Search for: All records

We demonstrate the potential of the state-of-the-art OpenAI GPT-4 large language model to engage in meaningful interactions with Astronomy papers using in-context prompting. To optimize for efficiency, we employ a distillation technique that effectively reduces the size of the original input paper by 50%, while maintaining the paragraph structure and overall semantic integrity. We then explore the model’s responses using a multi-document context (ten distilled documents). Our findings indicate that GPT-4 excels in the multi-document domain, providing detailed answers contextualized within the framework of related research findings. Our results showcase the potential of large language models for the astronomical community, offering a promising avenue for further exploration, particularly the possibility of utilizing the models for hypothesis generation.

Large-scale surveys will provide spectroscopy for ∼50 million resolved stars in the Milky Way and Local Group. However, these data will have a high degree of heterogeneity and most will be low-resolution (R< 10,000), posing challenges to measuring consistent and reliable stellar labels. Here, we introduce a framework for identifying and remedying these issues. By simultaneously fitting the full spectrum and Gaia photometry withthe Payne, we measure ∼30 abundances for eight metal-poor red giants in M15. From degraded quality Keck/HIRES spectra, we evaluate trends with resolution and signal-to-noise ratio (S/N) and find that (i) ∼20 abundances are recovered consistently within ≲0.1 dex agreement and with ≲0.05–0.15 dex systematic uncertainties from 10,000 ≲R≲ 80,000; (ii) for nine elements (C, Mg, Ca, Sc, Ti, Fe, Ni, Y, and Nd), this systematic precision and accuracy extends down toR∼ 2500; and (iii) while most elements do not exhibit strong S/N-dependent systematics, there are nonnegligible biases for four elements (C, Mg, Ca, and Dy) below S/N ∼ 10 pixel⁻¹. We compare statistical uncertainties from Markov Chain Monte Carlo sampling to the easier-to-compute Cramér–Rao bounds and find that they agree for ∼85% of elements, indicating the latter to be a reliable and faster way tomore »estimate uncertainties. Our analysis illustrates the great promise of low-resolution spectroscopy for stellar chemical abundance work in the low-metallicity regime, and ongoing improvements to stellar models (e.g., 3D-NLTE physics) will only further extend its viability to more stars, more elements, and higher precision and accuracy.

« less

Carbon-enhanced metal-poor (CEMP) stars comprise almost a third of stars with [Fe/H] < −2, although their origins are still poorly understood. It is highly likely that one sub-class (CEMP-s stars) is tied to mass-transfer events in binary stars, while another sub-class (CEMP-no stars) are enriched by the nucleosynthetic yields of the first generations of stars. Previous studies of CEMP stars have primarily concentrated on the Galactic halo, but more recently they have also been detected in the thick disc and bulge components of the Milky Way. Gaia DR3 has provided an unprecedented sample of over 200 million low-resolution (R ≈ 50) spectra from the BP and RP photometers. Training on the CEMP catalogue from the SDSS/SEGUE database, we use XGBoost to identify the largest all-sky sample of CEMP candidate stars to date. In total, we find 58 872 CEMP star candidates, with an estimated contamination rate of 12 per cent. When comparing to literature high-resolution catalogues, we positively identify 60–68 per cent of the CEMP stars in the data, validating our results and indicating a high completeness rate. Our final catalogue of CEMP candidates spans from the inner to outer Milky Way, with distances as close as r ∼ 0.8 kpc from the Galacticmore »centre, and as far as r > 30 kpc. Future higher resolution spectroscopic follow-up of these candidates will provide validations of their classification and enable investigations of the frequency of CEMP-s and CEMP-no stars throughout the Galaxy, to further constrain the nature of their progenitors.

« less

The majority of the Milky Way’s stellar halo consists of debris from our galaxy’s last major merger, the Gaia-Sausage-Enceladus (GSE). In the past few years, stars from the GSE have been kinematically and chemically studied in the inner 30 kpc of our galaxy. However, simulations predict that accreted debris could lie at greater distances, forming substructures in the outer halo. Here we derive metallicities and distances using Gaia DR3 XP spectra for an all-sky sample of luminous red giant stars, and map the outer halo with kinematics and metallicities out to 100 kpc. We obtain follow-up spectra of stars in two strong overdensities—including the previously identified outer Virgo Overdensity—and find them to be relatively metal rich and on predominantly retrograde orbits, matching predictions from simulations of the GSE merger. We argue that these are apocentric shells of GSE debris, forming 60–90 kpc counterparts to the 15–20 kpc shells that are known to dominate the inner stellar halo. Extending our search across the sky with literature radial velocities, we find evidence for a coherent stream of retrograde stars encircling the Milky Way from 50 to 100 kpc, in the same plane as the Sagittarius Stream but moving in the oppositemore »direction. These are the first discoveries of distant and structured imprints from the GSE merger, cementing the picture of an inclined and retrograde collision that built up our galaxy’s stellar halo.

« less

Abstract Modern Galactic surveys have revealed an ancient merger that dominates the stellar halo of our galaxy (Gaia–Sausage–Enceladus, GSE). Using chemical abundances and kinematics from the H3 Survey, we identify 5559 halo stars from this merger in the radial range r Gal = 6–60kpc. We forward model the full selection function of H3 to infer the density profile of this accreted component of the stellar halo. We consider a general ellipsoid with principal axes allowed to rotate with respect to the galactocentric axes, coupled with a multiply broken power law. The best-fit model is a triaxial ellipsoid (axes ratios 10:8:7) tilted 25° above the Galactic plane toward the Sun and a doubly broken power law with breaking radii at 12 kpc and 28 kpc. The doubly broken power law resolves a long-standing dichotomy in literature values of the halo breaking radius, being at either ∼15 kpc or ∼30 kpc assuming a singly broken power law. N -body simulations suggest that the breaking radii are connected to apocenter pile-ups of stellar orbits, and so the observed double-break provides new insight into the initial conditions and evolution of the GSE merger. Furthermore, the tilt and triaxiality of the stellar halo could implymore »that a fraction of the underlying dark matter halo is also tilted and triaxial. This has important implications for dynamical mass modeling of the galaxy as well as direct dark matter detection experiments.« less

Abstract Some studies of stars’ multielement abundance distributions suggest at least 5–7 significant dimensions, but others show that many elemental abundances can be predicted to high accuracy from [Fe/H] and [Mg/Fe] (or [Fe/H] and age) alone. We show that both propositions can be, and are, simultaneously true. We adopt a machine-learning technique known as normalizing flow to reconstruct the probability distribution of Milky Way disk stars in the space of 15 elemental abundances measured by APOGEE. Conditioning on T eff and log g minimizes the differential systematics. After further conditioning on [Fe/H] and [Mg/Fe], the residual scatter for most abundances is σ [ X /H] ≲ 0.02 dex, consistent with APOGEE’s reported statistical uncertainties of ∼0.01–0.015 dex and intrinsic scatter of 0.01–0.02 dex. Despite the small scatter, residual abundances display clear correlations between elements, which we show are too large to be explained by measurement uncertainties or by the finite sampling noise. We must condition on at least seven elements to reduce the correlations to a level consistent with the observational uncertainties. Our results demonstrate that cross-element correlations are a much more sensitive probe of a hidden structure than dispersion, and they can be measured precisely in a large samplemore »even if the star-by-star measurement noise is comparable to the intrinsic scatter. We conclude that many elements have an independent story to tell, even for the mundane disk stars and elements produced by the core-collapse and Type Ia supernovae. The only way to learn these lessons is to measure the abundances directly, and not merely infer them.« less

We report the discovery of Specter, a disrupted ultrafaint dwarf galaxy revealed by the H3 Spectroscopic Survey. We detected this structure via a pair of comoving metal-poor stars at a distance of 12.5 kpc, and further characterized it with Gaia astrometry and follow-up spectroscopy. Specter is a 25° × 1° stream of stars that is entirely invisible until strict kinematic cuts are applied to remove the Galactic foreground. The spectroscopic members suggest a stellar ageτ≳ 12 Gyr and a mean metallicity$〈[\mathrm{Fe}/\mathrm{H}]〉=-{1.84}_{-0.18}^{+0.16}$, with a significant intrinsic metallicity dispersion${\sigma }_{[\mathrm{Fe}/\mathrm{H}]}={0.37}_{-0.13}^{+0.21}$. We therefore argue that Specter is the disrupted remnant of an ancient dwarf galaxy. With an integrated luminosityM_V≈ −2.6, Specter is by far the least-luminous dwarf galaxy stream known. We estimate that dozens of similar streams are lurking below the detection threshold of current search techniques, and conclude that spectroscopic surveys offer a novel means to identify extremely low surface brightness structures.

Abstract The Sloan Digital Sky Survey (SDSS) has recently initiated its fifth survey generation (SDSS-V), with a central focus on stellar spectroscopy. In particular, SDSS-V's Milky Way Mapper program will deliver multiepoch optical and near-infrared spectra for more than 5 × 10 6 stars across the entire sky, covering a large range in stellar mass, surface temperature, evolutionary stage, and age. About 10% of those spectra will be of hot stars of OBAF spectral types, for whose analysis no established survey pipelines exist. Here we present the spectral analysis algorithm, ZETA-PAYNE, developed specifically to obtain stellar labels from SDSS-V spectra of stars with these spectral types and drawing on machine-learning tools. We provide details of the algorithm training, its test on artificial spectra, and its validation on two control samples of real stars. Analysis with ZETA-PAYNE leads to only modest internal uncertainties in the near-IR with APOGEE (optical with BOSS): 3%–10% (1%–2%) for T eff , 5%–30% (5%–25%) for v sin i , 1.7–6.3 km s −1 (0.7–2.2 km s −1 ) for radial velocity, <0.1 dex (<0.05 dex) for log g , and 0.4–0.5 dex (0.1 dex) for [M/H] of the star, respectively. We find a good agreement betweenmore »atmospheric parameters of OBAF-type stars when inferred from their high- and low-resolution optical spectra. For most stellar labels, the APOGEE spectra are (far) less informative than the BOSS spectra of these stars, while log g , v sin i , and [M/H] are in most cases too uncertain for meaningful astrophysical interpretation. This makes BOSS low-resolution optical spectra better for stellar labels of OBAF-type stars, unless the latter are subject to high levels of extinction.« less

Abstract The astrophysical origins of r -process elements remain elusive. Neutron star mergers (NSMs) and special classes of core-collapse supernovae (rCCSNe) are leading candidates. Due to these channels’ distinct characteristic timescales (rCCSNe: prompt, NSMs: delayed), measuring r -process enrichment in galaxies of similar mass but differing star formation durations might prove informative. Two recently discovered disrupted dwarfs in the Milky Way’s stellar halo, Kraken and Gaia-Sausage Enceladus (GSE), afford precisely this opportunity: Both have M ⋆ ≈ 10 8 M ⊙ but differing star formation durations of ≈2 Gyr and ≈3.6 Gyr. Here we present R ≈ 50,000 Magellan/MIKE spectroscopy for 31 stars from these systems, detecting the r -process element Eu in all stars. Stars from both systems have similar [Mg/H] ≈ −1, but Kraken has a median [Eu/Mg] ≈ −0.1 while GSE has an elevated [Eu/Mg] ≈ 0.2. With simple models, we argue NSM enrichment must be delayed by 500–1000 Myr to produce this difference. rCCSNe must also contribute, especially at early epochs, otherwise stars formed during the delay period would be Eu free. In this picture, rCCSNe account for ≈50% of the Eu in Kraken, ≈25% in GSE, and ≈15% in dwarfs with extended star formation durationsmore »like Sagittarius. The inferred delay time for NSM enrichment is 10×–100× longer than merger delay times from stellar population synthesis—this is not necessarily surprising because the enrichment delay includes time taken for NSM ejecta to be incorporated into subsequent generations of stars. For example, this may be due to natal kicks that result in r -enriched material deposited far from star-forming gas, which then takes ≈10 8 –10 9 yr to cool in these galaxies.« less