Search NASA⌕ Search

SEARCH · Search NASA

Results for “Normalization”

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 559 records · Page 31

Closeness to spheres of hypersurfaces with normal curvature bounded below

For a Riemannian manifold M{sup n+1} and a compact domain Ω⊂ M{sup n+1} bounded by a hypersurface ∂Ω with normal curvature bounded below, estimates are obtained in terms of the distance from O to ∂Ω for the angle between the geodesic line joining a fixed interior point O in Ω to a point on ∂Ω and the outward normal to the surface. Estimates for the width of a spherical shell containing such a hypersurface are also presented. Bibliography: 9 titles.

97 MATHEMATICS AND COMPUTING↗

THE LARSON-TINSLEY EFFECT IN THE ULTRAVIOLET: INTERACTING VERSUS 'NORMAL' SPIRAL GALAXIES

We compare the UV-optical colors of a well-defined set of optically selected pre-merger interacting galaxy pairs with those of normal spirals. The shorter wavelength colors show a larger dispersion for the interacting galaxies than for the spirals. This result can best be explained by higher star formation rates on average in the interacting galaxies, combined with higher extinctions on average. This is consistent with earlier studies which found that the star formation in interacting galaxies tends to be more centrally concentrated than in normal spirals, perhaps due to gas being driven into the center by the interaction. As noted in earlier studies, there is a large variation from galaxy to galaxy in the implied star formation rates of the interacting galaxies, with some galaxies having enhanced rates but others being fairly quiescent.

79 ASTRONOMY AND ASTROPHYSICS↗

NEAR-INFRARED POLARIMETRY OF A NORMAL SPIRAL GALAXY VIEWED THROUGH THE TAURUS MOLECULAR CLOUD COMPLEX

Few normal galaxies have been probed using near-infrared polarimetry, even though it reveals magnetic fields in the cool interstellar medium better than either optical or radio polarimetry. Deep H-band (1.6 {mu}m) linear imaging polarimetry toward Taurus serendipitously included the galaxy 2MASX J04412715+2433110 with adequate sensitivity and resolution to map polarization across nearly its full extent. The observations revealed the galaxy to be a steeply inclined ({approx}75 Degree-Sign ) disk type with a diameter, encompassing 90% of the Petrosian flux, of 4.2 kpc at a distance of 53 Mpc. Because the sight line passes through the Taurus Molecular Cloud complex, the foreground polarization needed to be measured and removed. The foreground extinction A{sub V} of 2.00 {+-} 0.10 mag and reddening E(H - K) of 0.125 {+-} 0.009 mag were also assessed and removed, based on analysis of Two Micron All Sky Survey, UKIRT Infrared Deep Sky Survey, Spitzer, and Wide-field Infrared Survey Explorer photometry using the Near-Infrared Color Excess, NICE-Revisited, and Rayleigh-Jeans Color Excess methods. Corrected for the polarized foreground, the galaxy polarization values range from 0% to 3%. The polarizations are dominated by a disk-parallel magnetic field geometry, especially to the northeast, while either a vertical field or single scattering of bulge light produces disk-normal polarizations to the southwest. The multi-kiloparsec coherence of the magnetic field revealed by the infrared polarimetry is in close agreement with short-wavelength radio synchrotron observations of edge-on galaxies, indicating that both cool and warm interstellar media of disk galaxies may be threaded by common magnetic fields.

79 ASTRONOMY AND ASTROPHYSICS↗

Star formation in the central regions of active and normal galaxies

We investigated the properties of the host galaxies of active galactic nuclei (AGNs). We used radio emission around the nuclei of the host galaxies to represent AGN activity and used infrared (IR) emission to represent the star-forming activity and stellar population of the host galaxies. We determined that active galaxies have higher stellar masses (SMs) within the central kiloparsec radius than normal galaxies do independent of the Hubble types of the host galaxies; but both active and normal galaxies exhibit similar specific star formation rates (SSFRs), ranging between 10 -10.5 and 10 -9.5 yr -1 . We find that the central SM surface density might be used as an indicator to identify AGNs. We also discovered that certain AGNs exhibit substantial inner stellar structures in the IR images; most of the AGNs with inner structures are Seyferts, whereas only a few LINERs exhibit inner structures. We note that the AGNs with inner structures show a positive correlation between the radio activity of the AGNs and the SFRs of the host galaxies, but the sources without inner structures show a negative correlation between the radio power and the SFRs. These results might be explained with a scenario of starburst–AGN evolution. In this scenario, AGN activities are triggered following a nuclear starburst; during the evolution, AGN activities are accompanied by SF activity in the inner regions of the host galaxies; at the final stage of the evolution, the AGNs might transform into LINERs, exhibiting weak SF activity in the central regions of the host galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

ENHANCING CORONAL STRUCTURES WITH THE FOURIER NORMALIZING-RADIAL-GRADED FILTER

Images of the corona have a high dynamic range which is excellent for quantitative photometric analysis. To understand the processes governing the solar corona, it is essential to have information about the absolute brightness as well as the underlying structure. However, due to the steep radial gradient of brightness in the images, and to the fact that structures closer to the solar disk have higher contrast than structures further from the disk, human vision cannot view the intricate structure of the corona in such images. The recently developed normalizing-radial-graded filter (NRGF) is an effective way for revealing the coronal structure. In this work, we present a more adaptive filter inspired by the NRGF, which we call the Fourier normalizing-radial-graded filter (FNRGF). It approximates the local average and the local standard deviation by a finite Fourier series. This method enables the enhancement of finer details, especially in regions of lower contrast. We also show how the influence of additive noise is reduced by a modification to the FNRGF. To illustrate the power of the method, the FNRGF is applied to images of emission from coronal forbidden lines observed during the 2010 July 11 total solar eclipse. It is also successfully applied to space-based observations of the low corona in the extreme ultraviolet and to white light coronagraph observations, thus demonstrating the validity of the FNRGF as a new tool that will help the interpretation of the information embedded in most types of coronal images.

79 ASTRONOMY AND ASTROPHYSICS↗

THE RELATIVE ABUNDANCE OF COMPACT AND NORMAL MASSIVE EARLY-TYPE GALAXIES AND ITS EVOLUTION FROM REDSHIFT z {approx} 2 TO THE PRESENT

We report on the evolution of the number density and size of early-type galaxies (ETGs) from z {approx} 2 to z {approx} 0. We select a sample of 563 massive (M > 10{sup 10} M{sub Sun }), passively evolving (specific star formation rate <10{sup -2} Gyr{sup -1}), and morphologically spheroidal galaxies at 0 < z < 2.5, using the panchromatic photometry and spectroscopic redshifts available in the Great Observatories Origins Deep Surveys fields. We combine Advanced Camera for Surveys and Wide Field Camera 3 Hubble Space Telescope images to study the morphology of our galaxies in their optical rest frame in the entire 0 < z < 2.5 range. We find that throughout the explored redshift range the passive galaxies selected with our criteria have weak morphological K-correction, with size being slightly smaller in the optical than in the UV rest frame (by {approx}20% and {approx}10% at z > 1.2 and z < 1.2, respectively). We measure a significant evolution of the mass-size relation of ETGs, with a fractional increment that is almost independent of the stellar mass. ETGs formed at z > 1 appear to be preferentially small, and the evolution of the mass-size relation at z < 1 is driven by both the continuous size growth of the compact galaxies and the appearance of new ETGs with large sizes. We also find that the number density of all passive ETGs increases rapidly, by a factor of five, from z {approx} 2 to z {approx} 1, and then more mildly by another factor of 1.5 from z {approx} 1 to z {approx} 0. We interpret these results as evidence that the bulk of the ETGs are formed at 1 < z < 3 through a mechanism that leaves very compact remnants. At z < 1 the compact ETGs grow gradually in size, becoming normal-size galaxies, and at the same time new ETGs with normal-large sizes are formed.

79 ASTRONOMY AND ASTROPHYSICS↗

A PHYSICAL MODEL FOR SN 2001ay, A NORMAL, BRIGHT, EXTREMELY SLOW DECLINING TYPE Ia SUPERNOVA

We present a study of the peculiar Type Ia supernova 2001ay (SN 2001ay). The defining features of its peculiarity are high velocity, broad lines, and a fast rising light curve, combined with the slowest known rate of decline. It is one magnitude dimmer than would be predicted from its observed {Delta}m{sub 15}, and shows broad spectral features. We base our analysis on detailed calculations for the explosion, light curves, and spectra. We demonstrate that consistency is key for both validating the models and probing the underlying physics. We show that this SN can be understood within the physics underlying the {Delta}m{sub 15} relation, and in the framework of pulsating delayed detonation models originating from a Chandrasekhar mass, M{sub Ch}, white dwarf, but with a progenitor core composed of 80% carbon. We suggest a possible scenario for stellar evolution which leads to such a progenitor. We show that the unusual light curve decline can be understood with the same physics as has been used to understand the {Delta}m{sub 15} relation for normal SNe Ia. The decline relation can be explained by a combination of the temperature dependence of the opacity and excess or deficit of the peak luminosity, {alpha}, measured relative to the instantaneous rate of radiative decay energy generation. What differentiates SN 2001ay from normal SNe Ia is a higher explosion energy which leads to a shift of the {sup 56}Ni distribution toward higher velocity and {alpha} < 1. This result is responsible for the fast rise and slow decline. We define a class of SN 2001ay-like SNe Ia, which will show an anti-Phillips relation.

79 ASTRONOMY AND ASTROPHYSICS↗

THE EGNoG SURVEY: GAS EXCITATION IN NORMAL GALAXIES AT z Almost-Equal-To 0.3

As observations of molecular gas in galaxies are pushed to lower star formation rate (SFR) galaxies at higher redshifts, it is becoming increasingly important to understand the conditions of the gas in these systems to properly infer their molecular gas content. The rotational transitions of the carbon monoxide (CO) molecule provide an excellent probe of the gas excitation conditions in these galaxies. In this paper, we present the results from the gas excitation sample of the Evolution of molecular Gas in Normal Galaxies (EGNoG) survey at the Combined Array for Research in Millimeter-wave Astronomy (CARMA). This subset of the full EGNoG sample consists of four galaxies at z Almost-Equal-To 0.3 with SFRs of 40-65 M {sub Sun} yr{sup -1} and stellar masses of Almost-Equal-To 2 Multiplication-Sign 10{sup 11} M {sub Sun }. Using the 3 mm and 1 mm bands at CARMA, we observe both the CO(J = 1 {yields} 0) and CO(J = 3 {yields} 2) transitions in these four galaxies in order to probe the excitation of the molecular gas. We report robust detections of both lines in three galaxies (and an upper limit on the fourth), with an average line ratio, r {sub 31} = L'{sub CO(3-2)}/L'{sub CO(1-0)}, of 0.46 {+-} 0.07 (with systematic errors {approx}< 40%), which implies sub-thermal excitation of the CO(J = 3 {yields} 2) line. We conclude that the excitation of the gas in these massive, highly star-forming galaxies is consistent with normal star-forming galaxies such as local spirals, not starbursting systems like local ultraluminous infrared galaxies. Since the EGNoG gas excitation sample galaxies are selected from the main sequence (MS) of star-forming galaxies, we suggest that this result is applicable to studies of MS galaxies at intermediate and high redshifts, supporting the assumptions made in studies that find molecular gas fractions in star-forming galaxies at z {approx} 1-2 to be an order of magnitude larger than what is observed locally.

79 ASTRONOMY AND ASTROPHYSICS↗

MODELING THE REDSHIFT EVOLUTION OF THE NORMAL GALAXY X-RAY LUMINOSITY FUNCTION

Emission from X-ray binaries (XRBs) is a major component of the total X-ray luminosity of normal galaxies, so X-ray studies of high-redshift galaxies allow us to probe the formation and evolution of XRBs on very long timescales ({approx}10 Gyr). In this paper, we present results from large-scale population synthesis models of binary populations in galaxies from z = 0 to {approx}20. We use as input into our modeling the Millennium II Cosmological Simulation and the updated semi-analytic galaxy catalog by Guo et al. to self-consistently account for the star formation history (SFH) and metallicity evolution of each galaxy. We run a grid of 192 models, varying all the parameters known from previous studies to affect the evolution of XRBs. We use our models and observationally derived prescriptions for hot gas emission to create theoretical galaxy X-ray luminosity functions (XLFs) for several redshift bins. Models with low common envelope efficiencies, a 50% twins mass ratio distribution, a steeper initial mass function exponent, and high stellar wind mass-loss rates best match observational results from Tzanavaris and Georgantopoulos, though they significantly underproduce bright early-type and very bright (L{sub x} > 10{sup 41}) late-type galaxies. These discrepancies are likely caused by uncertainties in hot gas emission and SFHs, active galactic nucleus contamination, and a lack of dynamically formed low-mass XRBs. In our highest likelihood models, we find that hot gas emission dominates the emission for most bright galaxies. We also find that the evolution of the normal galaxy X-ray luminosity density out to z = 4 is driven largely by XRBs in galaxies with X-ray luminosities between 10{sup 40} and 10{sup 41} erg s{sup -1}.

79 ASTRONOMY AND ASTROPHYSICS↗

STELLAR ORBITAL STUDIES IN NORMAL SPIRAL GALAXIES. I. RESTRICTIONS TO THE PITCH ANGLE

We built a family of non-axisymmetric potential models for normal non-barred or weakly barred spiral galaxies as defined in the simplest classification of galaxies: the Hubble sequence. For this purpose, a three-dimensional self-gravitating model for the spiral arm PERLAS is superimposed on the galactic axisymmetric potentials. We analyze the stellar dynamics varying only the pitch angle of the spiral arms, from 4 Degree-Sign to 40 Degree-Sign for an Sa galaxy, from 8 Degree-Sign to 45 Degree-Sign for an Sb galaxy, and from 10 Degree-Sign to 60 Degree-Sign for an Sc galaxy. Self-consistency is indirectly tested through periodic orbital analysis and through density response studies for each morphological type. Based on ordered behavior, periodic orbit studies show that, for pitch angles up to approximately 15 Degree-Sign , 18 Degree-Sign , and 20 Degree-Sign for Sa, Sb, and Sc galaxies, respectively, the density response supports the spiral arms' potential, a requisite for the existence of a long-lasting large-scale spiral structure. Beyond those limits, the density response tends to ''avoid'' the potential imposed by maintaining lower pitch angles in the density response; in that case, the spiral arms may be explained as transient features rather than long-lasting large-scale structures. In a second limit, from a phase-space orbital study based on chaotic behavior, we found that for pitch angles larger than {approx}30 Degree-Sign , {approx}40 Degree-Sign , and {approx}50 Degree-Sign for Sa, Sb, and Sc galaxies, respectively, chaotic orbits dominate the all phase-space prograde region that surrounds the periodic orbits sculpting the spiral arms and even destroying them. This result seems to be in good agreement with observations of pitch angles in typical isolated normal spiral galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Feasible voltage-tap based quench detection in a Ag/Bi-2212 coil enabled by fast 3D normal zone propagation

For this study, small insert solenoids have been built using a commercial Ag/Bi-2212 multifilamentary round wire, insulated with a new thin TiO 2 – polymer coating insulation (thickness in ~20 μm versus ~100 μm for a commonly used mullite braided sleeve insulation), and characterized in background magnetic field up to 14 T at 4.2 K to explore the high-field performance and quench detection of Bi-2212 magnets. The coil has no visible leakage and no electrical shorts after reaction, and it carries 280 A/mm -2 in a background field 14 T and generates an additional 1.7 T. A notable result is that, despite normal zones propagate slowly along the conductor, the hot spot temperature upon detection increases only from 40 K to 60 K when the resistive quench detection voltage threshold increases from 0.1 V to 1 V for all operating current density investigated, showing that quench detection using voltage taps is feasible for this coil. This is in a strong contrast to a coil we previously built to the same specifications but from wires insulated with the mullite braided sleeve insulation, for which the hot spot temperature upon detection increases from ~80 K to ~140 K while increasing from the detection voltage threshold from 0.1 V to 1 V, and thus for which quench detection using voltage taps presents significant risks, consistent with the common belief that the effectiveness of quench detection using voltage taps for superconducting magnets built using high temperature superconductors is seriously compromised by their slow normal zone propagation. This striking difference is ascribed to the fast transverse quench propagation enabled by thin insulation and improved thermal coupling between conductor turns. Finally, this work demonstrates that quench detection for high-temperature superconducting magnets highly depends on the design and construction of the coils such as insulation materials used and this dependence should be factored into the overall magnet design.

36 MATERIALS SCIENCE↗

Current profile and normalized beta control via feedback linearization and Lyapunov techniques

Abstract Simultaneous control of the current profile and normalized plasma beta is an essential control problem in the development of advanced tokamak scenarios. However, this control problem is especially challenging due to the nonlinear nature of the current, heat, and particle transport dynamics, as well as the difficulty to understand and accurately model such processes. In this work, a nonlinear, robust, model-based controller for the simultaneous regulation of the current profile and normalized beta has been designed using feedback linearization and Lyapunov redesign techniques. Feedback linearization avoids approximate linearization of the plasma dynamics, retaining the original physics content of the model. Moreover, the use of Lyapunov redesign techniques makes the controller robust against the uncertainties arising during the modeling process. The controller’s performance in the presence of unknown dynamics is tested in nonlinear, one-dimensional simulations using the Control Oriented Transport SIMulator (COTSIM) code, which employs plasma models that are significantly more complex than those employed for control synthesis.

Physics↗

Simultaneous access to high normalized density, current, pressure, and confinement in strongly-shaped diverted negative triangularity plasmas

Strongly-shaped diverted negative triangularity (NT) plasmas in the DIII-D tokamak demonstrate simultaneous access to high normalized density, current, pressure, and confinement. NT plasmas are shown to exist across an expansive parameter space compatible with high fusion power production, revealing surprisingly good core stability properties that compare favorably to conventional positive triangularity plasmas in DIII-D. Non-dimensionalizing the key parameters, expanded operating spaces featuring edge safety factors below 3, normalized betas above 3, Greenwald density fractions above 1, and high-confinement mode (H-mode) confinement qualities above 1 are observed, even simultaneously, and all with a robustly stable edge free from deleterious edge-localized mode instabilities. Scaling of the confinement time with engineering parameters reveals at least a linear dependence on plasma current although with significant power degradation, both in excess of expected H-mode scalings. These results increase confidence that NT plasmas are a viable approach to realize fusion power and open directions for future detailed study.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Background studies and normalization of signal events in the Mu2e experiment

The Mu2e experiment is currently being constructed at Fermilab to search for the neutrino-less conversion of negative muons into electrons in the field of an aluminum nucleus. The experiment aims at a sensitivity of four orders of magnitude higher than previous related experiments, which implies highly demanding accuracy requirements both in the design and during the operation. To achieve such a goal, two important tasks should be accomplished. First, it is essential to estimate precisely the particle yields and all the backgrounds that could mimic the monoenergetic conversion electron signal. Second, it is necessary to normalize the signal events accurately. The normalization of the signal events is planned to be done using a detector system made of an HPGe detector and a Lanthanum Bromide detector, which will measure the rate of muons stopped on the aluminum target by looking at the emitted characteristic X-and γ-rays of energies up to 1809 keV. Therefore, it is essential to evaluate the detector system's performance before the start of the actual experiment. In this study, the first task was addressed by an extensive campaign of Monte Carlo simulations to investigate the relevant parameters and their impact on the experiment's sensitivity. The second task was handled by taking advantage of the Helmholtz-Zentrum Dresden-Rossendorf (HZDR) pulsed Bremsstrahlung photon beam at the ELBE facility. The detector system was tested at the ELBE facility under timing and background conditions similar to the ones expected at the Mu2e experiment. The study presents and discusses the simulation results and the detector system testing campaign.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

HELIUM-IGNITED VIOLENT MERGERS AS A UNIFIED MODEL FOR NORMAL AND RAPIDLY DECLINING TYPE Ia SUPERNOVAE

The progenitors of Type Ia supernovae (SNe Ia) are still unknown, despite significant progress during the past several years in theory and observations. Violent mergers of two carbon-oxygen (CO) white dwarfs (WDs) are a candidate scenario suggested to be responsible for at least a significant fraction of normal SNe Ia. Here, we simulate the merger of two CO WDs using a moving-mesh code that allows for the inclusion of thin helium (He) shells (0.01 M{sub Sun }) on top of the WDs at an unprecedented numerical resolution. The accretion of He onto the primary WD leads to the formation of a detonation in its He shell. This detonation propagates around the CO WD and sends a converging shock wave into its core, known to robustly trigger a second detonation, as in the well-known double-detonation scenario for He-accreting CO WDs. However, in contrast to that scenario where a massive He shell is required to form a detonation through thermal instability, here the He detonation is ignited dynamically. Accordingly the required He-shell mass is significantly smaller, and hence its burning products are unlikely to affect the optical display of the explosion. We show that this scenario, which works for CO primary WDs with CO- as well as He-WD companions, has the potential to explain the different brightness distributions, delay times, and relative rates of normal and fast declining SNe Ia. Finally, we discuss extensions to our unified merger model needed to obtain a comprehensive picture of the full observed diversity of SNe Ia.

79 ASTRONOMY AND ASTROPHYSICS↗

MISALIGNMENT OF THE JET AND THE NORMAL TO THE DUSTY TORUS IN THE BROAD ABSORPTION LINE QSO FIRST J155633.8+351758

We performed Very Long Baseline Array observations of the broad absorption line quasar FIRST J155633.8+351758, ''the first radio loud BALQSO''. Our observations at 15.3 GHz partially resolved a secondary component at position angle (P.A.) Almost-Equal-To 35 Degree-Sign . We combine this determination of the radio jet projection on the sky plane, with the constraint that the jet is viewed within 14. Degree-Sign 3 of the line of sight (as implied by the high variability brightness temperature) and with the P.A. of the optical/UV continuum polarization in order to study the quasar geometry. Within the context of the standard model, the data indicates a ''dusty torus'' (scattering surface) with a symmetry axis tilted relative to the accretion disk normal and a polar broad absorption line outflow aligned with the accretion disk normal. We compare this geometry to that indicated by the higher resolution radio data, brightness temperature, and optical/UV continuum polarization P.A. of a similar high optical polarization BALQSO, Mrk 231. A qualitatively similar geometry is found in these two polar BALQSOs; the continuum polarization is determined primarily by the tilt of the dusty torus.

79 ASTRONOMY AND ASTROPHYSICS↗

SALPETER NORMALIZATION OF THE STELLAR INITIAL MASS FUNCTION FOR MASSIVE GALAXIES AT z ∼ 1

The stellar initial mass function (IMF) is a key parameter for studying galaxy evolution. Here we measure the IMF mass normalization for a sample of 68 field galaxies in the redshift range 0.7-0.9 within the Extended Groth Strip. To do this we derive the total (stellar + dark matter) mass-to-light [(M/L)] ratio using axisymmetric dynamical models. Within the region where we have kinematics (about one half-light radius), the models assume (1) that mass follows light, implying negligible differences between the slope of the stellar and total density profiles, (2) constant velocity anisotropy (β{sub z}≡1−σ{sub z}{sup 2}/σ{sub R}{sup 2}=0.2), and (3) that galaxies are seen at the average inclination for random orientations (i.e., i = 60°, where i = 90° represents edge-on). The dynamical models are based on anisotropic Jeans equations, constrained by Hubble Space Telescope/Advanced Camera for Surveys imaging and the central velocity dispersion of the galaxies, extracted from good-quality spectra taken by the DEEP2 survey. The population (M/L) are derived from full-spectrum fitting of the same spectra with a grid of simple stellar population models. Recent dynamical modeling results from the ATLAS{sup 3D} project and numerical simulations of galaxy evolution indicate that the dark matter fraction within the central regions of our galaxies should be small. This suggests that our derived total (M/L) should closely approximate the stellar M/L. Our comparison of the dynamical (M/L) and the population (M/L) then implies that for galaxies with stellar mass M {sub *} ≳ 10{sup 11} M {sub ☉}, the average normalization of the IMF is consistent with a Salpeter slope, with a substantial scatter. This is similar to what is found within a similar mass range for nearby galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Translation and rotation equivariant normalizing flow (TRENF) for optimal cosmological analysis

ABSTRACT Our Universe is homogeneous and isotropic, and its perturbations obey translation and rotation symmetry. In this work, we develop translation and rotation equivariant normalizing flow (TRENF), a generative normalizing flow (NF) model which explicitly incorporates these symmetries, defining the data likelihood via a sequence of Fourier space-based convolutions and pixel-wise non-linear transforms. TRENF gives direct access to the high dimensional data likelihood p(x|y) as a function of the labels y, such as cosmological parameters. In contrast to traditional analyses based on summary statistics, the NF approach has no loss of information since it preserves the full dimensionality of the data. On Gaussian random fields, the TRENF likelihood agrees well with the analytical expression and saturates the Fisher information content in the labels y. On non-linear cosmological overdensity fields from N-body simulations, TRENF leads to significant improvements in constraining power over the standard power spectrum summary statistic. TRENF is also a generative model of the data, and we show that TRENF samples agree well with the N-body simulations it trained on, and that the inverse mapping of the data agrees well with a Gaussian white noise both visually and on various summary statistics: when this is perfectly achieved the resulting p(x|y) likelihood analysis becomes optimal. Finally, we develop a generalization of this model that can handle effects that break the symmetry of the data, such as the survey mask, which enables likelihood analysis on data without periodic boundaries.

79 ASTRONOMY AND ASTROPHYSICS↗