Search

The exponential growth of astronomical literature poses significant challenges for researchers navigating and synthesizing general insights or even domain-specific knowledge. We present pathfinder , a machine learning framework designed to enable literature review and knowledge discovery in astronomy, focusing on semantic searching with natural language instead of syntactic searches with keywords. Utilizing state-of-the-art large language models (LLMs) and a corpus of 385,166 peer-reviewed papers from the Astrophysics Data System, pathfinder offers an innovative approach to scientific inquiry and literature exploration. Our framework couples advanced retrieval techniques with LLM-based synthesis to search astronomical literature by semantic context as a complement to currently existing methods that use keywords or citation graphs. It addresses complexities of jargon, named entities, and temporal aspects through time-based and citation-based weighting schemes. We demonstrate the tool’s versatility through case studies, showcasing its application in various research scenarios. The system’s performance is evaluated using custom benchmarks, including single-paper and multipaper tasks. Beyond literature review, pathfinder offers unique capabilities for reformatting answers in ways that are accessible to various audiences (e.g., in a different language or as simplified text), visualizing research landscapes, and tracking the impact of observatories and methodologies. This tool represents a significant advancement in applying artificial intelligence to astronomical research, aiding researchers at all career stages in navigating modern astronomy literature.

Resolving the environments of massive stars is crucial for understanding their formation mechanisms and their impact on galaxy evolution. An important open question is whether massive stars found in diffuse regions outside spiral arms formed in situ or migrated there after forming in denser environments. To address this question, we use multiresolution measurements of extinction in the Andromeda galaxy (M31) to probe the interstellar medium surrounding massive stars across galactic environments. We construct a catalog of 42,107 main-sequence massive star candidates ( M ≥ 8 M _⊙ ) using resolved stellar photometry from the Panchromatic Hubble Andromeda Treasury (PHAT) program, plus stellar and dust model fits from the Bayesian Extinction and Stellar Tool (BEAST). We quantify galactic environments by computing surrounding stellar densities of massive stars using kernel density estimation. We then compare high-resolution line-of-sight extinction estimates from the BEAST with 25 pc resolution dust maps from PHAT, measuring the total column density distribution of extinction. Our key finding is that, although the average total column density of dust increases with the density of massive stars, the average line-of-sight extinction toward massive stars remains constant across all environments. This suggests that massive stars have a uniform amount of dust in their immediate environment, regardless of their location in the galaxy. One possible explanation for these findings is that small molecular clouds are still capable of forming massive stars, even if they are not resolvable at 25 pc. These results indicate that massive stars are forming in the sparse regions of M31, as opposed to migrating there.

The Sun moves through the interstellar medium (ISM) at a velocity of ∼19 pc Myr ^−1 , making the conditions outside the solar system vary with time over millions of years. Today’s solar system is protected from interstellar particles by the heliosphere, the bubble formed by the solar wind as the Sun moves through the ISM, which engulfs the planets. There is geological evidence from ^60 Fe that Earth was in direct contact with the ISM 2–3 and 5–7 million years ago (MYA). Recent work argues that the Sun encountered a massive cold cloud 2 MYA as part of the Local Ribbon of Cold Clouds that shrunk the heliosphere and exposed Earth to the ISM. Here, we consider the effects of the passage of the Sun through the edge of the Local Bubble occurring at ${6.8}_{-0.4}^{+0.5}$ MYA assuming that the Sun encountered a cloud with a density of 900 cm ^−3 . If we consider additional turbulent motion within the cloud due to shocks, the density encountered can be as low as 283 cm ^−3 . Clouds of this density cover a small but nonzero (≲4.6%) fraction of the surface of the Local Bubble, making an encounter plausible. Using a state-of-the art magnetohydrodynamic model, we show that the heliosphere shrank to a scale smaller than Earth’s orbit, thereby exposing Earth to cold dense ISM, consistent with ^60 Fe evidence. The timing of the event matches perturbations observed in the paleoclimate record recovered from deep-sea sediments. The passage through the Local Bubble’s surface and contraction of the heliosphere therefore may have impacted the climate and biosphere significantly, suggesting a new driver of major events in Earth’s history.

Scylla is a deep Hubble Space Telescope (HST) survey of the stellar populations, interstellar medium, and star formation in the LMC and SMC. As a pure-parallel complement to the Ultraviolet Legacy Library of Young Stars as Essential Standards (ULLYSES) survey, Scylla obtained 342 orbits of ultraviolet (UV) through near-IR imaging of the LMC and SMC with Wide Field Camera 3. In this paper, we describe the science objectives, observing strategy, data reduction procedure, and initial results from our photometric analysis of 96 observed fields. Although our observations were constrained by ULLYSES primary exposures, we imaged all fields in at least two filters (F475W and F814W) and 64% of fields in at least three and as many as seven WFC3 filters spanning the UV to IR. Overall, we reach average 50% completeness of m _F225W = 26.0, m _F275W = 26.2, m _F336W = 26.9, m _F475W = 27.8, m _F814W = 25.5, m _F110W = 24.7, and m _F160W = 24.0 Vega mag in our photometric catalogs, which is faintward of the ancient main-sequence turnoff in all filters. The primary science goals of Scylla include characterizing the structure and properties of dust in the MCs, as well as their spatially resolved star formation and chemical enrichment histories. Our images and photometric catalogs, which represent the widest-area coverage of MCs with HST photometry to date, are available as a high-level science product at the Barbara A. Mikulski Archive for Space Telescopes.

We introduce the DESI LOW- Z Secondary Target Survey, which combines the wide-area capabilities of the Dark Energy Spectroscopic Instrument (DESI) with an efficient, low-redshift target selection method. Our selection consists of a set of color and surface brightness cuts, combined with modern machine-learning methods, to target low-redshift dwarf galaxies ( z < 0.03) between 19 < r < 21 with high completeness. We employ a convolutional neural network (CNN) to select high-priority targets. The LOW- Z survey has already obtained over 22,000 redshifts of dwarf galaxies ( M _* < 10 ^9 M _⊙ ), comparable to the number of dwarf galaxies discovered in the Sloan Digital Sky Survey DR8 and GAMA. As a spare fiber survey, LOW- Z currently receives fiber allocation for just ∼50% of its targets. However, we estimate that our selection is highly complete: for galaxies at z < 0.03 within our magnitude limits, we achieve better than 95% completeness with ∼1% efficiency using catalog-level photometric cuts. We also demonstrate that our CNN selections z < 0.03 galaxies from the photometric cuts subsample at least 10 times more efficiently while maintaining high completeness. The full 5 yr DESI program will expand the LOW- Z sample, densely mapping the low-redshift Universe, providing an unprecedented sample of dwarf galaxies, and providing critical information about how to pursue effective and efficient low-redshift surveys.

Contamination from galaxy fragments, identified as sources, is a major issue in large photometric galaxy catalogs. In this paper, we prove that this problem can be easily addressed with computer vision techniques. We use image cutouts to train a convolutional neural network (CNN) to identify cataloged sources that are in reality just star-formation regions and/or shreds of larger galaxies. The CNN reaches an accuracy ∼98% on our testing data sets. We apply this CNN to galaxy catalogs from three among the largest surveys available today: the Sloan Digital Sky Survey, the DESI Legacy Imaging Surveys, and the Panoramic Survey Telescope and Rapid Response System Survey. We find that, even when strict selection criteria are used, all catalogs still show a ∼5% level of contamination from galaxy shreds. Our CNN gives a simple yet effective solution to clean galaxy catalogs from these contaminants.

Diffuse interstellar bands (DIBs) are broad absorption features associated with interstellar dust and can serve as chemical and kinematic tracers. Conventional measurements of DIBs in stellar spectra are complicated by residuals between observations and best-fit stellar models. To overcome this, we simultaneously model the spectrum as a combination of stellar, dust, and residual components, with full posteriors on the joint distribution of the components. This decomposition is obtained by modeling each component as a draw from a high-dimensional Gaussian distribution in the data space (the observed spectrum)—a method we call “Marginalized Analytic Data-space Gaussian Inference for Component Separation” (MADGICS). We use a data-driven prior for the stellar component, which avoids missing stellar features not well modeled by synthetic spectra. This technique provides statistically rigorous uncertainties and detection thresholds, which are required to work in the low signal-to-noise regime that is commonplace for dusty lines of sight. We reprocess all public Gaia DR3 RVS spectra and present an improved 8621 Å DIB catalog, free of detectable stellar line contamination. We constrain the rest-frame wavelength to 8623.14 ± 0.087 Å (vacuum), find no significant evidence for DIBs in the Local Bubble from the 1/6th of RVS spectra that are public, and show unprecedented correlation with kinematic substructure in Galactic CO maps. We validate the catalog, its reported uncertainties, and biases using synthetic injection tests. We believe MADGICS provides a viable path forward for large-scale spectral line measurements in the presence of complex spectral contamination.

We investigate the kinematics of neutral gas in the Small Magellanic Cloud (SMC) and test the hypothesis that it is rotating in a disk. To trace the 3D motions of the neutral gas distribution, we identify a sample of young, massive stars embedded within it. These are stars with radial velocity measurements from spectroscopic surveys and proper motion measurements from Gaia, whose radial velocities match with dominant H i components. We compare the observed radial and tangential velocities of these stars with predictions from the state-of-the-art rotating disk model based on high-resolution 21 cm observations of the SMC from the Australian Square Kilometer Array Pathfinder telescope. We find that the observed kinematics of gas-tracing stars are inconsistent with disk rotation. We conclude that the kinematics of gas in the SMC are more complex than can be inferred from the integrated radial velocity field. As a result of violent tidal interactions with the Large Magellanic Cloud, nonrotational motions are prevalent throughout the SMC, and it is likely composed of distinct substructures overlapping along the line of sight. [ABSTRACT FROM AUTHOR]

We investigate the physical properties of structures seen in channel map observations of 21 cm neutral hydrogen (H i) emission. H i intensity maps display prominent linear structures that are well aligned with the ambient magnetic field in the diffuse interstellar medium (ISM). Some literature hold that these structures are “velocity caustics,” fluctuations imprinted by the turbulent velocity field, and are not three-dimensional density structures in the ISM. We test this hypothesis by stacking probes of the density field—broadband far-infrared (FIR) emission and the integrated H i column density ()—at the locations of linear H i intensity structures. We find that the H i intensity features are real density structures and not velocity caustics. We generalize the investigation to all small-scale structure in H i channel maps and analyze this correlation as a function of velocity channel width, finding no measurable contribution from velocity caustics to the H i channel map emission. Further, we find that small-scale H i channel map structures have elevated FIR/, implying that this emission originates from a colder, denser phase of the ISM than the surrounding material. The data are consistent with a multiphase diffuse ISM in which small-scale structures in narrow H i channel maps are preferentially cold neutral medium (CNM) that is anisotropically distributed and aligned with the local magnetic field. The shallow spatial power spectrum (SPS) of narrow H i channels is often attributed to velocity caustics. We conjecture instead that the small-scale structure and narrow line widths typical of CNM explain the observed relationship between the SPS and channel width. [ABSTRACT FROM AUTHOR]

Every quasar (quasi-stellar object; QSO) spectrum contains absorption-line signatures from the interstellar medium, disk-halo interface, and circumgalactic medium (CGM) of the Milky Way (MW). We analyze Hubble Space Telescope/Cosmic Origins Spectrograph (COS) spectra of 132 QSOs to study the significance and origin of Si iv absorption at km s−1 in the Galactic halo. The gas in the north predominantly falls in at km s−1, whereas in the south, no such pattern is observed. The Si iv column density has an average and a standard deviation of cm−2. At , does not significantly correlate with b, which cannot be explained by a commonly adopted flat-slab geometry. We propose a two-component model to reconstruct the –b distribution: a plane-parallel component to account for the MW’s disk-halo interface and a global component to reproduce the weak dependence on b. We find cm−2 and cm−2 on the basis of Bayesian analyses and block bootstrapping. The global component is most likely to have a Galactic origin, although its exact location is uncertain. If it were associated with the MW’s CGM, we would find for the cool gas at all velocities in the Galactic halo. Our analyses show that there is likely a considerable amount of gas at km s−1 hidden in the MW’s CGM. Along with this work, we make our QSO data set publicly available as the COS Quasar Database for Galactic Absorption Lines (COS-GAL). [ABSTRACT FROM AUTHOR]

The majority of hydrogen in the interstellar medium (ISM) is in atomic form. The transition from atoms to molecules and, in particular, the formation of the H2 molecule, is a key step in cosmic structure formation en route to stars. Quantifying H2 formation in space is difficult, due to the confusion in the emission of atomic hydrogen (H i) and the lack of a H2 signal from the cold ISM. Here we present the discovery of a rare, isolated dark cloud currently undergoing H2 formation, as evidenced by a prominent “ring” of H i self-absorption. Through a combined analysis of H i narrow self-absorption, CO emission, dust emission, and extinction, we directly measured, for the first time, the [H i]/[H2] abundance varying from 2% to 0.2%, within one region. These measured H i abundances are orders of magnitude higher than usually assumed initial conditions for protoplanetary disk models. None of the fast cloud formation model could produce such low atomic hydrogen abundance. We derived a cloud formation timescale of ∼6 × 106 years, consistent with the global Galactic star formation rate, and favoring the classical star formation picture over fast star formation models. Our measurements also help constrain the H2 formation rate, under various ISM conditions. [ABSTRACT FROM AUTHOR]

The local interstellar medium (ISM) is suffused with “dark” gas, identified by excess infrared and gamma-ray emission, yet undetected by standard ISM tracers such as neutral hydrogen (H i) or carbon monoxide emission. Based on observed dust properties from Planck, recent studies have argued that H i mixed with dust is strongly saturated and that dark gas is dominated by optically thick H i. We test this hypothesis by reproducing this model using data from Planck and new 21 cm emission maps from GALFA-H i—the first large-area 21 cm emission survey with comparable angular resolution to Planck. We compare the results with those from a large sample of H i column densities based on direct observations of H i optical depth, and find that the inferred column density corrections are significantly lower than those inferred by the Planck-based model. Further, we rule out the hypothesis that the pencil-beam H i absorption sight lines preferentially miss opaque “blobs” with small covering fraction, as these structures require densities and pressures that are incompatible with ISM conditions. Our results support the picture that excess dust emission in the local ISM is not dominated by optically thick H i, but is rather a combination of intrinsic changes in dust grain emissivities and H2 missed by CO observations. [ABSTRACT FROM AUTHOR]

We present an approach for comparing the detections and non-detections of Local Group (LG) dwarf galaxies in large H i surveys to the predictions of a suite of n-body simulations of the LG. This approach depends primarily on a set of empirical scaling relations to connect the simulations to the observations, rather than making strong theoretical assumptions. We then apply this methodology to the Galactic Arecibo L-band Feed Array Hi (GALFA-HI) Compact Cloud Catalog (CCC), and compare it to the suite Exploring the Local Volume In Simulations (ELVIS) of simulations. This approach reveals a strong tension between the naïve results of the model and the observations: while there are no LG dwarfs in the GALFA-HI CCC, the simulations predict ∼10. Applying a simple model of reionization can resolve this tension by preventing low-mass halos from forming gas. However, and if this effect operates as expected, the observations provide a constraint on the mass scale of the dwarf galaxy that reionization impacts. Combined with the observed properties of Leo T, the halo virial mass scale at which reionization impacts dwarf galaxy gas content is constrained to be , independent of any assumptions about star formation. [ABSTRACT FROM AUTHOR]

We present calibrations for star formation rate (SFR) indicators in the ultraviolet, mid-infrared, and radio-continuum bands, including one of the first direct calibrations of 150 MHz as an SFR indicator. Our calibrations utilize 66 nearby star-forming galaxies with Balmer-decrement-corrected luminosities, which span five orders of magnitude in SFR and have absolute magnitudes of . Most of our photometry and spectrophotometry are measured from the same region of each galaxy, and our spectrophotometry has been validated with SDSS photometry, so our random and systematic errors are small relative to the intrinsic scatter seen in SFR indicator calibrations. We find that the Wide-field Infrared Space Explorer W4 (22.8 μm), Spitzer 24 μm, and 1.4 GHz bands have tight correlations with the Balmer-decrement-corrected Hα luminosity, with a scatter of only 0.2 dex. Our calibrations are comparable to those from the prior literature for L∗ galaxies, but for dwarf galaxies, our calibrations can give SFRs that are far greater than those derived from most previous literature. [ABSTRACT FROM AUTHOR]

We report the detection of a largely ionized very high-velocity cloud (VHVC; km ) toward M33 with the Hubble Space Telescope/Cosmic Origin Spectrograph. The VHVC is detected in O i, C ii, Si ii, and Si iii absorption along five sightlines separated by . On sub-degree scales, the velocities and ionic column densities of the VHVC remain relatively smooth with standard deviations of ±14 km and ±0.15 dex between the sightlines, respectively. The VHVC has a metallicity of [O i/H i] = −0.56 ± 0.17 dex (Z = 0.28 ± 0.11 ). Despite the position–velocity proximity of the VHVC to the ionized Magellanic Stream, the VHVC’s higher metallicity makes it unlikely to be associated with the Stream, highlighting the complex velocity structure of this region of sky. We investigate the VHVC’s possible origin by revisiting its surrounding H i environment. We find that the VHVC may be (1) a MW CGM cloud, (2) related to a nearby H i VHVC—Wright’s Cloud, or (3) connected to M33's northern warp. Furthermore, the VHVC could be a bridge connecting Wright’s Cloud and M33's northern warp, which would make it a Magellanic-like structure in the halo of M33. [ABSTRACT FROM AUTHOR]

Near-infrared spectroscopy from APOGEE and wide-field optical photometry from Pan-STARRS1 have recently made precise measurements of the shape of the extinction curve possible for tens of thousands of stars, parameterized by R(V). These measurements revealed structures in R(V) with large angular scales, which are challenging to explain in existing dust paradigms. In this work, we combine three-dimensional maps of dust column density with R(V) measurements to constrain the three-dimensional distribution of R(V) in the Milky Way. We find that the variations in R(V) are correlated on kiloparsec scales. In particular, most of the dust within one kiloparsec in the outer Galaxy, including many local molecular clouds (Orion, Taurus, Perseus, California, and Cepheus), has a significantly lower R(V) than more distant dust in the Milky Way. These results provide new input to models of dust evolution and processing, and complicate the application of locally derived extinction curves to more distant regions of the Milky Way and to other galaxies. [ABSTRACT FROM AUTHOR]

We report the detection of accreting ionized gas at the disk–halo interface of the nearby galaxy M33. We analyze Hubble Space Telescope/Cosmic Origins Spectrograph absorption-line spectra of seven ultraviolet-bright stars evenly distributed across the disk of M33. We find Si iv absorption components consistently redshifted relative to the bulk M33's ISM absorption along all the sightlines. The Si iv detection indicates an enriched, disk-wide, ionized gas inflow toward the disk. This inflow is most likely multi-phase as the redshifted components can also be observed in ions with lower ionization states (e.g., S ii, P ii, Fe ii, Si ii). Kinematic modeling of the inflow is consistent with an accreting layer at the disk–halo interface of M33, which has an accretion velocity of at a distance of kpc above the disk. The modeling indicates a total mass of ∼3.9 × 107M⊙ for the accreting material at the disk–halo interface on the near side of the M33 disk, with an accretion rate of ∼2.9 M⊙ yr−1. The high accretion rate and the level of metal enrichment suggest the inflow is likely to be the fallback of M33 gas from a galactic fountain and/or the gas pulled loosed during a close interaction between M31 and M33. Our study of M33 is the first to unambiguously reveal the existence of a disk-wide, ionized gas inflow beyond the Milky Way, providing a better understanding of gas accretion in the vicinity of a galaxy disk. [ABSTRACT FROM AUTHOR]

The dust extinction curve is a critical component of many observational programs and an important diagnostic of the physics of the interstellar medium. Here we present new measurements of the dust extinction curve and its variation toward tens of thousands of stars, a hundred-fold larger sample than in existing detailed studies. We use data from the APOGEE spectroscopic survey in combination with ten-band photometry from Pan-STARRS1, the Two Micron All-Sky Survey, and Wide-field Infrared Survey Explorer. We find that the extinction curve in the optical through infrared is well characterized by a one-parameter family of curves described by R(V). The extinction curve is more uniform than suggested in past works, with , and with less than one percent of sight lines having . Our data and analysis have revealed two new aspects of Galactic extinction: first, we find significant, wide-area variations in R(V) throughout the Galactic plane. These variations are on scales much larger than individual molecular clouds, indicating that R(V) variations must trace much more than just grain growth in dense molecular environments. Indeed, we find no correlation between R(V) and dust column density up to . Second, we discover a strong relationship between R(V) and the far-infrared dust emissivity. [ABSTRACT FROM AUTHOR]

Using spectroscopically selected galaxies from the Sloan Digital Sky Survey we present a detection of reddening effects from the circumgalactic medium of galaxies which we attribute to an extended distribution of dust. We detect the mean change in the colors of “standard crayons” correlated with the presence of foreground galaxies at as a function of angular separation. Following Peek & Graves, we create standard crayons using passively evolving galaxies corrected for Milky Way reddening and color-redshift trends, leading to a sample with as little as 2% scatter in color. We devise methods to ameliorate possible systematic effects related to the estimation of colors, and we find an excess reddening induced by foreground galaxies at a level ranging from 10 to 0.5 mmag on scales ranging from 30 kpc to 1 Mpc. We attribute this effect to a large-scale distribution of dust around galaxies similar to the findings of Ménard et al. We find that circumgalactic reddening is a weak function of stellar mass over the range – and note that this behavior appears to be consistent with recent results on the distribution of metals in the gas phase. We also find that circumgalactic reddening has no detectable dependence on the specific star formation rate of the host galaxy. [ABSTRACT FROM AUTHOR]

A correction to the article "The Galfa-Hi Compact Cloud Catalog" that was published in the 2012 issue is presented.

We assess the fraction of the Milky Way (MW)'s circumgalactic medium (CGM) eluding detection due to its velocity being similar to gas in the disk. This is achieved using synthetic observations of the CGM in a simulated MW-mass galaxy that shows similar CGM kinematics to the MW and external L ∼ L galaxies. As viewed by a mock observer at a location similar to the Sun, only 50% (by mass) of the gas moves at high velocity ( or ) in the simulated CGM and would be observable. The low-velocity gas is thermodynamically similar to the high-velocity gas, indicating that the 50% observable fraction is applicable to spectral lines from the radio to the ultraviolet. We apply the observable mass fraction (50%) to current estimates of the MW’s CGM, and find a corrected total mass of for gas below 106 K within ∼15 kpc (this excludes the Magellanic System). This is less than the total mass of the CGM extending out to ∼150 kpc in other L ∼ L galaxies. However, we find similar O vi column densities when the discrepancy in path length between the MW and external galaxies is considered. The coherent spatial and kinematic distribution of low-velocity gas in the simulated CGM suggests that current H i observations of the MW’s CGM may miss large low-velocity H i complexes. In addition, current mass estimates of the MW’s CGM based on high-velocity line observations with distance constraints may miss a non-negligible fraction of gas in the outer halo, which can be obscured if it moves at a velocity similar to the gas in the lower halo. [ABSTRACT FROM AUTHOR]

The Local Leo Cold Cloud (LLCC, at a distance of 11–24 pc) was studied in its relation to the Local Hot Bubble (LHB) and the result suggested that much of the observed keV emission in that direction originates in front of the cloud. This placed a strong constraint on the distribution of X-ray emission within the LHB and called into question the assumption of a uniform distribution of X-ray emitting plasma within the Local Cavity. However, recent work has quantified the contribution of heliospheric solar wind charge exchange (SWCX) emission to the diffuse X-ray background measured by the ROSAT All-Sky Survey (RASS) at keV, and led to the consistency of pressure measurements between the LHB and the local cloud component of the complex of local interstellar clouds (CLICs) surrounding the Sun. In this paper we revisit the LLCC and improve the previous analysis by using higher resolution RASS data, a serendipitous ROSAT pointed observation, a rigorous treatment of the band-averaged X-ray absorption cross section, and models for the heliospheric and magnetospheric SWCX contributions. We find that the foreground emission to the cloud is in excess of the expected heliospheric (interplanetary plus near Earth) SWCX contribution but that it is marginally consistent with the range of possible LHB plasma path lengths between the LLCC and the CLICs given the currently understood plasma emissivity. [ABSTRACT FROM AUTHOR]