Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep”

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

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

At least 289 records · Page 16

Examining Aerosol Vertical Transport and Removal during Deep Convective Events

Atmospheric aerosols affect the global energy budget by scattering and absorbing sunlight (direct effects) and by changing the microphysical structure, lifetime, and coverage of clouds (indirect effects). Globally, the free troposphere is a major source of nucleation- and Aitken-mode aerosols due to the enhanced new particle formation rates at high altitudes. Recent studies have shown deep convective systems are capable of transporting these small aerosols from the free troposphere to the boundary layer by strong convective downdrafts and weaker downward motions in the stratiform regions. These vertically transported aerosols can grow into cloud condensation nuclei (CCN) and play a significant role in the global climate. During the deep convective processes, existing accumulation-mode aerosols that act as coagulation sinks of smaller particles are also removed by wet scavenging. Compared to the vertical transport of these particles by entrainment mixing, which is slower but more prevalent, the deep convective downdraft processes may be more rapid and efficient in the vertical transport of aerosols. However, most of the current climate models do not include this mechanism as a source of CCN, mainly because the frequency of deep convective events varies significantly with geographic location and thus their contributions to CCN are unpredictable. We target this critical gap in understanding the vertical transport and removal of aerosols by deep convections. We proposed to analyze a multi-year, multi-site measurement record available from the U.S. Department of Energy (DOE) ARM program, including the observations from the 2014/15 Observations and Modeling of the Green Ocean Amazon (GoAmazon) field campaign, the 2017/18 Aerosol and Cloud Experiments in the Eastern North Atlantic (ACE-ENA) field campaign, the 2018/19 Cloud, Aerosol, and Complex Terrain Interactions (CACTI) field campaign, the 2021/22 Tracking Aerosol Convection Interactions Experiment (TRACER) field campaign, and the long-term measurements collected at the Southern Great Plains (SGP) atmospheric observatory, where deep convective clouds were frequently observed. This project is aimed at the following three objectives: (1) Gaining a detailed and quantitative understanding of the aerosols transported by a convective downdraft and their evolution in the atmosphere; (2) Examining the wet scavenging mechanisms and efficiencies of aerosols at altitudes of deep convective systems based on ground and aircraft measurements; (3) Evaluating the contribution of deep convective systems to CCN as both a source and a sink of atmospheric aerosols and its seasonal variabilities.

54 ENVIRONMENTAL SCIENCES↗

Life Support for Deep Space and Mars

How should life support for deep space be developed? The International Space Station (ISS) life support system is the operational result of many decades of research and development. Long duration deep space missions such as Mars have been expected to use matured and upgraded versions of ISS life support. Deep space life support must use the knowledge base incorporated in ISS but it must also meet much more difficult requirements. The primary new requirement is that life support in deep space must be considerably more reliable than on ISS or anywhere in the Earth-Moon system, where emergency resupply and a quick return are possible. Due to the great distance from Earth and the long duration of deep space missions, if life support systems fail, the traditional approaches for emergency supply of oxygen and water, emergency supply of parts, and crew return to Earth or escape to a safe haven are likely infeasible. The Orbital Replacement Unit (ORU) maintenance approach used by ISS is unsuitable for deep space with ORU's as large and complex as those originally provided in ISS designs because it minimizes opportunities for commonality of spares, requires replacement of many functional parts with each failure, and results in substantial launch mass and volume penalties. It has become impractical even for ISS after the shuttle era, resulting in the need for ad hoc repair activity at lower assembly levels with consequent crew time penalties and extended repair timelines. Less complex, more robust technical approaches may be needed to meet the difficult deep space requirements for reliability, maintainability, and reparability. Developing an entirely new life support system would neglect what has been achieved. The suggested approach is use the ISS life support technologies as a platform to build on and to continue to improve ISS subsystems while also developing new subsystems where needed to meet deep space requirements.

life support↗

Global Survey of Selected Deep Underground Facilities; Examples of Geotechnical and Engineering Capabilities, Achievements, Challenges(Mines, Shafts, Tunnels, Boreholes, Sites and Underground Facilities for Nuclear Waste and Physics R&D):A Guide

This report presents tables, references, notes, and a synthesis of some notable geotechnical and engineering information that were applied in the creation of five interactive layer maps [http://gis.inl.gov/globalsites/ will redirect to 2018 update, http://gis.inl.gov/GlobalSurvey/] for selected: 1) deep mines and shafts; 2) existing, considered or planned radioactive waste management deep underground studies or disposal facilities 3) deep large diameter boreholes, 4) physics underground laboratories and facilities, and 5) deep open pit mines from around the world. These data are intended to facilitate user access to basic information and references regarding “deep underground” facilities, history, activities, and plans. In general, the interactive maps and database provide each facility’s approximate site location, geology, and engineered features (e.g.: access, geometry, depth, diameter, year of operations, groundwater, lithology, host unit name and age, basin; operator, management organization, geographic data, nearby cultural features, other). Although the survey is not all encompassing, it is a comprehensive review of many of the significant existing and historical underground facilities discussed in the literature addressing radioactive waste management and deep mined geologic disposal safety systems. The interactive map suite allows for a better appreciation of site area geographic and cultural settings. The global survey is intended to be used as a communication tool to support and inform: 1) interested parties and decision makers; 2) radioactive waste disposal and siting option evaluations, and 3) safety case development applicable to any mined geologic disposal facility as a demonstration (examples) of historical and current engineering and geotechnical capabilities available for use in deep underground facility siting, planning, construction, operations and monitoring.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Resolved gravity waves in the tropical stratosphere: Impact of horizontal resolution and deep convection parametrization

Convectively generated gravity waves (CGGWs) are important for numerical weather prediction due to their effect on the quasi-biennial oscillation (QBO) in the stratosphere. Here, using global ECMWF IFS simulations at TCo7999 (or 1.25 km), TCo2559 (or 3.9 km) and TCo1279 (or 7.8 km) horizontal resolutions, sensitivity of resolved CGGWs to the horizontal resolution and to the explicit versus parametrized representation of deep convection is elucidated during the westerly shear phase of the QBO. Parametrized deep convection is found to inhibit CGGWs, resulting in a twofold reduction in CGGW forcing. When deep convection is explicitly resolved, the total CGGW forcing is almost unchanged across the horizontal resolutions. However, the contribution of long and mesoscale CGGWs (with horizontal wavelengths 100 km ≤ $λ_h$ < $1,900$ km) to the total CGGW forcing decreases and the contribution of smaller-scale CGGWs (with $λ_h$ < $100$ km) increases as the horizontal resolution increases. At the maximum CGGW forcing altitude, at TCo7999 resolution 43% of the total CGGW forcing is due to long and mesoscale waves, whereas at TCo2559 and TCo1279 resolutions their contribution is 70% and 90%, respectively. While CGGW forcing from long and mesoscale waves is similar at TCo7999 resolution with explicit deep convection and at TCo1279 resolution with parametrized deep convection, CGGW forcing from these waves is artificially enhanced at TCo1279 and TCo2559 resolutions with explicit deep convection. This is due to the explicit deep convection being too strong and having too much variance for 100 km ≤ $λ_h$ < $1,900$ km. Therefore, parametrizations of deep convection and CGGWs (to account for forcing from waves with $λ_h$ < $100$ km) are required even at TCo2559 resolution. Additionally, resolved CGGW forcing at TCo7999 resolution is examined for the easterly shear phase of the QBO; similar to the westerly shear phase, the smaller-scale waves contribute 55% to the total CGGW forcing at the maximum CGGW forcing altitude.

54 ENVIRONMENTAL SCIENCES↗

Oil Biodegradation in Deep Marine Basins

Nine important hydrocarbon (oil) basins where offshore petroleum leases have been licensed are compared. These nine basins (Gulf of Mexico, Eastern Mediterranean’s Nile Deep-Sea Fan, Central Mediterranean and the Sirte Basin, North Sea, Caspian Sea, Angola, Trinidad and Tobago, Great Australian Bight, and Brazil’s Amazonian Deep-Sea Basin) are geographically separated and are impacted by very different water masses. The geochemical parameters of these basins are quite distinct, for example, salinities ranging from 39 psu in the Eastern Mediterranean to 12 psu in the Caspian. Additionally, parameters like temperatures of the bottom water are also very distinct, with the temperature in the deep water of the Eastern Mediterranean being between 12 °C and 14 °C and the temperature of the deep water in the North Sea being -2 °C. Each basin represents a unique ecosystem in which distinct microbes may thrive. These distinct environmental parameters may act to constrain the extent of hydrocarbon degradation in these basins. Another potential constraint on hydrocarbon degradation is the extent of natural hydrocarbon seeps in the area. Though many basins have similar if not 16S rRNA identical strains of oil-degrading bacteria, Colwellia psychrerythraea from different basins showed that a mixture of natural selection and neutral evolution has contributed to the divergence of these. Most if not all deep ocean basin microbial communities are dominated by Thaumarchaeota below 200 m. These microaerophilic, ammonium oxidizer, psychrophiles are very adapted to an oligotrophic lifestyle, and though many in this group will degrade oil, they are rapidly outcompeted by other bacteria in oil or high hydrocarbon intrusions, thus the virtual “canary in the coal mine.” Cometabolic biodegradation of oil is well documented but could be an important natural attenuation mechanism for oil in deep marine basins with episodic methane seeps. Microbial community structure can also predict concentrations of oil in deep basins. Many other synergistic effects require more research in environmental systems biology in deep marine basins.

Hazen, Terry C.↗

A record of deep-ocean dissolved O 2 from the oxidation state of iron in submarine basalts

The oxygenation of the deep ocean in the geological past has been associated with a rise in the partial pressure of atmospheric molecular oxygen (O 2 ) to near-present levels and the emergence of modern marine biogeochemical cycles. It has also been linked to the origination and diversification of early animals. It is generally thought that the deep ocean was largely anoxic from about 2,500 to 800 million years ago, with estimates of the occurrence of deep-ocean oxygenation and the linked increase in the partial pressure of atmospheric oxygen to levels sufficient for this oxygenation ranging from about 800 to 400 million years ago. Deep-ocean dissolved oxygen concentrations over this interval are typically estimated using geochemical signatures preserved in ancient continental shelf or slope sediments, which only indirectly reflect the geochemical state of the deep ocean. Here we present a record that more directly reflects deep-ocean oxygen concentrations, based on the ratio of Fe 3+ to total Fe in hydrothermally altered basalts formed in ocean basins. Our data allow for quantitative estimates of deep-ocean dissolved oxygen concentrations from 3.5 billion years ago to 14 million years ago and suggest that deep-ocean oxygenation occurred in the Phanerozoic (541 million years ago to the present) and potentially not until the late Palaeozoic (less than 420 million years ago).

58 GEOSCIENCES↗

Deep-field analytical calibration

The next generation of imaging surveys, including the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST), Euclid, and the Nancy Grace Roman Space Telescope, will provide unprecedented constraints on cosmology using weak gravitational lensing. To fully exploit this statistical power, shear measurement methods must achieve sub- per cent accuracy while mitigating systematic biases from noise, the point-spread function (PSF), blending, and shear-dependent detection. The analytical calibration framework (AnaCal) has demonstrated such accuracy but requires adding noise to images, reducing effective depth. We introduce Deep-Field Analytical Calibration (DEEP-FIELD AnaCal), an extension of AnaCal that uses deep-field images to compute shear responses while preserving the statistical power of wide-field data. We validate DEEP-FIELD AnaCal on isolated and blended galaxy image simulations with LSST-like conditions, finding it meets the stringent requirement of multiplicative bias $|m| < 3\times 10^{-3}$ at 99.7 per cent confidence. Compared to standard AnaCal applied to wide-field images, DEEP-FIELD AnaCal increases the effective galaxy number density from 17 to 30 arcmin$^{-2}$ for simulated 10-yr LSST data. With deep fields $10\times$ longer than the wide field, we find pixel noise variance in shear estimation is reduced by 30 per cent and overall uncertainty by $\sim 25~{{\ \rm per\ cent}}$. Finally, using the LSST Deep Drilling Fields strategy, we assess sample variance and find an equivalent calibration uncertainty of $\lesssim 0.3~{{\ \rm per\ cent}}$. These results demonstrate that DEEP-FIELD AnaCal offers a promising path to achieve the required shear calibration for upcoming weak lensing surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

Bioenergy sorghum’s deep roots: A key to sustainable biomass production on annual cropland

Abstract Bioenergy sorghum has high biomass yield potential, drought resilience, good nitrogen use efficiency, and a root system that contributes to the accumulation of soil organic carbon. In this study, field grown bioenergy sorghum root systems were analyzed during the growing season to characterize their depth, biomass, morphology, anatomy, and gene expression profiles. Bioenergy sorghum roots grew continuously during a 155‐day growing season producing ~175 nodal roots, accumulating ~7 Mg of dry biomass per hectare, and reaching >2 m deep in the soil profile. Nodal roots within 20 cm of the stem were 1–5 mm in diameter, whereas roots deeper in soil profiles were enriched in lateral roots with small diameters (~30–500 µm) enabling growth through soil macropores. In older field‐grown plants, roots with intact endodermal, vascular and inner root tissues were surrounded by degraded or aerenchyma‐filled epidermal and cortical cell layers. Transcriptome analysis of nodal, surface, and deep roots identified >2,500 differentially expressed genes involved in root growth, transport, adaptation, defense, and AMF–root interaction. Deep roots (180–240 cm) differentially expressed genes that regulate lateral root growth. Surface roots (0–20 cm) located mid‐row differentially expressed genes involved in nitrate transport, whereas ammonium transport genes were expressed in surface and deep roots and genes involved in phosphate transport were expressed in nodal, surface, and deep roots. Overall, bioenergy sorghum's long growing season enables root systems to grow deeper and accumulate more biomass than annual grain crops such as maize, attributes that could help restore annual cropland soil organic carbon levels and improve soil productivity. Deep roots active in nutrient transport are positioned to take‐up fertilizer leached deep into soil profiles mitigating potential nutrient run‐off. Bioenergy sorghum's large and deep root system is a key to sustainable production of biofuels, biopower, and bioproducts on annual cropland.

59 BASIC BIOLOGICAL SCIENCES↗

Deep Lynx Unidirectional Network Connector

The Deep Lynx UNC is a small utility written in Python that uses only basic Python libraries and the Deep Lynx Python Software Development Kit (SDK). It queries Deep Lynx for tasks that indicate some data to be run in the UNC's environment. Once a task is acquired, the UNC retrieves the associated data and provides it to some other application for processing, waits for output, and sends the output from the other application to Deep Lynx. As data flows only from the UNC to Deep Lynx, this software solves the problem of connecting Deep Lynx with applications and resources that exist in a network where the network can reach Deep Lynx, but Deep Lynx cannot reach that network (unidirectional network).

Browning, JerenM↗

DLIO: A DATA-CENTRIC BENCHMARK FOR DEEP LEARNING APPLICATIONS

SF-22-136 Deep learning has been shown as a successful method for various tasks, and its popularity results in numerous open-source deep learning software tools. Deep learning has been applied to a broad spectrum of scientific domains such as cosmology, particle physics, computer vision, fusion, and astrophysics. Scientists have performed a great deal of work to optimize the computational performance of deep learning frameworks. However, the same cannot be said for I/O performance. As deep learning algorithms rely on big-data volume and variety to effectively train neural networks accurately, I/O is a significant bottleneck on large-scale distributed deep learning training. DLIO, is a novel representative benchmark suite built based on the I/O profiling of the selected workloads. DLIO can be utilized to accurately emulate the I/O behavior of modern deep learning applications. Using DLIO, application developers and system software solution architects can identify potential I/O bottlenecks in their applications and guide optimizations to boost the I/O performance leading to lower training times. The storage vendor can also use DLIO as a guide for designing and optimize the storage and filesystem targeting at deep learning application.

ZHENG, HUIHUO↗

Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications

Abstract Robust quantification of predictive uncertainty is a critical addition needed for machine learning applied to weather and climate problems to improve the understanding of what is driving prediction sensitivity. Ensembles of machine learning models provide predictive uncertainty estimates in a conceptually simple way but require multiple models for training and prediction, increasing computational cost and latency. Parametric deep learning can estimate uncertainty with one model by predicting the parameters of a probability distribution but does not account for epistemic uncertainty. Evidential deep learning, a technique that extends parametric deep learning to higher-order distributions, can account for both aleatoric and epistemic uncertainties with one model. This study compares the uncertainty derived from evidential neural networks to that obtained from ensembles. Through applications of the classification of winter precipitation type and regression of surface-layer fluxes, we show evidential deep learning models attaining predictive accuracy rivaling standard methods while robustly quantifying both sources of uncertainty. We evaluate the uncertainty in terms of how well the predictions are calibrated and how well the uncertainty correlates with prediction error. Analyses of uncertainty in the context of the inputs reveal sensitivities to underlying meteorological processes, facilitating interpretation of the models. The conceptual simplicity, interpretability, and computational efficiency of evidential neural networks make them highly extensible, offering a promising approach for reliable and practical uncertainty quantification in Earth system science modeling. To encourage broader adoption of evidential deep learning, we have developed a new Python package, Machine Integration and Learning for Earth Systems (MILES) group Generalized Uncertainty for Earth System Science (GUESS) (MILES-GUESS) ( https://github.com/ai2es/miles-guess ), that enables users to train and evaluate both evidential and ensemble deep learning. Significance Statement This study demonstrates a new technique, evidential deep learning, for robust and computationally efficient uncertainty quantification in modeling the Earth system. The method integrates probabilistic principles into deep neural networks, enabling the estimation of both aleatoric uncertainty from noisy data and epistemic uncertainty from model limitations using a single model. Our analyses reveal how decomposing these uncertainties provides valuable insights into reliability, accuracy, and model shortcomings. We show that the approach can rival standard methods in classification and regression tasks within atmospheric science while offering practical advantages such as computational efficiency. With further advances, evidential networks have the potential to enhance risk assessment and decision-making across meteorology by improving uncertainty quantification, a longstanding challenge. This work establishes a strong foundation and motivation for the broader adoption of evidential learning, where properly quantifying uncertainties is critical yet lacking.

Schreck, John S.↗

Carbonate “clumped” isotope signatures in aragonitic scleractinian and calcitic gorgonian deep-sea corals

Deep-sea corals are a potentially valuable archive of the temperature and ocean chemistry of intermediate and deep waters. Living in near-constant temperature, salinity, and pH and having amongst the slowest calcification rates observed in carbonate-precipitating biological organisms, deep-sea corals can provide valuable constraints on processes driving mineral equilibrium and disequilibrium isotope signatures. Here we report new data to further develop “clumped” isotopes as a paleothermometer in deep-sea corals as well as to investigate mineral-specific, taxon-specific, and growth-rate-related effects. Carbonate clumped isotope thermometry is based on measurements of the abundance of the doubly substituted isotopologue 13 C 18 O 16 O 2 in carbonate minerals, analyzed in CO 2 gas liberated on phosphoric acid digestion of carbonates and reported as Δ 47 values. We analyzed Δ 47 in live-collected aragonitic scleractinian ( Enallopsammia sp.) and high-Mg calcitic gorgonian (Isididae and Coralliidae) deep-sea corals and compared results to published data for other aragonitic scleractinian taxa. Measured Δ 47 values were compared to in situ temperatures, and the relationship between Δ 47 and temperature was determined for each group to investigate taxon-specific effects. We find that aragonitic scleractinian deep-sea corals exhibit higher values than high-Mg calcitic gorgonian corals and the two groups of coral produce statistically different relationships between Δ 47 –temperature calibrations. These data are significant in the interpretation of all carbonate clumped isotope calibration data as they show that distinct Δ 47 –temperature calibrations can be observed in different materials recovered from the same environment and analyzed using the same instrumentation, phosphoric acid composition, digestion temperature and technique, CO 2 gas purification apparatus, and data handling. There are three possible explanations for the origin of these different calibrations. The offset between the corals of different mineralogy is in the same direction as published theoretical predictions for the offset between calcite and aragonite although the magnitude of the offset is different. One possibility is that the deep-sea coral results reflect high-Mg and aragonite crystals attaining nominal mineral equilibrium clumped isotope signatures due to conditions of extremely slow growth. In that case, a possible explanation for the attainment of disequilibrium bulk isotope signatures and equilibrium clumped isotope signatures by deep-sea corals is that extraordinarily slow growth rates can promote the occurrence of isotopic reordering in the interfacial region of growing crystals. We also cannot rule out a component of a biological “vital effect” influencing clumped isotope signatures in one or both orders of coral. Based on published experimental data and theoretical calculations, these biological vital effects could arise from kinetic isotope effects due to the source of carbon used for calcification, temperature- and pH-dependent rates of CO 2 hydration and/or hydroxylation, calcifying fluid pH, the activity of carbonic anhydrase, the residence time of dissolved inorganic carbon in the calcifying fluid, and calcification rate. A third possible explanation is the occurrence of variable acid digestion fractionation factors. Although a recent study has suggested that dolomite, calcite, and aragonite may have similar clumped isotope acid digestion fractionation factors, the influence of acid digestion kinetics on Δ 47 is a subject that warrants further investigation.

58 GEOSCIENCES↗

Moss Tolerance of Deep Space-Like Ionizing Radiation, Singular or Combined With Spaceflight Microgravity-Preparation for the BRIC-27 Experiment to the ISS

Understanding how terrestrial life perceives and tolerates deep space environments is essential to advancing human space exploration missions and searching for life beyond our solar system. Throughout evolution moss adapted to living in extreme environments from the edge of habitability. Moss inhabits areas with high UVB/UVC photon levels such as polar regions subject to ozone holes, high elevation mountain environments, and sites with elevated ionizing particle levels such as those found at nuclear plant accident sites, atomic bomb test sites or radioactive element-rich regions. We investigate if moss tolerance of ionizing radiation encounter terrestrially extends to resistance to deep space ionizing radiation. We exposed moss to high energy ion beams simulating Galactic Cosmic Ray (GCRSim) and Solar Particle Event (SPESim) (NSRL, BNL) that permeate deep space, and moss was unharmed. We also exposed moss to high doses of gamma rays as released in astrophysical events, and moss survived absorption of extremely high doses. Next, we ask if moss tolerance of deep space-like ionizing radiation observed terrestrially differs from that in other deep space environments. In the upcoming BRIC-27 spaceflight experiment we will sequentially expose Antarctic moss C.purpureus to GCRSim and SPESim followed by exposure to spaceflight microgravity and compare gene expression profiles in deep space-like ionizing radiation, singular or combined with spaceflight microgravity. Additionally, we will compare gene expression profiles in spaceflight microgravity, singular or combined with deep space-like ionizing radiation. Hence, BRIC-27 will advance our understanding of the combined effects of both deep space ionizing radiation and microgravity, which may have different and more profound effects on plant physiology and performance than each condition separately.

Agata Klaudia Zupanska↗

Scalable Deep Learning via I/O Analysis and Optimization

Scalable deep neural network training has been gaining prominence because of the increasing importance of deep learning in a multitude of scientific and commercial domains. Consequently, a number of researchers have investigated techniques to optimize deep learning systems. Much of the prior work has focused on runtime and algorithmic enhancements to optimize the computation and communication. Despite these enhancements, however, deep learning systems still suffer from scalability limitations, particularly with respect to data I/O. This situation is especially true for training models where the computation can be effectively parallelized, leaving I/O as the major bottleneck. In fact, our analysis shows that I/O can take up to 90% of the total training time. Thus, in this article, we first analyze LMDB, the most widely used I/O subsystem of deep learning frameworks, to understand the causes of this I/O inefficiency. Based on our analysis, we propose LMDBIO—an optimized I/O plugin for scalable deep learning. LMDBIO includes six novel optimizations that together address the various shortcomings in existing I/O for deep learning. Lastly, our experimental results show that LMDBIO significantly outperforms LMDB in all cases and improves overall application performance by up to 65-fold on a 9,216-core system.

97 MATHEMATICS AND COMPUTING↗

SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees

Deep unfolding networks have recently gained popularity for solving imaging inverse problems. However, the computational and memory complexity of data-consistency layers within traditional deep unfolding networks scales with the number of measurements, limiting their applicability to large-scale imaging inverse problems. We propose SGD-Net as a new methodology for improving the efficiency of deep unfolding through stochastic approximations of the data-consistency layers. Our theoretical analysis shows that SGD-Net can be trained to approximate batch deep unfolding networks to an arbitrary precision. Our simulations on intensity diffraction tomography and sparse-view computed tomography show that SGD-Net can match the performance of the traditional batch network at a fraction of training and testing complexity.Deep unfolding networks have recently gained popularity for solving imaging inverse problems. However, the computational and memory complexity of data-consistency layers within traditional deep unfolding networks scales with the number of measurements, limiting their applicability to large-scale imaging inverse problems. We propose SGD-Net as a new methodology for improving the efficiency of deep unfolding through stochastic approximations of the data-consistency layers. Our theoretical analysis shows that SGD-Net can be trained to approximate batch deep unfolding networks to an arbitrary precision. Our simulations on intensity diffraction tomography and sparse-view computed tomography show that SGD-Net can match the performance of the traditional batch network at a fraction of training and testing complexity.

97 MATHEMATICS AND COMPUTING↗

Data and scripts associated with the manuscript "Encoding Diel Hysteresis and the Birch Effect in Dryland Soil Respiration Models through Knowledge-Guided Deep Learning"

This package contains the data and scripts used in "Encoding Diel Hysteresis and the Birch Effect in Dryland Soil Respiration Models through Knowledge-Guided Deep Learning" (Jiang et al., 2022). The data.zip file contains the flux tower and automated chamber observations used for developing the deep learning model for modeling soil respiration. The scripts.zip file contains the Jupyter notebooks and python scripts for preprocessing the data, training the deep learning models, and postprocessing the results. The src.zip contains the source code for training the deep learning model, performing mutual information analysis, and plotting functions. The trained_models.zip contains multiple folders used for hosting the trained deep-learning models and the associated soil respiration predictions. The whole process is performed using python. We include the REAMD.md to document the python package requirements.Soil respiration in dryland ecosystems is challenging to model due to its complex interactions with environmental drivers. Knowledge-guided deep learning provides a much more effective means of accurately representing these complex interactions than traditional Q10-based models. Mutual information analysis revealed that future soil temperature shares more information with soil respiration than past soil temperature, consistent with their clockwise diel hysteresis. We explicitly encoded diel hysteresis, soil drying, and soil rewetting effects on soil respiration dynamics in a newly designed Long Short Term Memory (LSTM) model. The model takes both past and future environmental drivers as inputs to predict soil respiration. The new LSTM model substantially outperformed three Q10-based models and the Community Land Model when reproducing the observed soil respiration dynamics in a semi-arid ecosystem. The new LSTM model clearly demonstrated its superiority for temporally extrapolating soil respiration dynamics, such that the resulting correlation with observational data is up to 0.7 while the correlations of both Q10-based models and the Community Land Model (CLM) are less than 0.4. Our results underscore the high potential for knowledge-guided deep learning to replace Q10-based soil respiration modules in Earth system models.

54 ENVIRONMENTAL SCIENCES↗

Final Scientific/Technical Report: Real Time Particle Size and Settling Velocity for In-Situ Monitoring of Deep-Sea Polymetallic Nodule Mining

In the coming decade, it is anticipated that deep-sea mining activities will commence throughout the deep ocean, at depths around 4000-6000m where vast deposits of baseball-sized rocks called polymetallic nodules lay on the seabed. These nodules contain resources such as cobalt, copper and nickel which are needed to produce batteries for electric vehicles. Collector machines driving on the seabed and gathering these nodules will stir up clouds of sediment, and there is concern about how the resulting concentration of sediment in the near-bottom ocean water will affect deep and mid-ocean biology. There is currently no technology capable of comprehensively monitoring the concentrations and properties of stirred up sediment in the deep. This project addressed this critical gap by developing a deep-sea particle measurement system (Real-Time Size and Settling Velocity, RTSSV) that uses a multi-camera high-resolution video imaging system, with accompanying image processing, to measure the concentration, size distribution and settling speed of sediment. Importantly, these measurements are made in-situ, so that delicate sediment aggregates will not be disturbed, as is the case when collecting samples at depth to be measured in the lab. The instrument can be mounted on a variety of established deep-sea platforms such as moorings, AUVs and ROVs, providing data in near real time, where the key measurements need to be made—from the source of the sediment disturbance and throughout the resulting plume—vital for determining if deep-sea mining can be done in an environmentally responsible manner. The developed system (RTSSV) will possess the only such capability worldwide, giving a US-based group a substantial competitive advantage as this newly evolving global industry develops. As added value, the technology will also have widespread application to the offshore oil and gas industry, dredging industry, and to sediment science research.

47 OTHER INSTRUMENTATION↗

Measure Guideline: Deep Energy Enclosure Retrofit for Interior Insulation of Masonry Walls and Flat Roofs

Measure Guideline: Deep Energy Enclosure Retrofit for Zero Energy Ready House Flat Roofs BSC TO5 Task 7.4 Deep Energy Enclosure Retrofit (DEER) for Interior Insulation of Masonry Walls Lawrence, MA 01840 This Measure Guideline describes a deep energy enclosure retrofit solution for insulating mass masonry buildings from the interior. It describes the retrofit assembly, technical details, and installation sequence for retrofitting masonry walls. Interior insulation of masonry retrofits might adversely affect the durability of the wall. This guideline includes a review of decision criteria pertinent to retrofitting masonry walls from the interior and the possible risk of freeze-thaw damage. This Measure Guideline is intended to support contractors who are implementing an interior insulation-based high-performance enclosure retrofit for masonry buildings and designers who are looking to design such retrofits. It may also be helpful to building owners who wish to learn more about strategies that are available for deep energy enclosure retrofit of masonry residential buildings on the interior. This Measure Guideline is important to the high-performance retrofit industry because the solution described here provides an opportunity to retrofit masonry buildings with ambitious energy performance goals by insulating them from the interior without disturbing their exterior appearance. BSC TO5 Task 7.4 Deep Energy Enclosure Retrofit (DEER) for Zero Energy Ready House (ZERH) Flat Roofs House Jamaica Plain, MA 02130 This Measure Guideline provides design and construction information for a deep energy enclosure retrofit solution of a flat roof assembly. It describes the strategies and procedures for an exterior retrofit of a flat wood-framed roof with brick masonry exterior walls using exterior and interior (framing cavity) insulation. The approach supported in this guide could also be adapted for use with flat wood-framed roofs with wood-framed exterior walls. Designers, contractors, and building code officials will benefit from the information contained in this Measure Guideline. The guide may also be helpful to building owners who wish to learn more about strategies for deep energy enclosure retrofit of flat roofs. This Measure Guideline is important to the high-performance retrofit industry because it demonstrates techniques for retrofitting flat roofs from the exterior, which is less disruptive to the living space and allows the structure to remain occupied during the project. It also illustrates a solution for preparing homes to become zero energy ready.

Array↗