Reconstructing the Stripping History of the Sagittarius Stream with Neural Networks astropy ( 8 ; 6 ; 7 ) , Agama ( 86 ) , pytorch ( 72 )
Jian Zhang, Cuihua Du, Mingji Deng, Zhongcheng Li, Haoyang Liu · 2026
Read Reconstructing the Stripping History of the Sagittarius Stream with Neural Networks astropy ( 8 ; 6 ; 7 ) , Agama ( 86 ) , pytorch ( 72 ) by Jian Zhang, Cuihua Du, Mingji Deng, Zhongcheng Li, Haoyang Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-14308.txt.
Introduction
Zhang and four co-authors train a neural-network method on N-body simulations for stripping times of stream stars from phase-space coordinates. The stream comes from a dwarf's disruption, with wraps stripped during pericentric passages. SEGUE, APOGEE and LAMOST spectroscopy plus Gaia astrometry and StarHorse distances identify members. The method finds a metallicity gradient with stripping time, linear with slope near a third of a dex per Gyr. M 54, Terzan 7, Terzan 8 and Arp 2 look bound to it, while Pal 12, Whiting 1 and NGC 2419 were stripped about 1, 1, and 2 Gyr ago.
Catalog record
| CREATOR | Jian Zhang, Cuihua Du, Mingji Deng, Zhongcheng Li, Haoyang Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39796 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.14308
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Reconstructing the Stripping History of the Sagittarius Stream with Neural Networks astropy ( 8 ; 6 ; 7 ) , Agama ( 86 ) , pytorch ( 72 ) Jian Zhang, Cuihua Du, Mingji Deng, Zhongcheng Li, Haoyang Liu Abstract The Sagittarius (Sgr) Stream is produced by the ongoing disruption of the Sgr dwarf spheroidal (dSph) galaxy and is thought to contain multiple wraps that were stripped during different pericentric passages. In this study, we introduce a neural‐network–based method trained on N-body simulations to infer the stripping time of Sgr Stream stars directly from their phase-space coordinates. We combine spectroscopic data from SEGUE, APOGEE DR17, and LAMOST DR7 LRS with Gaia EDR3 astrometry and distance estimates from the latest StarHorse catalog to identify high-quality Sgr Stream members. Applying our method to these stars, we measure a clear metallicity gradient with stripping time, well described by a linear relation with slope ∼ 0.3 dex Gyr − 1 \sim 0.3~\mathrm{dex~Gyr^{-1}} . We further predict the stripping times of globular clusters previously suggested to originate from the Sgr dSph. M 54, Terzan 7, Terzan 8, and Arp 2 exhibit stripping times consistent with being currently bound to the Sgr remnant. Pal 12, Whiting 1, and NGC 2419 are inferred to have been stripped 0.9 ± 0.1 0.9\pm 0.1 , 1.1 ± 0.2 1.1\pm 0.2 , and 2.1 ± 0.2 2.1\pm 0.2 Gyr ago, respectively. For NGC 4147 and NGC 5634, whose membership in the Sgr system remains uncertain, our analysis suggests stripping times of 1.1 ± 0.4 1.1\pm 0.4 and 1.1 ± 0.1 1.1\pm 0.1 Gyr, respectively, if they are ultimately confirmed as genuine Sgr members. These results demonstrate that data-driven models of dynamical stripping histories offer a promising approach for reconstructing the formation and chemical evolution of the Sgr Stream. show][email protected] I Introduction The \Lambda CDM paradigm predicts that large-scale structures grow hierarchically through the continual accretion and merging of smaller systems ( 89 ; 80 ) . This is clearly reflected in the Milky Way, whose stellar halo contains numerous substructures, such as the Helmi Streams ( 38 ) , Gaia–Sausage–Enceladus ( 12 ; 37 ) , the Sgr Stream ( 42 ; 56 ) , Thamnos ( 49 ) , Sequoia ( 66 ) , and Arjuna, I’itoi, and Aleph ( 67 ) . Among these, the Sgr Stream is particularly noteworthy because it represents an ongoing merger: the tidal disruption of the Sgr dSph, the Milky Way’s nearest and third most massive satellite. Studying the Sgr Stream, and its interaction with the Milky Way, provides a unique opportunity to investigate the detailed physics of how galactic mergers unfold. Stellar streams represent the remnants of stars stripped from their progenitors by the tidal field of the Milky Way. They manifest as narrow, coherent structures on the sky. Streams born from globular clusters are typically thin and dynamically cold, whereas those arising from dwarf galaxies are broader and dynamically hotter ( 16 ) . The Sgr Stream is the archetype of the latter type, produced by the ongoing tidal disruption of the Sgr dSph. The Sgr dSph was first recognized by 42 as an overdensity in both position and velocity, and subsequent studies further mapped its outskirts and extended structure ( 3 ; 60 ; 61 ) . Since then, numerous studies have used a variety of stellar tracers—including RR Lyrae stars, blue horizontal branch stars, M giants, and main-sequence turnoff stars—together with techniques such as matched color–magnitude diagram (CMD) filtering and great-circle cell counts (GC3) to trace the tidal debris of the Sgr Stream across the sky ( 44 ; 56 ; 68 ; 14 ; 69 ; 20 ; 23 ; 13 ; 41 , e.g.,) . In parallel, dynamical modeling efforts, ranging from early N-body simulations ( 39 ; 40 ; 46 ; 50 ; 26 ; 73 ) to the landmark study of 51 , have progressively improved our understanding of the formation and evolution of the Sgr Stream. The Gaia mission has provided precise astrometric measurements for more than a billion stars in the Milky Way ( 30 ; 29 ; 27 ; 28 ) , fundamentally transforming our understanding of Galactic structure. For the Sgr Stream in particular, Gaia has enabled the blind identification of \sim 10^{5} high-probability members based primarily on proper motions ( 4 ; 43 ; 75 ) . These data have significantly improved our knowledge of the Sgr Stream’s present-day kinematics and enabled the construction of comprehensive dynamical models of the Sgr dSph and the Sgr Stream ( 85 ; 71 ; 88 ) . The chemical properties of stars in the Sgr dSph and the Sgr Stream provide important clues to the nature of the progenitor dwarf galaxy. Numerous studies have shown that stars in the Sgr Stream ([Fe/H] \sim-1 ) are, on average, more metal-poor than those in the bound core of the Sgr dSph ([Fe/H] \sim-0.4 ) ( 11 ; 18 ; 63 ; 17 ) . Metallicity gradients have been detected both within the Sgr dSph core ( 55 ; 64 ; 87 ) and along the Sgr Stream arms ( 23 ; 24 ; 22 ; 65 ) , suggesting that the Sgr progenitor possessed a radial metallicity gradient prior to its tidal disruption. However, measuring a metallicity gradient along the Sgr Stream is complicated by the overlap of multiple tidal wraps. Older wraps, stripped early from the progenitor’s outer regions, are expected to be more metal-poor, whereas younger wraps, removed from deeper layers, should be more metal-rich. The superposition of these wraps obscures any simple metallicity–distance or metallicity–longitude trend. The multi-wrap nature of the Sgr Stream has been substantiated in several recent studies. 33 used SEGUE ( 93 ) spectroscopy to identify two chemically and kinematically distinct sub-populations in both the leading and trailing arms. 45 , using the H3 Survey ( 19 ) , found a kinematically diffuse, metal-poor population that they attributed to the stellar halo of the Sgr progenitor. Most recently, 53 combined SEGUE spectroscopy with Gaia EDR3 astrometry to show that simulated particles stripped more than 2 Gyr ago exhibit present-day phase-space properties similar to low-metallicity Sgr Stream stars, whereas those stripped within the last 2 Gyr more closely resemble metal-rich ([Fe/H] > –1) Sgr Stream members. In this work, we address the challenge posed by overlapping tidal wraps by introducing a neural-network method, trained on N-body simulations, to infer the stripping time of Sgr Stream stars directly from their phase-space coordinates. This paper is organized as follows. Section II describes the observational datasets and the selection criteria used to identify Sgr Stream members. Section III introduces our neural-network framework for mapping phase-space coordinates to stripping time. Our results on the metallicity gradient are presented in Section IV . Section V discusses the implications and limitations of our findings, and Section VI summarizes our main conclusions. II Data II.1 SEGUE/APOGEE/LAMOST + Gaia and StarHorse This study extends the analysis presented in 53 (hereafter L23). L23 combined spectroscopic data from SEGUE, which is well suited for studies of distant halo substructures ( 92 ; 13 ; 33 ; 91 ) , with astrometry from Gaia EDR3 ( 28 ) .