Search NASA⌕ Search

SEARCH · Search NASA

Results for “modeling discrepancy”

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 415 records · Page 23
All sourcesNASA NTRSDOE OSTI
NASA NTRS1988

Studies of the dependence of the microwave radar cross section on ocean surface variables during the FASINEX experiment

The ability of theoretical radar cross section (RCS) models to predict the absolute magnitude of the ocean radar cross section under a wide variety of sea and atmospheric conditions was studied using experimental data from the FASINEX Experiment. This consists of RCS data from a Ku-band scatterometer mounted on an aircraft (10 separate flights were conducted), a wide variety of atmospheric measurements (including stress) and sea conditions. Theoretical models are tested. Where discrepancies are observed, revisions are hypothesized and evaluated.

Weissman, D. A.↗
NASA NTRS1989

Resolution of the 1,238-keV gamma-ray line from supernova 1987A

Observations of supernova 1987A from the maiden flight of the Gamma-Ray Imaging Spectrometer (GRIS) are reported. SN1987A was observed for a period of 11.1 hours on May 1, 1988. Line emission at 1238 keV and continuum emission from 60-800 keV were detected. A gaussian line profile gives an acceptable fit to the 1238 keV line. The best-fit parameters are: flux = 8.5(+ 2.3, - 2.2) x 10 to the -4th photons/sq cm/s; peak energy = 1235.4 (+ 2.2, - 2.4) keV; FWHM = 16.3 (+ 6.1, - 5.7) keV. No evidence is found for a supernova-produced red- or blueshift in the 1238 keV line. The measured linewidth is a factor of about two greater than model predictions, although the discrepancy represents only two standard deviations. The line profiles are characteristic of optically thin regions, whereas the intensity implies a mean optical depth of about two. Fragmentation or nonspherical geometry of the supernova shell are possible explanations of the data.

Teegarden, B. J.↗
NASA NTRS1992

Neutral winds in the lower thermosphere from Dynamics Explorer 2

Doppler line profile measurements of the OI lambda 557.7 nm 'green line' emission, made by the Fabry-Perot interferometer on Dynamics Explorer 2, have provided altitude profiles of the meridional component of the lower thermospheric neutral wind. The wind inversion technique of Nardi has been used to extract the neutral wind profiles from the line-of-sight measurements. Individual lambda 557.7 nm Doppler line profiles and inverted volume-emission-rate and neutral-wind profiles are presented. Neutral wind measurements from the about 120 km altitude level, obtained on multiple orbital passes over the summer hemisphere polar region, have been merged to produce a synthesized averaged 'vector' wind field in geomagnetic coordinates. The wind pattern exhibits a region of anticyclonic vorticity in the daytime sector of the magnetic polar cap. Averaged winds of about 300 m/sec in the equatorial direction are observed in the early morning sector. The satellite winds are in reasonable agreement with the predictions of the NCAR and UCL thermosphere general circulation models, with significant regional discrepancies evident in both magnitude and direction.

Killeen, T. L.↗
NASA NTRS1993

On the effect of acoustic coupling on random and harmonic plate vibrations

The effect of acoustic coupling on random and harmonic plate vibrations is studied using two numerical models. In the coupled model, the plate response is obtained by integration of the nonlinear plate equation coupled with the nonlinear Euler equations for the surrounding acoustic fluid. In the uncoupled model, the nonlinear plate equation with an equivalent linear viscous damping term is integrated to obtain the response of the plate subject to the same excitation field. For a low-level, narrow-band excitation, the two models predict the same plate response spectra. As the excitation level is increased, the response power spectrum predicted by the uncoupled model becomes broader and more shifted towards the high frequencies than that obtained by the coupled model. In addition, the difference in response between the coupled and uncoupled models at high frequencies becomes larger. When a high intensity harmonic excitation is used, causing a nonlinear plate response, both models predict the same frequency content of the response. However, the level of the harmonics and subharmonics are higher for the uncoupled model. Comparisons to earlier experimental and numerical results show that acoustic coupling has a significant effect on the plate response at high excitation levels. Its absence in previous models may explain the discrepancy between predicted and measured responses.

Frendi, A.↗
NASA NTRS1993

Effect of acoustic coupling on random and harmonic plate vibrations

The effect of acoustic coupling on random and harmonic plate vibrations is studied using two numerical models. In the coupled model, the plate response is obtained by integration of the nonlinear plate equation coupled with the nonlinear Euler equations for the surrounding acoustic fluid. In the uncoupled model, the nonlinear plate equation with an equivalent linear viscous damping term is integrated to obtain the response of the plate subject to the same excitation field. For a low-level, narrow-band excitation, the two models predict the same plate response spectra. As the excitation level is increased, the response power spectrum predicted by the uncoupled model becomes broader and more shifted towards the high frequencies than that obtained by the coupled model. In addition, the difference in response between the coupled and uncoupled models at high frequencies becomes larger. When a high intensity harmonic excitation is used, causing a nonlinear plate response, both models predict the same frequency content of the response. However, the level of the harmonics and subharmonics are higher for the uncoupled model. Comparisons to earlier experimental and numerical results show that acoustic coupling has a significant effect on the plate response at high excitation levels. Its absence in previous models may explain the discrepancy between predicted and measured responses.

Frendi, Abdelkader↗
NASA NTRS1992

Effect of mean velocity shear on the dissipation rate of turbulent kinetic energy

The dissipation rate of turbulent kinetic energy in incompressible turbulence is investigated using a two-scale DIA. The dissipation rate is shown to consist of two parts; one corresponds to the dissipation rate used in the current turbulence models of eddy-viscosity type, and another comes from the viscous effect that is closely connected with mean velocity shear. This result can elucidate the physical meaning of the dissipation rate used in the current turbulence models and explain part of the discrepancy in the near-wall dissipation rates between the current turbulence models and direct numerical simulation of the Navier-Stokes equation.

Yoshizawa, Akira↗
NASA NTRS1996

The Compton Mirror in NGC 4151

We show that the sharp cutoff in the hard X-ray spectrum of NGC 4151, unusual for Seyfert 1 galaxies, can be reconciled with the average Seyfert 1 spectrum if we assume that the central source is completely hidden from our line of sight by the thick part of the accretion disk or by the broad emission-line clouds. The observed X-ray radiation is produced by scattering of the Seyfert 1 type spectrum in the higher, cooler parts of the accretion disk corona, or in a wind. A sharp cutoff appears as a result of the Compton recoil effect. This model naturally explains a discrepancy regarding the inclination of the central source, inferred to be low (face-on) from observations of the iron K-alpha emission line, but inferred to be high on the basis of optical and UV observations.

Poutanen, Juri↗
NASA NTRS1997

On the Floating Point Performance of the i860 Microprocessor

The i860 microprocessor is a pipelined processor that can deliver two double precision floating point results every clock. It is being used in the Touchstone project to develop a teraflop computer by the year 2000. With such high computational capabilities it was expected that memory bandwidth would limit performance on many kernels. Measured performance of three kernels showed performance is less than what memory bandwidth limitations would predict. This paper develops a model that explains the discrepancy in terms of memory latencies and points to some problems involved in moving data from memory to the arithmetic pipelines.

Lee, King↗
NASA NTRS2010

Hard X-Ray Footprint Source Sized

RHESSI has detected compact hard (25 - 100 keV) X-ray sources that are <4 arcseconds (FWHM) in extent for certain flares (Dennis and Pernak (2009). These sources are believed to be at magnetic loop footpoints that are known from observations at other wavelengths to be very small. Flare ribbons seen in the W with TRACE, for example, are approx. 1 arcsecond in width, and white light flares show structure at the approx. 1 arcsecond level. However, Kontar and Jeffrey (2010) have shown that the measured extent should be >6 arcseconds, even if the X-ray emitting thick-target source is point-like. This is because of the strong albedo contribution in the measured energy range for a source located at the expected altitude of 1 Mm near the top of the chromosphere. This discrepancy between observations and model predictions may indicate that the source altitude is significantly lower than assumed or that the RHESSI image reconstruction procedures are not sensitive to the more diffuse albedo patch in the presence of a strong compact source. Results will be presented exploring the latter possibility using the Pixon image reconstruction procedure and other methods based on visibilities.

Dennis, Brian R.↗
NASA NTRS2016

Reflectivity Around the Gold M-Edges of X-ray Reflector of the Soft X-Ray Telescope Onboard ASTRO-H

The X-ray astronomy satellite ASTRO-H are equipped with two equivalent soft X-ray telescopes (SXT-I and SXT-S) which cover the energy band 0.3-12 keV. The X-ray reflectors of the SXTs are coated with a gold monolayer by means of the replication technique. A series of gold M absorption edges in the 2-4 keV band causes complex structures in the energy response of the SXTs. In the same band, there are astrophysically important emission lines from Si, Ar and S. Since the SXS has unprecedentedly high spectral resolution, we have measured the reflectivity around the gold M-edges in an extremely fine energy pitch at the synchrotron radiation facility KEK PF BL11-B, with the 2 eV pitch in 2100 eV to 4100 eV band that covers the entire series of the absorption edges (M-I through M-V) at grazing incident angles to the reflectors of 0.5, 0.8, 1.0, 1.2, 1.4 degree, and with a finer pitch of 0.25 eV in the 2200 eV to 2350 eV band where the two deepest M-IV and M-V edges are included. In the resultant reflectivity curves, we have clearly identified the fine structures associated with all the M-edges. Using these data, we calculated atomic scattering factor f1 as a function of X-ray energy, with which we have built the mirror response function which can be applied to the Suzaku spectra. As a result, we have found that discrepancy of the spectral model to the Suzaku data of 4U1630-472 (a black hole transient) and the Crab nebula around the M-edges are significantly reduced from those with the official Suzaku response.

ASTRO-H↗
NASA NTRS2018

Maps of Tethys' Thermophysical Properties

On 11th April 2015 Cassini's Composite Infrared Spectrometer (CIRS) made a series of observations of Tethys’ daytime anti-Saturn hemisphere over a nine-hour time period. During this time the sub-spacecraft position was remarkably stable (0.3° S to 3.9° S; 153.2° W to 221.8° W), and so these observations provide unprecedented coverage of diurnal temperature variations on Tethys’ anti-Saturn hemisphere. In 2012 a thermal anomaly was discovered at low latitudes on Tethys’ leading hemisphere; it appears cooler during the day and warmer at night than its surroundings (Howett et al., 2012) and is spatially correlated with a decrease in the IR3/UV3 visible color ratio (Schenk et al., 2011). The cause of this anomaly is believed to be surface alteration by high-energy electrons, which preferentially bombard low-latitudes of Tethys’ leading hemisphere (Schenk et al., 2011; Howett et al., 2012; Paranicas et al. 2014; Schaible et al., 2017). The thermal anomaly was quickly dubbed “Pac-Man” due to its resemblance to the 1980s video game icon. We use these daytime 2015 CIRS data, along with two sets of nighttime CIRS observations of Tethys (from 27 June 2007 and 17 August 2015) to make maps of bolometric Bond albedo and thermal inertia variations across the anti-Saturn hemisphere of Tethys (including the edge of its Pac-Man region). These maps confirm the presence of the Pac-Man thermal anomaly and show that while Tethys’ bolometric Bond albedo varies negligibly outside and inside the anomaly (0.69 plus or minus 0.02 inside, compared to 0.71 plus or minus 0.04 outside) the thermal inertia varies dramatically (29 plus or minus 10 J m−2 K−1 s−1/2 inside, compared to 9 plus or minus 4 J m−2 K−1 s−1/2 outside). These thermal inertias are in keeping with previously published values: 25 plus or minus 3 J m−2 K−1 s−1/2 inside, and 5 ± 1 J m−2 K−1 s−1/2 outside the anomaly (Howett et al., 2012). A detailed analysis shows that on smaller spatial-scales the bolometric Bond albedo does vary: increasing to a peak value at 180° W. For longitudes between approximately 100° W and approximately 160° W the thermal inertia increases from northern to southern latitudes, while the reverse is true for bolometric Bond albedo. The thermal inertia on Tethys generally increases towards the center of its leading hemisphere but also displays other notable small-scale variations. These thermal inertia and bolometric Bond albedo variations are perhaps due to differences in competing surface modification by E ring grains and high-energy electrons which both bombard Tethys’ leading hemisphere (but in different ways). A comparison between the observed temperatures and our best thermal model fits shows notable discrepancies in the morning warming curve, which may provide evidence of regional variations in surface roughness effects, perhaps again due to variations in surface alteration mechanisms.

Howett, C. J. A.↗
NASA NTRS2023

Comparison Between Ozone Column Depths and Methane Lifetimes Computed By One- and Three-Dimensional Models at Different Atmospheric O(2) Levels

The calculation of ozone column depth as a function of atmospheric O2concentration has been performed multiple times over the past 40 years using one-dimensional (1-D) photochemical models. Several of these calculations [1–3] have been done by the Kasting group. We refer to the current version of that model as the‘standard’1-D model (see §2). Recently, Cooke et al., henceforth C22 ,performed a similar calculation using a three-dimensional (3-D) coupled chemistry-climate model (WACCM6). They found substantially lower ozone column depths at lower pO2levels. For example, at0.01 times the present atmospheric level (PAL) of O(2), Segura et al. calculated 124 Dobson units(DU), whereas C22 calculated a mean column depth of just 66 DU—almost a factor of two lower. This comparison is shown in figure 1. C22 argued, correctly, that 3-D models should be able to do a better job on this problem than 1-D models because ozone column depths are affected by atmospheric transport, e.g. the Brewer–Dobson circulation in the stratosphere that blows ozone from the tropics towards the poles, as well as vertical transport that carries ozone downward from the stratosphere to the troposphere. Meridional transport is absent in 1-D models, and vertical transport is fixed, so model geometry almost certainly contributes to the observed discrepancies in ozone column depth. The 3-D model also has a variable tropopause height (higher at the equator, lower towards the poles), which is difficult to simulate in one dimension.

Atmospheric chemistry↗
NASA NTRS2004

Core Formation in the Earth: Constraints from Ni and Co

Due to their metal-loving nature, Ni and Co were strongly partitioned into the metallic core and were left depleted in the silicate mantle during core formation in the Earth. Based on experimental liquid metal- liquid silicate partition coefficients (D), studies have suggested that core formation in an early magma Ocean can explain the observed mantle depletions of Ni and Co [l-51. However, the conditions proposed by the magma ocean models have ranged from pressures of 24 to 59 GPa and temperatures of 2200 to < 4000 K. Furthermore, the proposed magma Ocean oxygen fugacities have differed by nearly two orders of magnitude.Chabot and Agee noted that the different models predicted contradictory behaviors for D(Ni) and D(Co) as a function of temperature. With the hope of resolving the discrepancies between the magma ocean models, we conducted a systematic experimental study to constrain the effects of temperature on D(Ni) and D(Co).

Chabot, N. L.↗
NASA NTRS1994

Dust extinction of the stellar continua in starburst galaxies: The ultraviolet and optical extinction law

We analyze the International Ultraviolet Explorer (IUE) UV and the optical spectra of 39 starburst and blue compact galaxies in order to study the average properties of dust extinction in extended regions of galaxies. The optical spectra have been obtained using an aperture which matches that of IUE, so comparable regions within each galaxy are sampled. The data from the 39 galaxies are compared with five models for the geometrical distribution of dust, adopting as extinction laws both the Milky Way and the Large Magellanic Cloud laws. The commonly used uniform dust screen is included among the models. We find that none of the five models is in satisfactory agreement with the data. In order to understand the discrepancy between the data and the models, we have derived an extinction law directly from the data in the UV and optical wavelength range. The resulting curve is characterized by an overall slope which is more gray than the Milky Way extinction law's slope, and by the absence of the 2175 A dust feature. Remarkably, the difference in optical depth between the Balmer emission lines H(sub alpha) and H(sub beta) is about a factor of 2 larger than the difference in the optical depth between the continuum underlying the two Balmer lines. We interpret this discrepancy as a consequence of the fact that the hot ionizing stars are associated with dustier regions than the cold stellar population is. The absence of the 2175 A dust feature can be due either to the effects of the scattering and clumpiness of the dust or to a chemical composition different from that of the Milky Way dust grains. Disentangling the two interpretations is not easy because of the complexity of the spatial distribution of the emitting regions. The extinction law of the UV and optical spectral continua of extended regions can be applied to the spectra of medium- and high-redshift galaxies, where extended regions of a galaxy are, by necessity, sampled.

Calzetti, Daniela↗
DOE OSTI2025

Anomalous electroweak physics unraveled via evidential deep learning

The ever-growing ecosystem of beyond standard model (BSM) calculations and parametrizations has motivated the development of systematic methods for making quantitative cross-comparisons over the wide range of possible models, especially with controllable uncertainties. In this setting, the language of uncertainty quantification (UQ) furnishes useful metrics for assessing statistical overlaps and discrepancies among BSM and related models. In this study, we leverage recent machine learning (ML) developments in evidential deep learning (EDL) for UQ to separate data (aleatoric) and knowledge (epistemic) uncertainties in a model-discrimination setting. We construct several potentially BSM-motivated scenarios for the anomalous electroweak interaction (AEWI) of neutrinos with nucleons in deep inelastic scattering ( v DIS). These scenarios are then quantitatively mapped, as a demonstration, alongside Monte Carlo replicas of the CT18 PDFs used to calculate the $\varDelta \chi ^{2}$ statistic for a typical multi-GeV v DIS experiment, CDHSW. Our framework effectively highlights areas of model agreement and provides a classification of out-of-distribution (OOD) samples. By offering the opportunity to quantitatively understand model overlaps, the approach presented in this work can help facilitate efficient BSM model exploration and exclusion for future New Physics searches.

AI↗
DOE OSTI2020

Experimentally Validated Computational Modeling of Creep and Creep-Cracking for Nuclear Concrete Structures

In a Nuclear Power Plant, one of the most important components is the concrete nuclear reactor cavity, which serves both a structural and protective function as the biological radiation shield. Given that creep has been identified as a major knowledge gap in the assessment of nuclear structures (NUREG/CR-7153), this work helps to further the understanding of creep behavior of massive concrete containment structures for decades to enable safe and long-term operation of these facilities. This project has developed a robust, experimentally validated model to predict creep in nuclear concrete structures for up 60 years using short-term creep data thereby enabling a longer service life of critical facilities and early detection of structural failure. The work presented in this report is a pairing of computational and experimental methods. For the first time, the time temperature superposition (TTS) principle was successfully used to generate a uniaxial creep compliance master curve to predict mortar creep response for up to 22,500 days (nearly 60 years) at a reference temperature of 20°C. These data were used as input into finite element analysis (FEA) codes that use highly realistic random, 3D concrete microstructures from reconstructed coarse limestone aggregates. Finite element analysis performed provides the ability to quickly upscale mortar viscoelastic behavior to long-term concrete creep/relaxation data. A master creep compliance curve, constructed from the TTS principle, spanning 27 years, was used to validate two and a half decades of simulated concrete creep. Concurrently, three different simulated wall specimens were designed to mimic the behavior of post-tensioned concrete nuclear containment facility vessel walls over time as a result of concrete creep. The specimens were designed with different thicknesses, transverse and longitudinal reinforcement ratios, and level of post-tensioning stress. Each specimen contained various instrumentation to measure internal concrete temperature, concrete strain, and post-tensioning strain hourly for over 3 years. The concrete creep model developed in this project, based on the FEA concrete simulations, was applied to simulate the structural-scale experiments of prestressed concrete walls conducted in this project using the Grizzly code. These models can represent the effects of reinforcing and prestressing. Although there are some discrepancies with the experimental data, the model can predict the overall trends of the creep response in these experiments. One of these experimental models was also applied to an extended time to demonstrate how the findings from this study can be used to predict the behavior of actual structures of interest that have been in service for extended periods of time.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRS1981

The Theory and Practice of Estimating the Accuracy of Dynamic Flight-Determined Coefficients

Means of assessing the accuracy of maximum likelihood parameter estimates obtained from dynamic flight data are discussed. The most commonly used analytical predictors of accuracy are derived and compared from both statistical and simplified geometrics standpoints. The accuracy predictions are evaluated with real and simulated data, with an emphasis on practical considerations, such as modeling error. Improved computations of the Cramer-Rao bound to correct large discrepancies due to colored noise and modeling error are presented. The corrected Cramer-Rao bound is shown to be the best available analytical predictor of accuracy, and several practical examples of the use of the Cramer-Rao bound are given. Engineering judgement, aided by such analytical tools, is the final arbiter of accuracy estimation.

Maine, R. E.↗
NASA NTRS1998

Design, Evaluation and GCM-Performance of a New Parameterization for Microphysics of Clouds with Relaxed Arakawa-Schubert Scheme (McRas)

A prognostic cloud scheme named McRAS (Microphysics of clouds with Relaxed Arakawa-Schubert Scheme) was developed with the aim of improving cloud-microphysics, and cloud-radiation interactions in GCMs. McRAS distinguishes convective, stratiform, and boundary-layer clouds. The convective clouds merge into stratiform clouds on an hourly time-scale, while the boundary-layer clouds do so instantly. The cloud condensate transforms into precipitation following the auto-conversion relations of Sundqvist that contain a parametric adaptation for the Bergeron-Findeisen process of ice crystal growth and collection of cloud condensate by precipitation. All clouds convect, advect, and diffuse both horizontally and vertically with a fully active cloud-microphysics throughout its life-cycle, while the optical properties of clouds are derived from the statistical distribution of hydrometeors and idealized cloud geometry. An evaluation of McRAS in a single column model (SCM) with the GATE Phase III data has shown that McRAS can simulate the observed temperature, humidity, and precipitation without discernible systematic errors. An evaluation with the ARM-CART SCM data in a cloud model intercomparison exercise shows reasonable but not an outstanding accurate simulation. Such a discrepancy is common to almost all models and is related, in part, to the input data quality. McRAS was implemented in the GEOS II GCM. A 50 month integration that was initialized with the ECMWF analysis of observations for January 1, 1987 and forced with the observed sea-surface temperatures and sea-ice distribution and vegetation properties (biomes, and soils), with prognostic soil moisture, snow-cover, and hydrology showed a very realistic simulation of cloud process, incloud water and ice, and cloud-radiative forcing (CRF). The simulated ITCZ showed a realistic time-mean structure and seasonal cycle, while the simulated CRF showed sensitivity to vertical distribution of cloud water which can be easily altered by the choice of time constant and incloud critical cloud water amount regulators for auto-conversion. The CRF and its feedbacks also have a profound effect on the ITCZ. Even though somewhat weaker than observed, the McRAS-GCM simulation produces robust 30-60 day oscillations in the 200 hPa velocity potential. Two ensembles of 4-summer (July, August, September) simulations, one each for 1987 and 1988 show that the McRAS-GCM simulates realistic and statistically significant precipitation differences over India, Central America, and tropical Africa. Several seasonal simulations were performed with McRAS-GEOS II GCM for the summer (June-July- August) and winter (December-January-February) periods to determine how the simulated clouds and CRFs would be affected by: i) advection of clouds; ii) cloud top entrainment instability, iii) cloud water inhomogeneity correction, and (iv) cloud production and dissipation in different cloud-processes. The results show that each of these processes contributes to the simulated cloud-fraction and CRF.

Sud, Y. C.↗