Search NASA⌕ Search

SEARCH · Search NASA

Results for “large scale network”

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

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

At least 937 records · Page 52
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Real-time signal detection for Cyclotron Radiation Emission Spectroscopy measurements using antenna arrays

Cyclotron Radiation Emission Spectroscopy (CRES) is a technique for precision measurement of the energies of charged particles, which is being developed by the Project 8 Collaboration to measure the neutrino mass using tritium beta-decay spectroscopy. Project 8 seeks to use the CRES technique to measure the neutrino mass with a sensitivity of 40 meV, requiring a large supply of tritium atoms stored in a multi-cubic meter detector volume. Antenna arrays are one potential technology compatible with an experiment of this scale, but the capability of an antenna-based CRES experiment to measure the neutrino mass depends on the efficiency of the signal detection algorithms. Here, in this paper, we develop efficiency models for three signal detection algorithms and compare them using simulations from a prototype antenna-based CRES experiment as a case-study. The algorithms include a power threshold, a matched filter template bank, and a neural network based machine learning approach, which are analyzed in terms of their average detection efficiency and relative computational cost. It is found that significant improvements in detection efficiency and, therefore, neutrino mass sensitivity are achievable, with only a moderate increase in computation cost, by utilizing either the matched filter or machine learning approach in place of a power threshold, which is the baseline signal detection algorithm used in previous CRES experiments by Project 8.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

Signal Processing Based Method for Real-Time Anomaly Detection in High-Performance Computing

Performance anomalies can manifest as irregular execution times or abnormal execution events for many reasons, including network congestion and resource contention. Detecting such anomalies in real-time by analyzing the details of performance traces at scale is impractical due to the sheer volume of data High-Performance Computing (HPC) applications produce. In this paper, we propose formulating HPC performance anomaly detection as a signal-processing problem where anomalies can be treated as noise. We evaluate our proposed method in comparison with two other commonly used anomaly detection techniques of varying complexity based on their detection accuracy and scalability. Since real-time in-situ anomaly detection at a large scale requires lightweight methods that can handle a large volume of streaming data, we find that our proposed method provides the best trade-off. We then implement the proposed method in Chimbuko, the first online, distributed, and scalable workflow-level performance trace analysis framework. We compare our proposed signal-based anomaly detection algorithm with two other methods using a function of their accuracy, F1 score, and detection overhead. Our experiments demonstrate that our proposed approach achieves a 99% improvement for the benchmark datasets and a 93% improvement with Chimbuko traces.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2024

Driving Uptake for Energy Efficiency Financing Programs: Marketing and Outreach, Partnership Networks, and Program Design Considerations

Many energy efficiency financing programs could achieve greater uptake and impact by more effectively recruiting participants. This report examines some of the primary factors that have contributed to high participant uptake among successful financing programs. We review best practices in partnerships (Chapter 2), direct marketing (Chapter 3), and program design (Chapter 4) that facilitate robust participation. This report is primarily designed for state and local governments that have established energy efficiency financing programs or are considering doing so and are seeking insight into how they can ramp up program participation. In disseminating lessons learned from well-established programs that have experienced success in their target markets, the objective is to help scale up the large number of energy efficiency financing programs that seek to replicate these successes. This report can inform states, local governments, and other entities that will establish or expand clean energy financing programs with funding made available under the Infrastructure Investment and Jobs Act and the Inflation Reduction Act.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2025

Harnessing distributed GPU computing for generalizable graph convolutional networks in power grid reliability assessments

Although machine learning (ML) has emerged as a powerful tool for rapidly assessing grid contingencies, prior studies have largely considered a static grid topology in their analyses. This limits their application, since they need to be re-trained for every new topology. Here, this paper explores the development of generalizable graph convolutional network (GCN) models by pre-training them across a range of grid topologies and contingency types. We found that a GCN model with auto-regressive moving average (ARMA) layers with a line graph representation of the grid offered the best predictive performance in predicting voltage magnitudes (VM) and voltage angles (VA). We introduced the concept of phantom nodes to consider disparate grid topologies with a varying number of nodes and lines. For pre-training the GCN ARMA model across a variety of topologies, distributed graphics processing unit (GPU) computing afforded us significant training scalability. The predictive performance of this model on grid topologies that were part of the training data is substantially better than the direct current (DC) approximation. Although direct application of the pre-trained model to topologies that are not part of the grid is not particularly satisfactory, fine-tuning with small amounts of data from a specific topology of interest significantly improves predictive performance. In general, this paper highlights the feasibility of training large-scale GNN models to assess the reliability of power grids by considering a wide variety of grid topologies and contingency types. With the advent of foundational models in ML and the exponential increase in GPU computing clusters, generalizable ML models will significantly enhance how utilities manage power systems and make decisions in real-time or near-real-time.

24 - POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2023

Ab initio study of the structure and properties of amorphous silicon hydride from accelerated molecular dynamics simulations

This paper presents a large-scale ab initio simulation study of amorphous silicon hydride (a-Si 1-x H x ) with an emphasis on the structure and properties of the material across a range of hydrogen concentration by combining accelerated molecular dynamics (MD) simulations with first-principles density-functional calculations. The accelerated MD scheme relied on classical metadynamics, which enabled the development of 2500+ high-quality structural models of a-Si 1-x H x , with system sizes ranging from 150 to 6000 atoms and hydrogen concentrations vary from 6 to 20 at. %. The resulting amorphous networks were found to be completely free from any coordination defects and that they all exhibited a pristine band-gap in their electronic spectrum. The microstructural properties of hydrogen distributions were examined with an emphasis on the presence of isolated and clustered environments of hydrogen atoms. The results were compared with experimental data obtained from X-ray diffraction, infrared spectroscopy and nuclear magnetic resonance studies.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Model fusion with physics-guided machine learning: Projection-based reduced-order modeling

The unprecedented amount of data generated from experiments, field observations, and large-scale numerical simulations at a wide range of spatiotemporal scales has enabled the rapid advancement of data-driven and especially deep learning models in the field of fluid mechanics. Although these methods are proven successful for many applications, there is a grand challenge of improving their generalizability. This is particularly essential when data-driven models are employed within outer-loop applications like optimization. In this work, we put forth a physics-guided machine learning (PGML) framework that leverages the interpretable physics-based model with a deep learning model. Leveraging a concatenated neural network design from multi-modal data sources, the PGML framework is capable of enhancing the generalizability of data-driven models and effectively protects against or inform about the inaccurate predictions resulting from extrapolation. We apply the PGML framework as a novel model fusion approach combining the physics-based Galerkin projection model and long- to short-term memory (LSTM) network for parametric model order reduction of fluid flows. We demonstrate the improved generalizability of the PGML framework against a purely data-driven approach through the injection of physics features into intermediate LSTM layers. Our quantitative analysis shows that the overall model uncertainty can be reduced through the PGML approach, especially for test data coming from a distribution different than the training data. Moreover, we demonstrate that our approach can be used as an inverse diagnostic tool providing a confidence score associated with models and observations. The proposed framework also allows for multi-fidelity computing by making use of low-fidelity models in the online deployment of quantified data-driven models.

42 ENGINEERING↗
DOE OSTI2023

AOI-2, A Novel Access Control Blockchain Paradigm for Cybersecure Sensor Infrastructure in Fossil Power Generation Systems

Fossil power generation systems are increasingly vulnerable to attack from both cybercriminals as well as internal threats. These vulnerabilities demand that emerging technologies such as blockchains be utilized to secure the data involved in the information flows within the Supervisory Control and Data Acquisition (SCADA) systems of the fossil power generation plants. The publicly accessible blockchain protocols, although secure, are visible to everyone. Even private blockchains currently are unable to support different levels of access to different participants, which is a critical requirement for the existing SCADA systems running the power plants. In light of the above, novel blockchain protocols that are specifically adapted to fossil power generation environments need to be developed in order to achieve the goal of cybersecure sensor networks. In this work, we address this question by creating a novel blockchain technology, namely smart private ledger, for cybersecure communication within the fossil power generation systems. A lab-scale sensor network consisting of strain and temperature sensors is constructed to develop the ledger. The technology has hierarchical access control which is compatible with the existing SCADA systems in fossil power plants. The sensor data is used with cryptographic digital signatures and secret sharing protocols within the nodes of the blockchain technology. The research results will lead to cybersecurity for machine-to-machine interactions, infrastructure for secure data logging for sensors, decentralized data storage, and second-layer technologies for high volume machine-to-machine interactions in the power plants. The work aims to largely address the concerns for the security of distributed sensor networks in such systems that can be compromised by insider threats and by cybercriminals. The research has led to the training of the next generation of engineers and scientists in the important areas of sensor engineering and blockchain technology.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2024

Spatially Local Surrogate Modeling of Subgrid-Scale Effects in Idealized Atmospheric Flows: A Deep Learned Approach Using High-Resolution Simulation Data

Abstract We introduce a machine learned surrogate model from high-resolution simulation data to capture the subgrid-scale effects in dry, stratified atmospheric flows. We use deep neural networks (NNs) to model the spatially local state differences between a coarse-resolution simulation and a high-resolution simulation. The setup enables the capture of both dissipative and antidissipative effects in the state differences. The NN model is able to accurately capture the state differences in offline tests outside the training regime. In online tests intended for production use, the NN-coupled coarse simulation has higher accuracy over a significant period of time compared to the coarse-resolution simulation without any correction. We provide evidence of the capability of the NN model to accurately capture high-gradient regions in the flow field. With the accumulation of the errors, the NN-coupled simulation becomes computationally unstable after approximately 90 coarse simulation time steps. Insights gained from these surrogate models further pave the way for formulating stable, complex, physics-based spatially local NN models which are driven by traditional subgrid-scale turbulence closure models. Significance Statement Flows in the atmosphere are highly chaotic and turbulent, comprising flow structures of broad scales. For effective computational modeling of atmospheric flows, the effects of the small- and large-scale structures need to be captured by the simulations. Capturing the small-scale structures requires fine-resolution simulations. Even with the current state-of-the-art supercomputers, it can be prohibitively expensive to simulate these flows when computed for the entire earth over climate time scales. Thus, it is necessary to focus on the larger-scale structures using a coarse-resolution simulation while capturing the effects of the smaller-scale structures using some parameterization (approximation) scheme and incorporating it into the coarse-resolution simulation. We use machine learning to model the effects of the small-scale structures (subgrid-scale effects) in atmospheric flows. Data from a fine-resolution simulation is used to compute the missing subgrid-scale effects in coarse-resolution simulations. We then use machine learning models to approximate these differences between the coarse- and fine-resolution simulations. We see improved accuracy for the coarse-resolution simulations when corrected using these machine learned models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Ab initio study of the structure and properties of amorphous silicon hydride from accelerated molecular dynamics simulations

This paper presents a large-scale ab initio simulation study of amorphous silicon hydride (a-Si 1-x H x ) with an emphasis on the structure and properties of the material across a range of hydrogen concentration by combining accelerated molecular dynamics (MD) simulations with first-principles density-functional calculations. The accelerated MD scheme relied on classical metadynamics, which enabled the development of 2600+ high-quality structural models of a-Si 1-x H x , with system sizes ranging from 150 to 6,000 atoms and hydrogen concentrations vary from 6 to 20 at. %. The resulting amorphous networks were found to be completely free from any coordination defects and that they all exhibited a pristine band-gap in their electronic spectrum. The microstructural properties of hydrogen distributions were examined with great emphasis on the presence of isolated and clustered environments of hydrogen atoms. The results were compared with a suite of experimental data obtained from x-ray diffraction, infrared spectroscopy, spectroscopic ellipsometry and nuclear magnetic resonance studies.

36 MATERIALS SCIENCE↗
DOE OSTI2020

A multirate mass transfer model to represent the interaction of multicomponent biogeochemical processes between surface water and hyporheic zones (SWAT-MRMT-R 1.0)

Abstract. Surface water quality along river corridors can be modulated by hyporheic zones (HZs) that are ubiquitous and biogeochemically active. Watershed management practices often ignore the potentially important role of HZs as a natural reactor. To investigate the effect of hydrological exchange and biogeochemical processes on the fate of nutrients in surface water and HZs, a novel model, SWAT-MRMT-R, was developed coupling the Soil and Water Assessment Tool (SWAT) watershed model and the reaction module from a flow and reactive transport code (PFLOTRAN). SWAT-MRMT-R simulates concurrent nonlinear multicomponent biogeochemical reactions in both the channel water and its surrounding HZs, connecting the channel water and HZs through hyporheic exchanges using multirate mass transfer (MRMT) representation. Within the model, HZs are conceptualized as transient storage zones with distinguished exchange rates and residence times. The biogeochemical processes within HZs are different from those in the channel water. Hyporheic exchanges are modeled as multiple first-order mass transfers between the channel water and HZs. As a numerical example, SWAT-MRMT-R is applied to the Hanford Reach of the Columbia River, a large river in the United States, focusing on nitrate dynamics in the channel water. Major nitrate contaminants entering the Hanford Reach include those from the legacy waste, irrigation return flows (irrigation water that is not consumed by crops and runs off as point sources to the stream), and groundwater seepage resulting from irrigated agriculture. A two-step reaction sequence for denitrification and an aerobic respiration reaction is assumed to represent the biogeochemical transformations taking place within the HZs. The spatially variable hyporheic exchange rates and residence times in this example are estimated with the basin-scale Networks with EXchange and Subsurface Storage (NEXSS) model. Our simulation results show that (1), given a residence time distribution, how the exchange fluxes to HZs are approximated when using MRMT can significantly change the amount of nitrate consumption in HZs through denitrification and (2) source locations of nitrate have a different impact on surface water quality due to the spatially variable hyporheic exchanges.

58 GEOSCIENCES↗
DOE OSTI2024

Measure this, not that: Optimizing the cost and model-based information content of measurements

Model-based design of experiments (MBDoE) is a powerful framework for selecting and calibrating science-based mathematical models from data. Here, this work extends popular MBDoE workflows by proposing a convex mixed integer (non)linear programming (MINLP) to optimize the selection of measurements. The solver MindtPy is modified to support calculating the D-optimality objective and its gradient via an external package, scipy, using the grey-box module in Pyomo. The new approach is demonstrated in two case studies: estimating highly correlated kinetics from a batch reactor and estimating transport parameters in a large-scale rotary packed bed for CO 2 capture. Both case studies show how examining the Pareto optimal trade-offs between information content measured by A- and D-optimality versus measurement budget offers practical guidance for selecting measurements for scientific experiments.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

The Historical Development of Large‐Scale Paleoclimate Field Reconstructions Over the Common Era

Abstract Climate field reconstructions (CFRs) combine modern observational data with paleoclimatic proxies to estimate climate variables over spatiotemporal grids during time periods when widespread observations of climatic conditions do not exist. The Common Era (CE) has been a period over which many seasonally‐ and annually‐resolved CFRs have been produced on regional to global scales. CFRs over the CE were first produced in the 1970s using dendroclimatic records and linear regression‐based approaches. Since that time, many new CFRs have been produced using a wide range of proxy data sets and reconstruction techniques. We assess the early history of research on CFRs for the CE, which provides context for our review of advances in CFR research over the last two decades. We review efforts to derive gridded hydroclimatic CFRs over continental regions using networks of tree‐ring proxies. We subsequently explore work to produce hemispheric‐ and global‐scale CFRs of surface temperature using multi‐proxy data sets, before specifically reviewing recently‐developed data assimilation techniques and how they have been used to produce simultaneous reconstructions of multiple climatic fields globally. We then review efforts to develop standardized and digitized databases of proxy networks for use in CFR research, before concluding with some thoughts on important next steps for CFR development.

Smerdon, Jason E.↗
DOE OSTI2022

Understanding Biomass and Polymer Feedstock Variability Through Analytical Pyrolysis and Two Dimensional Gas Chromatography and Mass Spectrometry

Moving toward a circular carbon economy requires enabling reuse of carbon-based macromolecules like those in biomass and polymers. Meeting this goal requires increased upcycling of waste products into higher value products. For biomass waste such as corn stover, pine needles, or bark, one approach is using pyrolysis to convert these feedstocks into bio-oil that can be used for liquid fuel or transformed into other hydrocarbon based products, like bio-polymers. Waste plastics can be upcycled or recycled into new products, or alternatively converted into bio-oil by pyrolysis. Macromolecules from biomass and polymers are challenging to convert into liquid fuel or chemical feedstocks, due to large and sometimes heterogeneous networks of polymeric bonds. We work with the Idaho National Laboratory’s Biomass Feedstock National User Facility and Department of Energy’s Feedstock Conversion Interface Consortium to develop a molecular understanding of biomass variability that occurs naturally and as a result of storage or preprocessing techniques applied to feedstocks, to understand chemical reactions occurring during pyrolysis and other biomass or polymer conversion techniques. Using two dimensional gas chromatography mass spectrometry coupled to analytical pyrolysis provides insight into the molecular changes that occur during the pyrolysis event, and these insights can extend to bench or pilot scale pyrolysis conversions. Understanding pyrolysis of biomass and polymers at a molecular level contributes to development of new tools to predict and manipulate the optimum conditions and identify the best storage, preprocessing, and conversion techniques to maximize output of specific high-value chemical species in the conversion to fuels, or to characterize feedstocks for blending before pyrolysis. Identification and understanding of the molecular level changes resulting from novel preprocessing techniques that can be used to control and manipulate the chemical output of bench or pilot scale pyrolysis is another focus of investigation, including the use of gamma irradiation, acid, or enzyme pre-treatments. We also focus on the impacts of storage conditions and techniques, which can affect moisture levels, degradation, and impacts from naturally occurring microbes in the environment.

09 BIOMASS FUELS↗
DOE OSTI2021

Reconstructing patchy reionization with deep learning

The precision anticipated from next-generation cosmic microwave background (CMB) surveys will create opportunities for characteristically new insights into cosmology. Secondary anisotropies of the CMB will have an increased importance in forthcoming surveys, due both to the cosmological information they encode and the role they play in obscuring our view of the primary fluctuations. Quadratic estimators have become the standard tools for reconstructing the fields that distort the primary CMB and produce secondary anisotropies. While successful for lensing reconstruction with current data, quadratic estimators will be suboptimal for the reconstruction of lensing and other effects at the expected sensitivity of the upcoming CMB surveys. In this paper we describe a convolutional neural network, ResUNet-CMB, that is capable of the simultaneous reconstruction of two sources of secondary CMB anisotropies, gravitational lensing and patchy reionization. Furthermore, we show that the ResUNet-CMB network significantly outperforms the quadratic estimator at low noise levels and is not subject to the lensing-induced bias on the patchy reionization reconstruction that would be present with a straightforward application of the quadratic estimator.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

End–to–End Metasurface Design for Temperature Imaging via Broadband Planck‐Radiation Regression

A theoretical framework is presented for temperature imaging from long-wavelength infrared (LWIR) thermal radiation (e.g., 8–12 µm) through the end-to-end design of a metasurface-optics frontend and a computational-reconstruction backend. A new nonlinear reconstruction algorithm, “Planck regression”, is introduced to reconstruct the temperature map from a gray scale sensor image, even in the presence of severe chromatic aberration, by exploiting black body and optical physics particular to thermal imaging. This algorithm is combined with an end-to-end approach that optimizes manufacturable, single-layer metasurfaces to yield the most accurate reconstruction. The designs demonstrate high-quality, noise-robust reconstructions of arbitrary temperature maps (including completely random images) in simulations of an ultra-compact thermal-imaging device. Here, it is also shown that Planck regression is much more generalizable to arbitrary images than a straightforward neural-network reconstruction, which requires a large training set of domain-specific images.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Projections of winter polynyas and their biophysical impacts in the Ross Sea Antarctica

Abstract This study investigates winter polynyas in the southern Ross Sea, Antarctica where several polynyas are known to form. Coastal polynyas are areas of lower sea ice concentration and/or thickness along the coast that are otherwise surrounded by more extensive, thicker sea ice pack. Polynyas are also locations where organisms can exploit both the ice substrate and pelagic resources. Using a self organizing map algorithm, we identify polynya events in the Community Earth System Model Version 2 Large Ensemble (CESM2-LE). The neural network algorithm is able to identify polynya events without imposing an ice concentration or thickness threshold, as is often done when identifying polynyas. The CESM2-LE produces a wintertime polynya feature comparable in size and location to the Ross Sea polynya, and during polynya events there are large turbulent heat fluxes and export of sea ice from the Ross Sea. In the CESM2-LE polynya event frequency is projected to decrease sharply in the later twentyfirst century, leading to increasing sea ice concentrations and thicknesses in the region. The drivers of the polynya frequency decline are likely both large scale circulation changes and local atmosphere and ocean feedbacks. If declines in wintertime polynya frequency over the twentyfirst century do occur they may impact Antarctic Bottom Water formation and local net primary productivity. Thus, better understanding potential local and unexpected sea ice changes in the Ross Sea is important for both assessing climate system impacts and ecological impacts on the Ross Sea ecosystem, which is currently protected by an internationally recognized marine protected area.

Meteorology & Atmospheric Sciences↗
DOE OSTI2020

Pump-and-Treat Impacts on Aquifer Conditions and the Development of System-Wide Performance Strategies - 20377

The pump-and-treat (P and T) system on the Hanford Site's Central Plateau currently treats multiple contaminants within the 200-ZP-1 operable unit, including nitrate through its biological treatment processes. Although it was initially designed to support the 200-ZP-1 Remedial Action Objectives, the P and T system has evolved in use to address contaminants from other regions of the aquifer as well. The P and T influents are biogeochemically distinct, treating groundwaters with elevated concentrations of cyanide, uranium, technetium-99, and nitrate. Given the varying geochemical signatures from current and potential future sources, operational conditions, and amendments considered for treatment processes, there is a need to identify the impacts that the P and T effluent has on the aquifer, and how the aquifer affects P and T performance. To this end, laboratory efforts have focused on characterizing P and T effluent and its biogeochemical effects on the sediments through a series repacked column experiments. The P and T effluent is rich in particulate matter. A systematic inspection of this particulate matter in the P and T effluent showed significant levels of manganese/iron oxides, calcite crystals, relatively large organic matter molecules, and biomass. Therefore, the column experiments were designed and conducted for four different treatment systems: artificial groundwater, filtered P and T effluent, non-filtered P and T effluent, and chlorinated P and T effluent. Different levels of microbiologically- and geochemically-induced changes in the columns occurred and modified the flow systems. For example, physical filtration occurred in the sediments where the particulate material had accumulated in the top portion of the column. Reactive transport modeling of the laboratory experiments is being carried out so that the reaction network can be extrapolated to the field scale. These simulations will provide predictive evaluations of system performance based on P and T operational changes and aquifer mass loading and capacity estimates. This information can be used to address combined subsurface and operational challenges, such as the need for frequent well redevelopment to maintain injection capacity. This information also provides the technical basis for decisions related to the P and T system operations that support remedy optimization efforts and short- and long-term remedy decisions related to multiple operable units on the Hanford Site Central Plateau. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

Enhancing Streamflow Forecast and Extracting Insights Using Long-Short Term Memory Networks With Data Integration at Continental Scales

Recent observations with varied schedules and types (moving average, snapshot, or regularly spaced) can help to improve streamflow forecasts, but it is challenging to integrate them effectively. Based on a long short-term memory (LSTM) streamflow model, we tested multiple versions of a flexible procedure we call data integration (DI) to leverage recent discharge measurements to improve forecasts. DI accepts lagged inputs either directly or through a convolutional neural network unit. DI ubiquitously elevated streamflow forecast performance to unseen levels, reaching a record continental-scale median Nash-Sutcliffe Efficiency coefficient value of 0.86. Integrating moving-average discharge, discharge from the last few days, or even average discharge from the previous calendar month could all improve daily forecasts. Directly using lagged observations as inputs was comparable in performance to using the convolutional neural network unit. Importantly, we obtained valuable insights regarding hydrologic processes impacting LSTM and DI performance. Before applying DI, the base LSTM model worked well in mountainous or snow-dominated regions, but less well in regions with low discharge volumes (due to either low precipitation or high precipitation-energy synchronicity) and large interannual storage variability. DI was most beneficial in regions with high flow autocorrelation: it greatly reduced baseflow bias in groundwater-dominated western basins and also improved peak prediction for basins with dynamical surface water storage, such as the Prairie Potholes or Great Lakes regions. However, even DI cannot elevate performance in high-aridity basins with 1-day flash peaks. Despite this limitation, there is much promise for a deep-learning-based forecast paradigm due to its performance, automation, efficiency, and flexibility.

54 ENVIRONMENTAL SCIENCES↗