - August 2024

This dataset contains processed fiber optic measurements collected during the extended cross-well circulation test at the Utah FORGE site in August 2024. The data was acquired from the 16B(78)-32 well using distributed fiber optic sensing (DFOS) technology, including distributed temperature sensing (DTS) on multimode fiber and distributed strain sensing (DSS) measurements on single-mode fiber. The dataset includes Brillouin absolute strain, Rayleigh frequency shift (RFS) DSS strain change, RFS DSS strain change rate, and temperature (DTS) data, all stored in HDF5 format. Time coordinates are provided in UTC, and depth measurements are given in measured depth relative to the rotary kelly bushing (MD RKB) in feet. The spatial sampling on the RFS DSS strain data and the Brillouin Absolute Total Strain data is 20 cm and the spatial sampling on the DTS data is 1m. The dataset is accompanied by a report from Neubrex, which provides further documentation.

15 GEOTHERMAL ENERGY↗
DOE OSTI2026

SPT-3G D1: Axion early dark energy with CMB experiments and DESI observations

We present the most up-to-date constraints on axion early dark energy (AEDE) from cosmic microwave background (CMB) and baryon acoustic oscillation (BAO) measurements. In particular, we assess the impact of data from ground-based CMB experiments, the South Pole Telescope (SPT) and the Atacama Cosmology Telescope (ACT)—both with and without Planck—on constraints on AEDE. We also highlight the impact that BAO information from the Dark Energy Spectroscopic Instrument (DESI) has on these constraints. From CMB data alone, we do not find statistically significant evidence for the presence of AEDE, and we find only moderate reduction in the Hubble tension. From the latest SPT data alone, we find the maximal fractional contribution of AEDE to the cosmic energy budget is f EDE < 0.12 at 95% confidence level (CL), and the Hubble tension between the SPT and Supernovae, H0, for the Equation of State of Dark Energy (SH0ES) results is reduced to the 2.3 σ level. When combining the latest SPT, ACT, and Planck datasets, we find f EDE < 0.070 at 95% CL and the Hubble tension at the 3.6 σ level. In contrast, adding DESI data to the CMB datasets results in mild preference for AEDE and, in some cases, non-negligible reduction in the Hubble tension. From SPT + DESI , we find f EDE = 0.08 1 - 0.052 + 0.037 at 68% CL, and the Hubble tension reduces to 1.5 σ . From the combination of DESI with all three CMB experiments, we get f EDE = 0.05 5 - 0.047 + 0.024 at 68% CL and a weak preference for AEDE over Λ CDM . This data combination, in turn, reduces the Hubble tension to 2.6 σ . We highlight that this shift in parameters when adding the DESI dataset is a manifestation of the discrepancy currently present between DESI and CMB experiments in the concordance model Λ CDM .

Khalife, A. R. [Paris, Inst. Astrophys.] (ORCID:00↗
DOE OSTI2023

Systematic Validation of Ensemble Cloud-Process Simulations using Polarimetric Radar Observations and Simulator over the NASA Wallops Flight Facility

The BiLateral Operational Storm-Scale Observation and Modeling (BLOSSOM) project was initiated to establish a long-term supersite to improve understanding of cloud physical states and processes as well as to support satellite and climate model programs over the WFF site via a bilateral approach of storm-scale observations and process modeling. This study highlights a noble systematic validation framework of the BLOSSOM ensemble cloud-process simulations through mixed-phase, light-rain, and deep-convective precipitation cases. The framework consists of creating a domain-shifted ensemble of large-scale forcing datasets, and configuring and performing cloud-process simulations with three different bulk microphysics schemes. Validation uses NASA S-band dual-POLarimetric radar (NPOL) observations in the form of statistical composites and skill scores via a polarimetric radar simulator and newly developed CfRad Data tool (CfRAD). While the simulations capture the overall structures of the reflectivity composites, polarimetric signals are still poorly simulated, mainly due to a lack of representation of ice microphysics diversity in shapes, orientation distributions, and their complex mixtures. Despite the limitation, this new ensemble-based validation framework demonstrates that 1) no particular forcing or microphysics scheme outperforms the rest and 2) the skill scores of coarse- and fine-resolution ensemble simulations with different domain-shifted forcing and microphysics schemes are highly correlated with each other with no clear improvement. On the other hand, this suggests that coarse-resolution ensemble simulations are relevant for selecting the best meteorological forcing and microphysics scheme before conducting computationally demanding large eddy simulations (LESs) in support of aircraft and satellite instrument development as well as cloud-precipitation-convection parameterizations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Dataset for Cruz-O'Byrne et al (2026): "Divergent biogeochemical responses in upland coastal forest soils to repeated flooding and shifts in water chemistry"

Hydrologic disturbances from accelerated sea-level rise and the increasing frequency and intensity of storms and tidal flooding are altering biogeochemical processes in upland coastal forests, transforming these ecosystems into wetlands. However, the initial effects of flooding on belowground biogeochemistry and the mechanisms driving greenhouse gas dynamics and soil organic matter stability during the early stages of this transition remain poorly understood. This dataset presents the results of a mesocosm experiment conducted in a controlled, highly instrumented laboratory environment, in which freshwater and brackish water pulses were applied to intact soil monoliths from a temperate upland coastal forest to examine how floodwater chemistry influences soil biogeochemistry and organo-mineral interactions. All data files are plain-text CSV (comma-separated value), and no special software is required to read them. Details about the content of each file are available in the document “Dataset_readme”. The dataset consists of the following data: • rcruzobyrne_moisture: Soil volumetric water content (VWC) • rcruzobyrne_GHG: Headspace greenhouse gas (GHG) concentration and fluxes • rcruzobyrne_methane_isotopes: Headspace methane isotope signature • rcruzobyrne_porewater: Porewater chemistry • rcruzobyrne_CDOM: Porewater colored dissolved organic matter (CDOM) • rcruzobyrne_FTIR: Soil Fourier-transform infrared (FTIR) spectroscopy Details of the experimental setup, data collection, and data analysis are provided in the manuscript by Cruz-O’Byrne et al (2026) Divergent biogeochemical responses in upland coastal forest soils to repeated flooding and shifts in water chemistry. Biogeochemistry. https://doi.org/10.1007/s10533-026-01340-0

EARTH SCIENCE > ATMOSPHERE > GREENHOUSE GAS↗
DOE OSTI2020

Harmonization of global land use change and management for the period 850–2100 (LUH2) for CMIP6

Abstract. Human land use activities have resulted in large changes to the biogeochemical and biophysical properties of the Earth's surface, with consequences for climate and other ecosystem services. In the future, land use activities are likely to expand and/or intensify further to meet growing demands for food, fiber, and energy. As part of the World Climate Research Program Coupled Model Intercomparison Project (CMIP6), the international community has developed the next generation of advanced Earth system models (ESMs) to estimate the combined effects of human activities (e.g., land use and fossil fuel emissions) on the carbon–climate system. A new set of historical data based on the History of the Global Environment database (HYDE), and multiple alternative scenarios of the future (2015–2100) from Integrated Assessment Model (IAM) teams, is required as input for these models. With most ESM simulations for CMIP6 now completed, it is important to document the land use patterns used by those simulations. Here we present results from the Land-Use Harmonization 2 (LUH2) project, which smoothly connects updated historical reconstructions of land use with eight new future projections in the format required for ESMs. The harmonization strategy estimates the fractional land use patterns, underlying land use transitions, key agricultural management information, and resulting secondary lands annually, while minimizing the differences between the end of the historical reconstruction and IAM initial conditions and preserving changes depicted by the IAMs in the future. The new approach builds on a similar effort from CMIP5 and is now provided at higher resolution (0.25∘×0.25∘) over a longer time domain (850–2100, with extensions to 2300) with more detail (including multiple crop and pasture types and associated management practices) using more input datasets (including Landsat remote sensing data) and updated algorithms (wood harvest and shifting cultivation); it is assessed via a new diagnostic package. The new LUH2 products contain > 50 times the information content of the datasets used in CMIP5 and are designed to enable new and improved estimates of the combined effects of land use on the global carbon–climate system.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2020

Harmonization of global land use change and management for the period 850–2100 (LUH2) for CMIP6

Human land use activities have resulted in large changes to the biogeochemical and biophysical properties of the Earth's surface, with consequences for climate and other ecosystem services. In the future, land use activities are likely to expand and/or intensify further to meet growing demands for food, fiber, and energy. As part of the World Climate Research Program Coupled Model Intercomparison Project (CMIP6), the international community has developed the next generation of advanced Earth system models (ESMs) to estimate the combined effects of human activities (e.g., land use and fossil fuel emissions) on the carbon–climate system. A new set of historical data based on the History of the Global Environment database (HYDE), and multiple alternative scenarios of the future (2015–2100) from Integrated Assessment Model (IAM) teams, is required as input for these models. With most ESM simulations for CMIP6 now completed, it is important to document the land use patterns used by those simulations. Here we present results from the Land-Use Harmonization 2 (LUH2) project, which smoothly connects updated historical reconstructions of land use with eight new future projections in the format required for ESMs. The harmonization strategy estimates the fractional land use patterns, underlying land use transitions, key agricultural management information, and resulting secondary lands annually, while minimizing the differences between the end of the historical reconstruction and IAM initial conditions and preserving changes depicted by the IAMs in the future. The new approach builds on a similar effort from CMIP5 and is now provided at higher resolution (0.25°×0.25°) over a longer time domain (850–2100, with extensions to 2300) with more detail (including multiple crop and pasture types and associated management practices) using more input datasets (including Landsat remote sensing data) and updated algorithms (wood harvest and shifting cultivation); it is assessed via a new diagnostic package. The new LUH2 products contain > 50 times the information content of the datasets used in CMIP5 and are designed to enable new and improved estimates of the combined effects of land use on the global carbon–climate system.

land use land cover change↗
DOE OSTI2023

Advancing the central role of non-model biorepositories in predictive modeling of emerging pathogens

The COVID-19 pandemic demonstrated the insufficiency of a reactive approach to emerging zoonotic pathogens. With spillover increasing in frequency as environments change and the human footprint continues to grow, pandemic prevention will require predictive models that can identify (i) potential zoonoses with a high likelihood of emergence and (ii) environmental or other features that may trigger a shift in host, vector, or pathogen baselines associated with emergence and/or spillover. Artificial intelligence (AI), and particularly its machine learning and deep learning branches, holds enormous potential for detecting shifts in large-scale biodiversity and disease datasets (genomic, ecological, geospatial, etc.). Such algorithms can be trained to identify subtle patterns in large volumes of data to yield insights into complex phenomena for which we have limited knowledge of the true cause(s) or predictor(s), as is the case for emerging infectious diseases.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Improving the Quality of Geothermal Data Through Data Standards and Pipelines Within the Geothermal Data Repository: Preprint

For machine learning outputs to be applicable to real world problems, high quality data are needed to ensure high quality results. With the more recent emphasis on machine learning in geothermal, there is an increasing need for greater focus on the quality of the data available for use in these projects. For example, Geothermal Operational Optimization Using Machine Learning (GOOML) utilized large quantities of geothermal power plant operational data to inform power plant operational configurations to maximize power generation. High quality datasets result from dependable sensors or devices collecting data, high frequency of measurements, sufficient data points, adequate metadata, reliable storage of data, and sufficient data curation. Another component that contributes to high quality data is reusability, which can be enhanced through data standardization. Data Standardization creates consistency in formatting and contents of like datasets, lessening preprocessing requirements and ensuring adequate information provided by a given dataset. The Geothermal Data Repository (GDR) aims to help improve data quality through automated data standardization for high-value datasets through the implementation of data pipelines alongside reliable and accessible long-term storage for datasets. As such, the GDR has decided to shift away from recommending the use of Excel-based content models and towards the implementation of automated data pipelines. This takes the burden of data standardization off the user and project team and will increase the availability of standardized geothermal data available through the GDR. A set of recommendations, or a data standard for each data type will exist with each data pipeline in order to advise data collection for maximum usability for future research. This paper serves to describe the GDR's proposed transition towards data standardization through automated data pipelines, to discuss the need for and value of such a shift, and to call for suggestions from the community regarding the most useful data standards and pipelines.

data↗
NASA NTRS2022

Optimizing Landsat Next Shortwave Infrared Bands for Crop Residue Characterization

This study focused on optimizing the placement of shortwave infrared (SWIR) bands for pixel-level estimation of fractional crop residue cover (f R ) for the upcoming Landsat Next mission. We applied an iterative wavelength shift approach to a database of crop residue field spectra collected in Beltsville, Maryland, USA (n = 916) and computed generalized two- and three-band spectral indices for all wavelength combinations between 2000 and 2350 nm, then used these indices to model field-measured f R . A subset of the full dataset with a Normalized Difference Vegetation Index (NDVI) < 0.3 threshold (n = 643) was generated to evaluate green vegetation impacts on f R estimation. For the two-band wavelength shift analyses applied to the NDVI < 0.3 dataset, a generalized normalized difference using 2226 nm and 2263 nm bands produced the top f R estimation performance (R 2 = 0.8222; RMSE = 0.1296). These findings were similar to the established two-band Shortwave Infrared Normalized Difference Residue Index (SINDRI) (R 2 = 0.8145; RMSE = 0.1324). Performance of the two-band generalized normalized difference and SINDRI decreased for the full-NDVI dataset (R 2 = 0.5865 and 0.4144, respectively). For the three-band wavelength shift analyses applied to the NDVI < 0.3 dataset, a generalized ratio-based index with a 2031–2085–2216 nm band combination, closely matching established Cellulose Absorption Index (CAI) bands, was top performing (R 2 = 0.8397; RMSE = 0.1231). Three-band indices with CAI-type wavelengths maintained top f R estimation performance for the full-NDVI dataset with a 2036–2111–2217 nm band combination (R 2 = 0.7581; RMSE = 0.1548). The 2036–2111–2217 nm band combination was also top performing in f R estimation (R 2 = 0.8690; RMSE = 0.0970) for an additional analysis assessing combined green vegetation cover and surface moisture effects. Our results indicate that a three-band configuration with band centers and wavelength tolerances of 2036 nm (±5 nm), 2097 nm (±14 nm), and 2214 (±11 nm) would optimize Landsat Next SWIR bands for f R estimation.

Landsat Next↗
DOE OSTI2025

Aspen Open Jets: unlocking LHC data for foundation models in particle physics

Foundation models are deep learning models pre-trained on large amounts of data which are capable of generalizing to multiple datasets and/or downstream tasks. This work demonstrates how data collected by the CMS experiment at the Large Hadron Collider can be useful in pre-training foundation models for HEP. Specifically, we introduce the AspenOpenJets (AOJs) dataset, consisting of approximately 178 M high p T jets derived from CMS 2016 Open Data. We show how pre-training the OmniJet-α foundation model on AOJs improves performance on generative tasks with significant domain shift: generating boosted top and QCD jets from the simulated JetClass dataset. In addition to demonstrating the power of pre-training of a jet-based foundation model on actual proton–proton collision data, we provide the ML-ready derived AOJs dataset for further public use.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Neural Image Compression: Generalization, Robustness, and Spectral Biases

Recent advances in neural image compression (NIC) have resulted in models which are starting to outperform traditional codecs. While this has led to growing excitement about using these methods in real-world applications, the successful adoption of any machine learning system (including NIC) in the wild requires it to generalize (and be robust) to unseen distribution shifts at deployment time. Unfortunately, current research lacks comprehensive datasets and informative tools to evaluate and understand compression performance in real-world settings. To bridge this crucial gap, first, this paper presents a comprehensive benchmark suite to evaluate the out-of-distribution (OOD) performance of image compression methods. Specifically, we design CLIC-C and Kodak-C by introducing 15 common corruptions to popular CLIC and Kodak benchmarks. Next, we propose spectrally inspired introspection tools to gain a deeper understanding of errors introduced by image compression methods as well as their OOD performance. To this end, we carry out a detailed performance comparison of the classical codec with various variants of NIC (e.g., original, variable rate, pruned), revealing intriguing findings that challenge our current understanding of the strengths and limitations of NIC. Finally, we corroborate our empirical findings with theoretical analysis, providing an in-depth view of the OOD performance of NIC. Our benchmarks, spectral introspection tools, and findings provide a crucial bridge to the real-world adoption of NIC. We hope that our work will propel future efforts in designing more robust and generalizable NIC methods.

neural networks, Variational Autoencoder, robustne↗
DOE OSTI2023

Dynamic subcanopy leaf traits drive resistance of net primary production across a disturbance severity gradient

Across the globe, the forest carbon sink is increasingly vulnerable to an expanding array of low- to moderate-severity disturbances. However, some forest ecosystems exhibit functional resistance (i.e., the capacity of ecosystems to continue functioning as usual) following disturbances such as extreme weather events and insect or fungal pathogen outbreaks. Unlike severe disturbances (e.g., stand-replacing wildfires), moderate severity disturbances do not always result in near-term declines in forest production because of the potential for compensatory growth, including enhanced subcanopy production. Community-wide shifts in subcanopy plant functional traits, prompted by disturbance-driven environmental change, may play a key mechanistic role in resisting declines in net primary production (NPP) up to thresholds of canopy loss. However, the temporal dynamics of these shifts, as well as the upper limits of disturbance for which subcanopy production can compensate, remain poorly characterized. In this study, we leverage a 4-year dataset from an experimental forest disturbance in northern Michigan to assess subcanopy community trait shifts as well as their utility in predicting ecosystem NPP resistance across a wide range of implemented disturbance severities. Through mechanical girdling of stems, we achieved a gradient of severity from 0% (i.e., control) to 45, 65, and 85% targeted gross canopy defoliation, replicated across four landscape ecosystems broadly representative of the Upper Great Lakes ecoregion. We found that three of four examined subcanopy community weighted mean (CWM) traits including leaf photosynthetic rate (p = 0.04), stomatal conductance (p = 0.07), and the red edge normalized difference vegetation index (p < 0.0001) shifted rapidly following disturbance but before widespread changes in subcanopy light environment triggered by canopy tree mortality. Surprisingly, stimulated subcanopy production fully compensated for upper canopy losses across our gradient of experimental severities, achieving complete resistance (i.e., no significant interannual differences from control) of whole ecosystem NPP even in the 85% disturbance treatment. Additionally, we identified a probable mechanistic switch from nutrient-driven to light-driven trait shifts as disturbance progressed. Our findings suggest that remotely sensed traits such as the red edge normalized difference vegetation index (reNDVI) could be particularly sensitive and robust predictors of production response to disturbance, even across compositionally diverse forests. The potential of leaf spectral indices to predict post-disturbance functional resistance is promising given the capabilities of airborne to satellite remote sensing. We conclude that dynamic functional trait shifts following disturbance can be used to predict production response across a wide range of disturbance severities.

resistance↗
DOE OSTI2026

CONUS-wide Projected Flood Frequency and Uncertainty Estimates, Version 1.0

This dataset presents a large-ensemble of CONUS-wide projected flood frequency and uncertainty estimates across ~2.7 million NHDPlusV2 river reaches over the CONUS. The framework producing this dataset leverages a multi-model, uncertainty-aware modeling framework that allows evaluating shifts in flood frequences at the stream reach level across the CONUS. CONUS-wide ensemble streamflow projections generated from hydrologic simulations driven by downscaled and bias-corrected Coupled Model Intercomparison Project Phase 6 (CMIP6) outputs are used to derive these flood frequency and uncertainty estimates over the period 1980 - 2099. A spatially consistent regional L-moment algorithm is applied across clusters defined by the US Hydrologic Unit Code Subregions (HUC4s and HUC8s) and NHDPlusV2 stream orders to estimate flood frequencies. The dataset also includes at-site based flood estimates that allow for the comparison between local and regional approach-based estimates, assess projected changes, and characterize their uncertainties. For more reliable estimation of rare flood frequencies such as 500 and 1000-year return periods, super-ensemble based estimates are also included in the dataset. This dataset is derived to support the "Impact-Informed Dam Safety Risk Assessment for Securing Hydropower Assests" project for the US Department of Energy (DOE) Hydropower and Hydrokinetic Office (H2O). For further details, refer to Kao et al. (2022), Ghimire et al. (2023), Ghimire et al. (2025), and Hosking and Wallis (1997).

Ghimire, Ganesh [ORNL] (ORCID:0000000242843941)↗