Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Competition”

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

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

At least 1,567 records · Page 87
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Aqueous Carbon Capture Using Guanidinium-Functionalized Hollow Fiber Sorbent Contactors

As part of the growing suite of technologies aimed at combatting rising temperatures, negative emissions technologies have become a powerful tool in the global effort to minimize the consequences of human-induced climate change. Among these, carbon removal from aqueous sources, which contain much higher carbon concentrations than the atmosphere, remains largely unexplored. Indeed, developing robust and efficient carbon capture materials for usage in complex aqueous environments remains a significant challenge. Here, we explore the potential of functionalizing polyvinylidene fluoride (PVDF) hollow fiber contactors grafted with a guanidinium-derived polymer sorbent for carbon removal from aqueous sources, including saline waters. Computational screening against amine-based analogs is utilized to identify guanidinium as a promising motif for bicarbonate (HCO 3 – ) ions binding. To leverage this finding, synthesis of a guanidinium polymer and subsequent covalent grafting onto PVDF hollow fibers is employed to structured polymer–sorbent–grafted hollow fiber contactors. Our prototype achieves an initial HCO 3 – removal of 34% with an increase to 98% after four cycles. The functionalized fibers demonstrate aqueous stability over 13 adsorption/desorption cycles in model NaHCO 3 solutions where regeneration is facilitated by a mild pH swing. Importantly, the system maintains selective performance in the presence of competitive chloride ions over multiple cycles; carbon removal remained above 10% even at high (10:1) NaCl/NaHCO 3 ratios. These findings demonstrate the feasibility of sorbent-based aqueous carbon removal and highlight its potential as a promising approach for negative emissions.

carbon capture↗
DOE OSTI2020

High Mass Analysis with a Fourier Transform Ion Cyclotron Resonance Mass Spectrometer: From Inorganic Salt Clusters to Antibody Conjugates and Beyond

Analysis of proteins and complexes under native mass spectrometric (MS) and solution conditions was typically performed using time-of-flight (ToF) analyzers, due to their routine high m/z transmission and detection capabilities. However, over recent years, the ability of Orbitrap-based mass spectrometers to transmit and detect a range of high molecular weight species is well documented. Herein, we describe how a 15 Tesla Fourier transform ion cyclotron resonance mass spectrometer (15 T FT-ICR MS) is more than capable of analyzing a wide range of ions in the high m/z scale (>5000), in both positive and negative instrument polarities, ranging from the inorganic cesium iodide salt clusters; a humanized IgG1k monoclonal antibody (mAb; 148.2 kDa); an IgG1-mertansine drug conjugate (148.5 kDa, drug-to-antibody ratio; DAR 2.26); an IgG1-siRNA conjugate (159.1 kDa; ribonucleic acid to antibody ratio; RAR 1); the membrane protein aquaporin-Z (97.2 kDa) liberated from a C8E4 detergent micelle; the empty MSP1D1-nanodisc (142.5 kDa) and the tetradecameric chaperone protein complex GroEL (806.2 kDa; GroEL dimer at 1.6 MDa). We also investigate different regions of the FT-ICR MS that impact ion transmission and desolvation. Finally, we demonstrate how the transmission of these species and resultant spectra are highly consistent with those previously generated on both quadrupole-ToF (Q-ToF) and Orbitrap instrumentation. This article serves as an impactful example of how FT-ICR mass analyzers are competitive to Q-ToFs and Orbitraps for high mass detection at high m/z.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Aerosol Impacts on Mesoscale Convective Systems Forming Under Different Vertical Wind Shear Conditions

Following our previous study of wind shear effect on mesoscale convective system (MCS) organization under a clean atmospheric condition using the Weather Research and Forecasting model coupled with Spectral-Bin Microphysics, we conduct sensitivity simulations by increasing cloud condensation nuclei (CCN) concentration and investigate aerosol impacts on MCSs forming under different wind shear conditions. We find that, aerosols induce stronger updrafts and downdrafts in all MCSs. The stronger updrafts and enlarged convective core area contribute to larger vertical mass fluxes and enhance precipitation, indicating convective invigoration. Increased updraft speed below 8-km altitude mainly results from enhanced condensational heating, and increased water loading effect is not large enough to overcome the increase in thermal buoyancy. Interestingly, above 8-km altitude we see reduced updraft speed by the increased aerosols due to reduced vertical pressure perturbation gradient force, and weakened depositional growth due to competition of water vapor by enhanced condensation. The accumulated rainfall and mean rain rate are increased, with an increased occurrence frequency of heavy rain. The increase of rain rate is seen in both convective and stratiform regions, with a more significant increase in convective rain rate. In general, we see a higher frequency of deep convective cloud in the polluted condition because of invigorated convection, and more stratiform/anvil clouds, but a lower frequency of shallow warm clouds. The unanimously invigorated MCSs by aerosols under various wind shear conditions revealed by this study has an important implication in weather and climate at warm and humid regions that are influenced by pollution.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Getting to the Root of Plant–Mediated Methane Emissions and Oxidation in a Thermokarst Bog

Vascular plants are important in the wetland methane cycle, but their effect on production, oxidation, and transport has high uncertainty, limiting our ability to predict emissions. In a permafrost–thaw bog in Interior Alaska, we used plant manipulation treatments, field–deployed planar optical oxygen sensors, direct measurements of methane oxidation, and microbial DNA analyses to disentangle mechanisms by which vascular vegetation affect methane emissions. Vegetation operated on top of baseline methane emissions, which varied with proximity to the thawing permafrost margin. Emissions from vegetated plots increased over the season, resulting in cumulative seasonal methane emissions that were 4.1–5.2 g m –2 season –1 greater than unvegetated plots. Mass balance calculations signify these greater emissions were due to increased methane production (3.0–3.5 g m –2 season –1 ) and decreased methane oxidation (1.1–1.6 g m –2 season –1 ). Minimal oxidation occurred along the plant–transport pathway, and oxidation was suppressed outside the plant pathway. Our data indicate suppression of methane oxidation was stimulated by root exudates fueling competition among microbes for electron acceptors. This contention is supported by the fact that methane oxidation and relative abundance of methanotrophs decreased over the season in the presence of vegetation, but methane oxidation remained steady in unvegetated treatments; oxygen was not detected around plant roots but was detected around silicone tubes mimicking aerenchyma; and oxygen injection experiments suggested that oxygen consumption was faster in the presence of vascular vegetation. In conclusion, root exudates are known to fuel methane production, and our work provides evidence they also decrease methane oxidation.

58 GEOSCIENCES↗
DOE OSTI2022

Modeling the Economic and Environmental Impacts of Land Scarcity Under Deep Uncertainty

Abstract Land scarcity is increasing over time, driven by complex multisector dynamics. The impacts of land scarcity on the economy and environment are multi‐faceted and regional, so any action to convert land will contain inherent tradeoffs. These impacts are complicated by the deeply uncertain evolution of the various sectors influencing land scarcity. A need therefore exists to provide multi‐metric and multi‐sector assessments that are robust to myriad uncertainties. Land conservation effectively limits the supply of productive land, while biofuel consumption increases the demand and competition for that land, and how these dynamics individually and jointly propagate to economic and environmental impacts is an important open question. To address this, we adopt the Global Change Analysis Model (GCAM) that has representations of various important systems including the climate, macroeconomic, energy, agriculture and land, and water resources systems. Various scenarios of increased land demand (from biofuels) and decreased land supply (from conservation) under various socioeconomic scenarios drawn from the SSPs were simulated using GCAM. We find that while biofuel consumption and land conservation reduce carbon emissions, this comes at the cost of higher food prices, reduced crop production, and increased water withdrawals. Additionally, some regions experience these tradeoffs more severely than others and are more heavily impacted from the same biofuel mandate or by an additional percent of protected land. These and other findings highlight the importance of multisector modeling frameworks that capture many cross‐sector linkages, and acknowledge the important uncertainties confronting the human‐Earth system when making any analysis of land scarcity impacts.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Improving Seasonal Forecast Using Probabilistic Deep Learning

The path toward realizing the potential of seasonal forecasting and its socioeconomic benefits relies on improving general circulation model (GCM) based dynamical forecast systems. To improve dynamical seasonal forecasts, it is crucial to set up forecast benchmarks, and clarify forecast limitations posed by model initialization errors, formulation deficiencies, and internal climate variability. With huge costs in generating large forecast ensembles, and limited observations for forecast verification, the seasonal forecast benchmarking and diagnosing task proves challenging. Here, we develop a probabilistic deep learning-based statistical forecast methodology, drawing on a wealth of climate simulations to enhance seasonal forecast capability and forecast diagnosis. By explicitly modeling the internal climate variability and GCM formulation differences, the proposed Conditional Generative Forecasting (CGF) methodology enables bypassing crucial barriers in dynamical forecast, and offers a top-down viewpoint to examine how complicated GCMs encode the seasonal predictability information. We apply the CGF methodology for global seasonal forecast of precipitation and 2 m air temperature, based on a unique data set consisting 52,201 years of climate simulation. Results show that the CGF methodology can faithfully represent the seasonal predictability information encoded in GCMs. We successfully apply this learned relationship in real-world seasonal forecast, achieving competitive performance compared to dynamical forecasts. Using this CGF as benchmark, we reveal the impact of insufficient forecast spread sampling that limits the skill of the considered dynamical forecast system. Finally, we introduce different strategies for composing ensembles using the CGF methodology, highlighting the potential for leveraging the strengths of multiple GCMs to achieve advantgeous seasonal forecast.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Trade and Climate Mitigation Interactions Create Agro–Economic Opportunities With Social and Environmental Trade–Offs in Latin America and the Caribbean

The Latin America and the Caribbean (LAC) region plays key roles in both meeting global agricultural demands and maintaining carbon sinks due to its abundant land and water resources. In this study we use the Global Change Analysis Model (GCAM) to evaluate the opportunities and challenges posed by two global-scale drivers: agricultural market integration (i.e., reduction of trade barriers) and land-based climate mitigation policy. We evaluate their potential individual and combined impacts on agricultural production and trade revenues across LAC’s economies through mid-century, as well as the resulting impacts on agricultural consumers and integrated land-water-climate systems across the region’s diverse sub-regions. Increased global market integration results in increased agricultural production and trade revenues for many LAC economies driven by their evolving comparative advantages. Climate mitigation measures on CO 2 and non-CO 2 greenhouse gases increase revenues due to increased agricultural prices from land competition and emissions abatement. The combined outcomes from both drivers are complex and sometimes non-linear, highlighting the importance of understanding the interactions between multiple drivers. Our results show that increased agricultural production and trade opportunities pose significant trade-offs that require careful multi-sectoral planning, such as emissions reduction challenges, potential loss of livestock production when pursuing land-based climate mitigation strategies, increased consumer expenditures, and changes in land-use or water withdrawals, resulting in deforestation or water scarcity pressures. There is considerable heterogeneity in economic and environmental outcomes across LAC sub-regions and agricultural commodities, illustrating the value of considering outcomes at finer scales.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Whistler Waves Associated With Electron Beams in Magnetopause Reconnection Diffusion Regions

Abstract Whistler waves are often observed in magnetopause reconnection associated with electron beams. We analyze seven MMS crossings surrounding the electron diffusion region (EDR) to study the role of electron beams in whistler excitation. Waves have two major types: (a) Narrow‐band waves with high ellipticities and (b) broad‐band waves that are more electrostatic with significant variations in ellipticities and wave normal angles. While both types of waves are associated with electron beams, the key difference is the anisotropy of the background population, with perpendicular and parallel anisotropies, respectively. The linear instability analysis suggests that the first type of wave is mainly due to the background anisotropy, with the beam contributing additional cyclotron resonance to enhance the wave growth. The second type of broadband waves are excited via Landau resonance, and as seen in one event, the beam anisotropy induces an additional cyclotron mode. The results are supported by particle‐in‐cell simulations. We infer that the first type occurs downstream of the central EDR, where background electrons experience Betatron acceleration to form the perpendicular anisotropy; the second type occurs in the central EDR of guide field reconnection. A parametric study is conducted with linear instability analysis. A beam anisotropy alone of above ∼3 likely excites the cyclotron mode waves. Large beam drifts cause Doppler shifts and may lead to left‐hand polarizations in the ion frame. Future studies are needed to determine whether the observation covers a broader parameter regime and to understand the competition between whistler and other instabilities.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

On the Scaling and Growth Limit of Fluvial Dunes

Abstract Bedforms emerge in a variety of shapes and sizes when a granular erodible surface is subject to a strong enough shear flow, as observed in topographic data from submarine canyons, rivers, deserts, and planetary bodies. The two salient features of bedforms are the ability to collectively transport particles, and to generate form drag thus increasing flow resistance. These two mechanisms are in competition and contribute to force bedforms of increasing size migrating more slowly. In a dedicated large‐scale open channel flow facility, hierarchies of fluvial bedforms were generated and measured in equilibrium conditions. The corresponding scale‐dependent migration velocity and mass flux contributions were quantified in the wave number and frequency domains. Experimental results are paired with a validated set of theoretical models to demonstrate that ripples or dunes reach an equilibrium state when drag partitioning ensures high enough frictional drag to sustain the bedload transport of sediment, and low enough form drag to enable the migration of the largest bedform size. This mechanism is inferred to constrain the growth of bedforms when sediment supply is not the limiting factor.

58 GEOSCIENCES↗
DOE OSTI2022

Learning Coagulation Processes With Combinatorial Neural Networks

Abstract Simulating the evolution of a coagulating aerosol or cloud of droplets in a key problem in atmospheric science. We present a proof of concept for modeling coagulation processes using a novel combinatorial neural network (CombNN) architecture. Using two types of data from a high‐detail particle‐resolved aerosol simulation, we show that CombNN models outperform standard neural networks and are competitive in accuracy with traditional state‐of‐the‐art sectional models. These CombNN models could have application in learning coarse‐grained coagulation models for multi‐species aerosols and for learning coagulation models from observed size‐distribution data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Agrivoltaics in Color: Going From Light Spectra to Biomass

Abstract Agrivoltaics (AV), conceived in the early 1980s, promise to ameliorate competition between solar energy generation and crop production for arable land. The premise behind AV is that excess light not used in photosynthesis can be used for energy production. There are opportunities for maximizing photosynthesis by targeting particular wavelengths (e.g., red) to be transmitted through semi‐transparent photovoltaic (PV) cells depending on crop type and environmental conditions. Camporese and Abou Najm (2022, https://doi.org/10.1029/2022EF002900 ) developed a numerical model that accommodates the various wavelengths of the incoming light spectrum to predict photosynthesis, stomatal conductance, and transpiration. This commentary seeks to place those and other recent findings about the modifications to the plant micro‐environment by PV cells in the context of maximum attainable aboveground biomass.

Katul, Gabriel G.↗
DOE OSTI2023

Compounding Uncertainties in Economic and Population Growth Increase Tail Risks for Relevant Outcomes Across Sectors

Understanding the long-term effects of population and GDP changes requires a multisectoral and regional understanding of the coupled human-Earth system, as the long-term evolution of this coupled system is influenced by human decisions and the Earth system. This study investigates the impact of compounding economic and population growth uncertainties on long-term multisectoral outcomes. We use the Global Change Analysis Model (GCAM) to explore the influence of compounding and feedback between future GDP and population growth on four key sectors: final energy consumption, water withdrawal, staple food prices, and CO 2 emissions. The results show that uncertainties in GDP and population compound, resulting in a magnification of tail risks for outcomes across sectors and regions. Compounding uncertainties significantly impact metrics such as CO 2 emissions and final energy consumption, particularly at the upper tail at both global and regional levels. However, the impact of staple food prices and water withdrawal depends on regional factors. Additionally, an alternative low-carbon transition scenario could compound uncertainties and increase tail risk, particularly in staple food prices, highlighting the influence of emergent constraints on land availability and food-energy competition for land use. The findings underscore the importance of considering and adequately accounting for compounding uncertainties in key drivers of multisectoral systems to enhance our comprehensive understanding of the complex nature of multisectoral systems. The paper provides valuable insights into the potential implications of compounding uncertainties.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Subsurface Redox Interactions Regulate Ebullitive Methane Flux in Heterogeneous Mississippi River Deltaic Wetland

Abstract As interfaces connecting terrestrial and ocean ecosystems, coastal wetlands develop temporally and spatially complex redox conditions, which drive uncertainties in greenhouse gas emission as well as the total carbon budget of the coastal ecosystem. To evaluate the role of complex redox reactions in methane emission from coastal wetlands, a coupled reactive‐transport model was configured to represent subsurface biogeochemical cycles of carbon, nitrogen, and sulfur, along with production and transport of multiple gas species through diffusion and ebullition. This model study was conducted at multiple sites along a salinity gradient in the Barataria Basin at the Mississippi River Deltaic Plain. Over a freshwater to saline gradient, simulated total flux of methane was primarily controlled by its subsurface production and consumption, which were determined by redox reactions directly (e.g., methanogenesis, methanotrophy) and indirectly (e.g., competition with sulfate reduction) under aerobic and/or anaerobic conditions. At fine spatiotemporal scales, surface methane fluxes were also strongly dependent on transport processes, with episodic ebullitive fluxes leading to higher spatial and temporal variability compared to the gradient‐driven diffusion flux. Ebullitive methane fluxes were determined by methane fraction in total ebullitive gas and the frequency of ebullitive events, both of which varied with subsurface methane concentrations and other gas species. Although ebullition thresholds are constrained by local physical factors, this study indicates that redox interactions not only determine gas composition in ebullitive fluxes but can also regulate ebullition frequency through gas production.

03 NATURAL GAS↗
DOE OSTI2024

Simulating Global Terrestrial Carbon and Nitrogen Biogeochemical Cycles With Implicit and Explicit Representations of Soil Microbial Activity

Abstract Nutrient limitation is widespread in terrestrial ecosystems. Accordingly, representations of nitrogen (N) limitation in land models typically dampen rates of terrestrial carbon (C) accrual, compared with C‐only simulations. These previous findings, however, rely on soil biogeochemical models that implicitly represent microbial activity and physiology. Here we present results from a biogeochemical model testbed that allows us to investigate how an explicit versus implicit representation of soil microbial activity, as represented in the MIcrobial‐MIneral Carbon Stabilization (MIMICS) and Carnegie‐Ames‐Stanford Approach (CASA) soil biogeochemical models, respectively, influence plant productivity, and terrestrial C and N fluxes at initialization and over the historical period. When forced with common boundary conditions, larger soil C pools simulated by the MIMICS model reflect longer inferred soil organic matter (SOM) turnover times than those simulated by CASA. At steady state, terrestrial ecosystems experience greater N limitation when using the MIMICS‐CN model, which also increases the inferred SOM turnover time. Over the historical period, however, warming‐induced acceleration of SOM decomposition over high latitude ecosystems increases rates of N mineralization in MIMICS‐CN. This reduces N limitation and results in faster rates of vegetation C accrual. Moreover, as SOM stoichiometry is an emergent property of MIMICS‐CN, we highlight opportunities to deepen understanding of sources of persistent SOM and explore its potential sensitivity to environmental change. Our findings underscore the need to improve understanding and representation of plant and microbial resource allocation and competition in land models that represent coupled biogeochemical cycles under global change scenarios.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Two-Way Option Contracts That Facilitate Adaptive Water Reallocation in the Western United States

Many water markets in the western United States (U.S.) have the ability to reallocate water temporarily during drought, often as short-term water rights leases from lower value irrigated activities to higher value urban uses. Regulatory approval of water transfers, however, typically takes time and involves high transaction costs that arise from technical and legal analyses, discouraging short-term leasing. This leads municipalities to protect against drought-related shortfalls by purchasing large volumes of infrequently used permanent water rights. High transaction costs also result in municipal water rights rarely being leased back to irrigators in wet or normal years, reducing agricultural productivity. This research explores the development of a multi-year two-way option (TWO) contract that facilitates leasing from agricultural-to-urban users during drought and leasing from urban-to agricultural users during wet periods. The modeling framework developed to assess performance of the TWO contracts includes consideration of the hydrologic, engineered, and institutional systems governing the South Platte River Basin in Colorado where there is growing competition for water between municipalities (e.g., the city of Boulder) and irrigators. The modeling framework is built around StateMod, a network-based water allocation model used by state regulators to evaluate water rights allocations and potential rights transfers. Results suggest that the TWO contracts could allow municipalities to maintain supply reliability with significantly reduced rights holdings at lower cost, while increasing agricultural productivity in wet and normal years. Additionally, the TWO contracts provide irrigators with additional revenues via net payments of option fees from municipalities.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Large Divergence of Projected High Latitude Vegetation Composition and Productivity Due To Functional Trait Uncertainty

Abstract Vegetation distribution and composition are expected to change in northern high latitudes under rapid warming, which regulates ecosystem functions but remains challenging to predict. Vegetation change arises from the interplay of chronic climate trends such as warming and transient demographic processes of recruitment, growth, competition, and mortality. Most predictive models overlooked the role of demographic dynamics controlled by plant traits. Here, we simulate vegetation dynamics at the Kougarok Hillslope site in Alaska under historical and future climates using the E3SM Land Model coupled to the Functionally Assembled Terrestrial Simulator (ELM‐FATES). To evaluate the roles of plant traits, we parameterize the model with 5,265 trait configurations representing diverse physiological and demographic strategies. Results show current modeled biomass, composition, and productivity are most sensitive to traits controlling photosynthetic capacity, carbon allocation, allometry, and phenology. Among all trait configurations, ∼5% reproduce in situ biomass and plant functional type (PFT) composition measured in 2016, that are indistinguishable from these two observed ecosystem states. Notably, these same trait configurations produce diverging biomass, composition, and productivity under future climate, where the uncertainty attributable to traits is twice the change attributable to climate change. The variation of projected productivity arises from emerging PFT composition under novel climate regimes, primarily explained by traits controlling cold‐induced mortality, recruitment, and allometry. Our findings highlight the importance and uncertainty of demographic dynamics and its interaction with climate change in shaping Arctic vegetation change. Improved model predictions will likely benefit from explicit consideration of vegetation demography and better constraints of critical traits.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

A New Ecosystem Model for Arctic Phytoplankton Phenology From Ice‐Covered to Open‐Water Periods: Implications for Future Sea Ice Retreat Scenarios

Abstract To predict ecosystem change in the Arctic Ocean, understanding Arctic phytoplankton phenology is essential. We develop a marine ecosystem model focusing on phytoplankton dynamics and the competition for multiple resources based on knowledge from in situ data obtained in the Chukchi Sea. The model is designed to include the unique ecological characteristics of the Arctic Ocean. To the best of our knowledge, this is the first model successfully simulating the current paradigm for Arctic phytoplankton phenology, including both under‐ice blooms and succession of phytoplankton groups. Sensitivity experiments show that the dominance of diatoms can be sustained by lateral transport of a high‐silicate water mass (Pacific Winter Water). Experiments for future scenarios show that sea ice retreat drives a decrease in large diatoms and an increase in the other relatively small non‐diatom groups. These results suggest that dominant phytoplankton groups can be shifted from diatoms to non‐diatoms under ongoing Arctic Ocean sea ice retreat.

Choi, Jang‐Geun↗
DOE OSTI2025

Deep Learning for Subsurface Flow: A Comparative Study of U‐Net, Fourier Neural Operators, and Transformers in Underground Hydrogen Storage

Subsurface flow research is essential for the sustainable management of natural resources and the environment. Deep learning (DL) has significantly advanced this field by developing efficient and accurate surrogate models to replace computationally expensive physics‐based simulations. These surrogate models are commonly used to predict the spatiotemporal evolution of state variables, such as gas saturation and reservoir pressure, in heterogeneous geological formations. Despite the various DL models applied to this task, there is a lack of studies systematically comparing their performance. This absence of comparative analysis leads to somewhat arbitrary DL model selection in subsurface flow research, resulting in suboptimal performance and potentially inaccurate predictions. To bridge this gap, we conduct a systematic comparison study of three popular DL architectures—U‐Net, Fourier Neural Operators (FNO), and Segmentation Transformer (SETR)—in surrogate modeling of underground hydrogen storage (UHS). We focus on UHS due to its promise of enhancing clean energy resilience and its cyclic operational conditions that represent common scenarios in various subsurface applications. We evaluate the models based on accuracy, training cost, and inference speed. The comparison shows that U‐Net achieves the highest accuracy, followed by SETR and FNO. Despite its lower accuracy, FNO has the highest inference speed. SETR offers competitive accuracy with the least training memory usage, demonstrating the potential of transformers in learning subsurface flow. Our results provide guidance for selecting DL models for surrogate modeling in a wide range of subsurface flow problems.

42 ENGINEERING↗