Search NASA⌕ Search

SEARCH · Search NASA

Results for “parameter reduction”

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 1,747 records · Page 97

Observation and simulation of beam-beam induced emittance growth in RHIC

In the recent years the peak luminosity of the RHIC polarized proton run has been improved. However, as a consequence, the luminosity lifetime is reduced. The beam emittance growth during the beam storage is a main contributor to the luminosity lifetime reduction, and it seems to be caused mainly by the beam-beam effect during collision. A simulation study of the emittance growth is performed with RHIC machine parameters using the LIFETRAC code [1]. The initial results of this study were reported in an earlier paper [2]. We present, in this paper, an in depth investigation of the emittance growth for a range of beam-beam parameters and bunch lengths. The simulation results are also compared to the available data from experimental measurements.

43 PARTICLE ACCELERATORS↗

Dynamic modeling and adaptive control for space stations

Of all large space structural systems, space stations present a unique challenge and requirement to advanced control technology. Their operations require control system stability over an extremely broad range of parameter changes and high level of disturbances. During shuttle docking the system mass may suddenly increase by more than 100% and during station assembly the mass may vary even more drastically. These coupled with the inherent dynamic model uncertainties associated with large space structural systems require highly sophisticated control systems that can grow as the stations evolve and cope with the uncertainties and time-varying elements to maintain the stability and pointing of the space stations. The aspects of space station operational properties are first examined, including configurations, dynamic models, shuttle docking contact dynamics, solar panel interaction, and load reduction to yield a set of system models and conditions. A model reference adaptive control algorithm along with the inner-loop plant augmentation design for controlling the space stations under severe operational conditions of shuttle docking, excessive model parameter errors, and model truncation are then investigated. The instability problem caused by the zero-frequency rigid body modes and a proposed solution using plant augmentation are addressed. Two sets of sufficient conditions which guarantee the globablly asymptotic stability for the space station systems are obtained.

Ih, C. H. C.↗

Local ammonia storage and ammonia inhibition in a monolithic copper-beta zeolite SCR catalyst

Selective catalytic reduction of NO with NH{sub 3} was studied on a Cu-beta zeolite catalyst, with specific focus on the distributed NH{sub 3} capacity utilization and inhibition. In addition, several other relevant catalyst parameter distributions were quantified including the SCR zone, or catalyst region where SCR occurs, and NO and NH{sub 3} oxidation. We show that the full NH{sub 3} capacity (100% coverage) is used within the SCR zone for a range of temperatures. By corollary, unused NH{sub 3} capacity exists downstream of the SCR zone. Consequently, the unused capacity relative to the total capacity is indicative of the portion of the catalyst unused for SCR. Dynamic NH{sub 3} inhibition distributions, which create local transient conversion inflections, are measured. Dynamic inhibition is observed where the gas phase NH{sub 3} and NO concentrations are high, driving rapid NH{sub 3} coverage buildup and SCR. Accordingly, we observe dynamic inhibition at low temperatures and in hydrothermally aged states, but predict its existence very near the catalyst front in higher conversion conditions where we did not specifically monitor its impact. While this paper addresses some general distributed SCR performance parameters including Oxidation and SCR zone, our major new contributions are associated with the NH{sub 3} capacity saturation within the SCR zone and dynamic inhibition distributions and the associated observations. These new insights are relevant to developing accurate models, designs and control strategies for automotive SCR catalyst applications.

54 ENVIRONMENTAL SCIENCES↗

Observations of the Hot Horizontal Branch Stars in the Metal-Rich Bulge Globular Cluster NGC 6388

The metal-rich bulge globular cluster NGC 6388 shows a distinct blue horizontal-branch tail in its colour-magnitude diagram (Rich et al. 1997) and is thus a strong case of the well-known 2nd Parameter Problem. In addition, its horizontal branch (HB) shows an upward tilt toward bluer colours, which cannot be explained by canonical evolutionary models. Several non-canonical scenarios have been proposed to explain these puzzling observations. In order to test the predictions of these scenarios, we have obtained medium resolution spectra to determine the atmospheric parameters of a sample of the blue HB stars in NGC 6388.Using the medium resolution spectra, we determine effective temperatures, surface gravities and helium abundances by fitting the observed Balmer and helium lines with appropriate theoretical stellar spectra. As we know the distance to the cluster, we can verify our results by determining masses for the stars. During the data reduction we took special care to correctly subtract the background, which is dominated by the overlapping spectra of cool stars. The cool blue tail stars in our sample with T(sub eff) approximately 10000 K have lower than canonical surface gravities, suggesting that these stars are, on average, approximately equal to 0.4 mag brighter than canonical HB stars in agreement with the observed upward slope of the HB in NGC 6388. Moreover, the mean mass of these stars agrees well with theoretical predictions. In contrast, the hot blue tail stars in our sample with T(sub eff) greater than or equal to 12000 K show significantly lower surface gravities than predicted by any scenario, which can reproduce the photometric observations. Their masses are also too low by about a factor of 2 compared to theoretical predictions. The physical parameters of the blue HB stars at about 10,000 K support the helium pollution scenario. The low gravities and masses of the hot blue tail stars, however, are probably caused by problems with the data reduction, most likely due to remaining background light in the spectra, which would affect the fainter hot blue tail stars much more strongly than the brighter cool blue tail stars. Our study of the hot blue tail stars in NGC 6388 illustrates the obstacles which are encountered when attempting to determine the atmospheric parameters of hot HB stars in very crowded fields using ground-based observations. We discuss these obstacles and offer possible solutions for future projects.

Moehler, S.↗

Scaling Analysis of Vortex Shedding Frequency between In line Plates

The Global Threat Reduction initiative (GTRI) was established by the National Nuclear Security Administration (NNSA) in 2004. One of the highlights of the program is to convert research and test reactors at home and abroad from Highly Enriched Uranium (HEU) to Low Enriched Uranium (LEU) fuels. By switching directly from HEU to LEU, an LEU reactor produces fewer neutrons than an HEU reactor and in some cases; a direct switch from HEU to LEU results in the reactor being unable to achieve criticality. Under certain circumstances, criticality is achieved and the original neutron fluxes are restored with LEU fuel by adding more fuel elements to the core. Not all reactors, exemplified by the Advanced Test Reactor (ATR) at the Idaho National Lab (INL), have the luxury of simply adding more fuel to the core. In such cases, entirely new designs must be created and tested. The ATR utilizes plate type fuel. During the design and testing of the ATR fuel, vortex shedding was identified as potential source for experiment and fuel failure. For fuel conversion, elements must be tested in pile prior to the qualification of the new fuel design. For some test elements, the primary test plate is followed by a secondary handling plate. It is known that objects downstream influence the vortex shedding. While vortex shedding is an important and well-studied phenomenon, not only in the context of nuclear engineering, but in fields as diverse as aeronautical engineering and biology, there is a significant gap in the understanding of vortex shedding in regards to two bluff bodies in line direct line with one another. The study provides a scaling analysis of the frequency of vortex shedding for plates in line. The scaling analysis is then compared to numerical simulation of two in line plates. The study provided a scaling analysis and suggested parameter for determining the expected trends in the Strouhal number for a plate with some object downstream. The analysis made the consideration of differing length scales as the cause for a reduction in vortex shedding frequency as well as looked at the potential impact of boundary layer growth on vortex shedding for a flat plate. The comparison between the expected scaling and numerical data was less than ideal, but there are several potential causes. One of the biggest issues with the comparison is the uncertainty associated with the defect velocity. The frequency was well established in several cases, but the velocity was not always well established resulting in a larger uncertainty. In the cases where the velocity was near zero (i.e. b/d = 0.5), the uncertainty dominated since the corrected Strouhal number is divided by the defect velocity. The Reynolds numbers were not incredibly large, ranging from 350-1600. The low Reynolds numbers potentially undermine the principle assumption made in the scaling analysis, i.e. Reynolds number is negligibly small. Additionally, the exact definition of the 'reduced velocity' was never explicitly defined, merely inferred from the assumption that the x length scale and y length scale were in fact different. Another potential problem is the defect velocities in the lower Reynolds number cases were not much smaller than the free stream velocity. Further work needs to be done at higher Reynolds numbers and with more data points to ensure the assumptions made are valid as well as confirm the accuracy of the scaled relationship. (authors)

42 ENGINEERING↗

Life cycle assessment and residue leaching: The importance of parameter, scenario and leaching data selection

Highlights: • Relevance of metal leaching in waste management system LCAs was assessed. • Toxic impacts from leaching could not be disregarded. • Uncertainty of toxicity, due to background activities, determines LCA outcomes. • Parameters such as pH and L/S affect LCA results. • Data modelling consistency and coverage within an LCA are crucial. - Abstract: Residues from industrial processes and waste management systems (WMSs) have been increasingly reutilised, leading to landfilling rate reductions and the optimisation of mineral resource utilisation in society. Life cycle assessment (LCA) is a holistic methodology allowing for the analysis of systems and products and can be applied to waste management systems to identify environmental benefits and critical aspects thereof. From an LCA perspective, residue utilisation provides benefits such as avoiding the production and depletion of primary materials, but it can lead to environmental burdens, due to the potential leaching of toxic substances. In waste LCA studies where residue utilisation is included, leaching has generally been neglected. In this study, municipal solid waste incineration bottom ash (MSWI BA) was used as a case study into three LCA scenarios having different system boundaries. The importance of data quality and parameter selection in the overall LCA results was evaluated, and an innovative method to assess metal transport into the environment was applied, in order to determine emissions to the soil and water compartments for use in an LCA. It was found that toxic impacts as a result of leaching were dominant in systems including only MSWI BA utilisation, while leaching appeared negligible in larger scenarios including the entire waste system. However, leaching could not be disregarded a priori, due to large uncertainties characterising other activities in the scenario (e.g. electricity production). Based on the analysis of relevant parameters relative to leaching, and on general results of the study, recommendations are provided regarding the use of leaching data in LCA studies.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Local hybrid functionals with orbital-free mixing functions and balanced elimination of self-interaction error

The recently introduced density overlap regions indicator (DORI) [P. de Silva and C. Corminboeuf, J. Chem. Theory Comput. 10(9), 3745–3756 (2014)] is a density-dependent scalar field revealing regions of high density overlap between shells, atoms, and molecules. In this work, we exploit its properties to construct local hybrid exchange-correlation functionals aiming at balanced reduction of the self-interaction error. We show that DORI can successfully replace the ratio of the von Weizsäcker and exact positive-definite kinetic energy densities, which is commonly used in mixing functions of local hybrids. Additionally, we introduce several semi-empirical parameters to control the local and global admixture of exact exchange. The most promising of our local hybrids clearly outperforms the underlying semi-local functionals as well as their global hybrids.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Dynamics of O{sup 8+}+H electron capture in Debye plasmas

The electron capture in O{sup 8+}-H collisions is studied by the two-center atomic-orbital close-coupling method when the interactions of charged particles are screened and have a Yukawa form. Atomic orbitals and eigenenergies of n{<=}7 states on O{sup 7+} and n{<=}2 states on H are calculated as a function of the interaction screening parameter and used in the atomic-orbital close-coupling dynamics scheme to calculate the electron capture cross sections in the energy range of 0.2-50 keV/u. It is shown that the degree of interaction screening determines the reduction in electron binding energies, the number of open electron capture channels, and the strength of the exchange couplings, thus affecting the entire collision dynamics and the magnitude and energy behavior of state-selective cross sections. The changes in electron binding energies and capture cross sections when the interaction screening varies introduce dramatic changes in the radiation spectrum of O{sup 7+}(nl) capture states with respect to the unscreened interaction case. The state-selective electron capture cross sections in this collision system, as well as the intensities of a number of charge exchange spectral lines, for a number of representative screening parameter values are presented and discussed.

74 ATOMIC AND MOLECULAR PHYSICS↗

Designing a New Fuel for HFIR-Performance Parameters for LEU Core Configurations

An engineering design study for a fuel that would enable the conversion of the High Flux Isotope Reactor from highly enriched uranium to low enriched uranium fuel is ongoing as part of an effort sponsored by the U.S. Department of Energy's National Nuclear Security Administration through the Global Threat Reduction Initiative. Given the unique fuel and core design and high power density of the reactor and the requirement that the impact of the fuel change on the core performance and operation be minimal, this conversion study presents a complex and challenging task, requiring improvements in the computational models currently used to support the operation of the reactor and development of new models that would take advantage of newly available simulation methods and tools. The computational models used to search for a fuel design that would meet the requirements for the conversion study and the results obtained with these models are presented and discussed. Estimates of relevant reactor performance parameters for the low enriched uranium fuel core are presented and compared to the corresponding data for the currently operating highly enriched uranium fuel core.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Validation of On-Orbit Methodology for the Assessment of Cardiac Function and Changes in the Circulating Volume Using Ultrasound and Braslet-M Occlusion Cuffs

The objective of this joint U.S. - Russian project was the development and validation of an in-flight methodology to assess a number of cardiac and vascular parameters associated with circulating volume and its manipulation in long-duration space flight. Responses to modified Valsalva and Mueller maneuvers were measured by cardiac and vascular ultrasound (US) before, during, and after temporary volume reduction by means of Braslet-M thigh occlusion cuffs (Russia). Materials and Methods: The study protocol was conducted in 14 sessions on 9 ISS crewmembers, with an average exposure to microgravity of 122 days. Baseline cardiovascular measurements were taken by echocardiography in multiple modes (including tissue Doppler of both ventricles) and femoral and jugular vein imaging on the International Space Station (ISS). The Braslet devices were then applied and measurements were repeated after >10 minutes. The cuffs were then released and the hemodynamic recovery process was monitored. Modified Valsalva and Mueller maneuvers were used throughout the protocol. All US data were acquired by the HDI-5000 ultrasound system aboard the ISS (ATL/Philips, USA) during remotely guided sessions. The study protocol, including the use of Braslet-M for this purpose, was approved by the ISS Human Research Multilateral Review Board (HRMRB). Results: The effects of fluid sequestration on a number of echocardiographic and vascular parameters were readily detectable by in-flight US, as were responses to respiratory maneuvers. The overall volume status assessment methodology appears to be valid and practical, with a decrease in left heart lateral E (tissue Doppler) as one of the most reliable measures. Increase in the femoral vein cross-sectional areas was consistently observed with Braslet application. Other significant differences and trends within the extensive cardiovascular data were also observed. (Decreased - RV and LV preload indices, Cardiac Output, LV E all maneuvers, LV Stroke Volume). Conclusions: This Study: 1) Addressed specific aspects of operational space medicine and space physiology, including assessment of circulating volume disturbances 2) Expanded the applications of diagnostic ultrasound imaging and Doppler techniques in microgravity. 3) Used respiratory maneuvers against the background of acute circulating volume manipulations which appear to enhance our ability to noninvasively detect volume-dependency in a number of cardiac and vascular parameters. 4) Determined that Tei index is not clinically changed therefore contractility not altered in the face of reduced preload. 5) Determined that increased Femoral Vein Area indicating blood being sequestered in lower extremities correlates with reduced preload and cardiac output. 6) That Braslet may be the only feasible means of acutely treating high pressure pulmonary edema in reduced gravity environments.

Hamilton, Douglas↗

Deep multiscale model learning

The objective of this article is to design novel multi-layer neural networks for multiscale simulations of flows taking into account the observed fine data and physical modeling concepts. Our approaches use deep learning techniques combined with local multiscale model reduction methodologies to predict flow dynamics. Using reduced-order model concepts is important for constructing robust deep learning architectures since the reduced-order models provide fewer degrees of freedom. We consider flow dynamics in porous media as multi-layer networks in this work. More precisely, the solution (e.g., pressures and saturation) at the time instant n+1 depends on the solution at the time instant n and input parameters, such as permeability fields, forcing terms, and initial conditions. One can regard the solution as a multi-layer network, where each layer, in general, is a nonlinear forward map and the number of layers relates to the internal time steps. We will rely on rigorous model reduction concepts to define unknowns and connections between layers. It is critical to use reduced-order models for this purpose, which will identify the regions of influence and the appropriate number of variables. Furthermore, due to the lack of available observed fine data, the reduced-order model can provide us sufficient inexpensive data as needed. The designed deep neural network will be trained using both coarse simulation data which is obtained from the reduced-order model and observed fine data. We will present the main ingredients of our approach and numerical examples. Numerical results show that using deep learning with data generated from multiscale models as well as available observed fine data, we can obtain an improved forward map which can better approximate the fine scale model.

97 MATHEMATICS AND COMPUTING↗

Parametric Testing of Chevrons on Single Flow Hot Jets

A parametric family of chevron nozzles have been studied, looking for relationships between chevron geometric parameters, flow characteristics, and far-field noise. Both cold and hot conditions have been run at acoustic Mach number 0.9. Ten models have been tested, varying chevron count, penetration, length, and chevron symmetry. Four comparative studies were defined from these datasets which show: that chevron length is not a major impact on either flow or sound; that chevron penetration increases noise at high frequency and lowers it at low frequency, especially for low chevron counts; that chevron count is a strong player with good low frequency reductions being achieved with high chevron count without strong high frequency penalty; and that chevron asymmetry slightly reduces the impact of the chevron. Finally, it is shown that although the hot jets differ systematically from the cold one, the overall trends with chevron parameters is the same.

Bridges, James↗

SU-E-J-07: IGRT Gently: Evaluating Imaging Dose in Phantoms of Different Sizes

Purpose: IGRT imaging procedures have emerged as a common method of patient position verification in radiotherapy, though imaging dose is generally neglected in the treatment plan. Consequently, evaluating and optimizing the dose from these procedures is worthwhile. This process is especially important for children, who are more radiosensitive than adults. The aim of this work was to gain some understanding of the relative doses involved with various XVI-preset parameters for an “adult” and “child” phantom set, with the hopes that imaging dose for a child can be reduced. Methods: 32 and 16cm CTDI-phantoms were used as surrogates for adult and child torsos, respectively. Dose was measured in the central and peripheral chamber positions of the phantoms. CBCT scans were made for both phantoms using Elekta’s Chest-preset to establish a dose baseline. The child-phantom was then scanned using the Elekta Head and Neck (HN) preset. A modified HN-preset (named Peds Abd-pelvis) was also created with a doubled mAs to maintain a reduction in dose to the child-phantom (relative to the baseline), while providing clinically-usable image quality. Results: The baseline dose to the child-phantom from the Chest-preset was 310% that of the adult-phantom for the center chamber position and 150% at the periphery. An average dose reduction of 97% was obtained in the childphantom by switching from the Chest-preset to the HN-preset, while the Peds Abd-pelvis-preset similarly reduced the dose by an average of 92%. Conclusion: XVI-preset parameters significantly affect dose, and should be optimized to reduce dose, while ensuring clinically-usable image quality. Using a modified imaging preset (Peds Abd-pelvis-preset) greatly reduced the dose to the child-phantom compared to the dose for the Chest-preset for both the child and adult-phantoms. This outcome provides support for the development of child-specific protocols for IGRT imaging in pediatric patients.

60 APPLIED LIFE SCIENCES↗

Reduction of the type Ia supernova host galaxy step in the outer regions of galaxies

Using 1533 type Ia supernovae (SNe Ia) from the 5-yr sample of the Dark Energy Survey (DES), we investigate the relationship between the projected galactocentric separation of the SNe and their host galaxies and their light curves and standardization. We show, for the first time, that the difference in SN Ia post-standardization brightnesses between high- and low-mass hosts reduces from $0.078\pm 0.011$ mag in the full sample to $0.036 \pm 0.018$ mag for SNe Ia located in the outer regions of their host galaxies, while increasing to $0.100 \pm 0.014$ mag for SNe in the inner regions. The difference in the size of the mass step between inner and outer regions is $0.064\pm 0.023$ mag. In these inner regions, the step can be reduced (but not removed) using a model where the $R_V$ of dust along the line of sight to the SN changes as a function of galaxy properties. We investigate the remaining difference using the distributions of the SN Ia stretch parameter to test the inferred age of SN progenitors. Comparing red (older) environments only, outer regions have a higher proportion of high-stretch SNe and a more homogeneous stretch distribution. However, this effect cannot explain the reduction in significance of any Hubble residual step in outer regions. We conclude that the standardized distances of SNe Ia located in the outer regions of galaxies are less affected by their global host galaxy properties than those in the inner regions.

79 ASTRONOMY AND ASTROPHYSICS↗

Utilizing Environmental Tracers to Reduce Groundwater Flow and Transport Model Parameter Uncertainties

Abstract Non‐uniqueness in groundwater model calibration is a primary source of uncertainty in groundwater flow and transport predictions. In this study, we investigate the ability of environmental tracer information to constrain groundwater model parameters. We utilize a pilot point calibration procedure conditioned to subsets of observed data including: liquid pressures, tritium ( 3 H), chlorofluorocarbon‐12 (CFC‐12), and sulfur hexafluoride (SF 6 ) concentrations; and groundwater apparent ages inferred from these environmental tracers, to quantify uncertainties in the heterogeneous permeability fields and infiltration rates of a steady‐state 2‐D synthetic aquifer and a transient 3‐D model of a field site located near Riverton, Wyoming (USA). To identify the relative data worth of each observation data type, the post‐calibration uncertainties of the optimal parameters for a given observation subset are compared to that from the full observation data set. Our results suggest that the calibration‐constrained permeability field uncertainties are largest when liquid pressures are used as the sole calibration data set. We find significant reduction in permeability uncertainty and increased predictive accuracy when the environmental tracer concentrations, rather than apparent groundwater ages, are used as calibration targets in the synthetic model. Calibration of the Riverton field site model using environmental tracer concentrations directly produces infiltration rate estimates with the lowest uncertainties, however; permeability field uncertainties remain similar between the environmental tracer concentration and apparent groundwater age calibration scenarios. This work provides insight on the data worth of environmental tracer information to calibrate groundwater models and highlights potential benefits of directly assimilating environmental tracer concentrations into model parameter estimation procedures.

54 ENVIRONMENTAL SCIENCES↗

Experimental clean combustor program: Phase 3: Turbulence measurement addendum

Airflow turbulence parameters were measured in the high pressure, high temperature flow stream leaving the compressor of an operating gas turbine engine. A water cooled hot film turbulence probe was used to determine the turbulence intensity and length scale in the compressor exit flow stream of the CF6-50 engine. Data were obtained only at idle operating conditions. At engine power levels above idle conditions, durability problems and erratic data readings were encountered with the turbulence measurement probes. Turbulence measurements were made at three radial immersions at a point 15.5 cm downstream of the compressor outlet guide vanes in the compressor exit diffuser. The passage height at this point is 5.54 cm. Data reduction was accomplished with a time-data fast Fourier transform (FFT) system. This system finds the power spectral density distribution (PSD) of a large number of data samples using a direct Fourier transform algorithm and finds the autocorrelation parameter for these data by doing an FFT analysis of the PSD curves for a series of time delay intervals.

Taylor, J. R.↗

Model reduction for fractured porous media: a machine learning approach for identifying main flow pathways

Discrete fracture networks (DFN) are often used to model flow and transport in fractured porous media. The accurate resolution of flow and transport behavior on a large DFN involving thousands of fractures is computationally expensive. This makes uncertainty quantification studies of quantities of interest such as travel time through the network computationally intractable, since hundreds to thousands of runs of the DFN model are required to get bounds on the uncertainty of the predictions. Prior works on the subject demonstrated that the complexity of a DFN could be reduced by considering a sub-network of it (often termed a “backbone” sub-network), one whose flow and transport properties were then shown to be similar to that of the full network. The technique is tantamount to partitioning the complete set of fractures of a network into two disjoint sets, one of which is the backbone sub-network while the other its complement. It is in this context that we present a system-reduction technique for DFNs using supervised machine learning via a Random Forest Classifier that selects a backbone sub-network from the full set of fractures. The in-sample errors (in terms of precision and recall scores) of the trained classifier are found to be very accurate indicators of the out-of-sample errors, thus exhibiting that the classifier generalizes well to test data. Moreover, this system-reduction technique yields sub-networks as small as 12% of the full DFN that still recover transport characteristics of the full network such as the peak dosage and tailing behavior for late times. Most importantly, the sub-networks remain connected, and their size can be controlled by a single dimensionless parameter. Furthermore, measures of KL-divergence and KS-statistic for the breakthrough curves of the sub-networks with respect to the full network show physically realistic trends in that the measures decrease monotonically as the size of the sub-networks increase. In conclusion, the computational efficiency gained by this technique depends on the size of the sub-network, but large reductions in computational time can be expected for small sub-networks, yielding as much as 90% computational savings for sub-networks that are as small as 10-12% of the full network.

97 MATHEMATICS AND COMPUTING↗

Accurate and efficient prediction of fine-resolution hydrologic and carbon dynamic simulations from coarse-resolution models

The topography, and the biotic and abiotic parameters are typically upscaled to make watershed-scale hydrologic-biogeochemical models computationally tractable. However, upscaling procedure can produce biases when nonlinear interactions between different processes are not fully captured at coarse resolutions. Here, we applied the Proper Orthogonal Decomposition Mapping Method (PODMM) to downscale the field solutions from a coarse (7 km) resolution grid to a fine (220 m) resolution grid. PODMM trains a reduced-order model (ROM) with coarse-resolution and fine-resolution solutions, here obtained using PAWS+CLM, a quasi-3-D watershed processes model that has been validated for many temperate watersheds. Subsequent fine-resolution solutions were approximated based only on coarse-resolution solutions and the ROM. The approximation errors were efficiently quantified using an error estimator. By jointly estimating correlated variables and temporally varying the ROM parameters, we further reduced the approximation errors by up to 20%. We also improved the method's robustness by constructing multiple ROMs using different set of variables, and selecting the best approximation based on the error estimator. The ROMs produced accurate downscaling of soil moisture, latent heat flux, and net primary production with O(1000) reduction in computational cost. The subgrid distributions were also nearly indistinguishable from the ones obtained using the fine-resolution model. Compared to coarse-resolution solutions, biases in upscaled ROM solutions were reduced by up to 80%. This method has the potential to help address the long-standing spatial scaling problem in hydrology and enable long-time integration, parameter estimation, and stochastic uncertainty analysis while accurately representing the heterogeneities.

58 GEOSCIENCES↗