Draft version October 3, 2019 Typeset using LATEX twocolumn style in AASTeX62 Variations in the width, density, and direction of the Palomar 5 tidal tails Ana Bonaca,1 Sarah Pearson,2 Adrian M. Price-Whelan,2, 3 Arjun Dey,4 Marla Geha,5 Nitya Kallivayalil,6 John Moustakas,7 Ricardo Muñoz,8 Adam D. Myers,9 David J. Schlegel,10 and Francisco Valdes4 1Center for Astrophysics | Harvard & Smithsonian, 60 Garden Street, Cambridge, MA 02138, USA 2Center for Computational Astrophysics, Flatiron Institute, 162 Fifth Avenue, NY 10010, USA 3Department of Astrophysical Sciences, Princeton University, 4 Ivy Lane, Princeton, NJ 08544, USA 4National Optical Astronomy Observatory, 950 North Cherry Avenue, Tucson, AZ 85719, USA 5Department of Astronomy, Yale University, New Haven, CT 06520, USA 6Department of Astronomy, University of Virginia, 530 McCormick Road, Charlottesville, VA 22904, USA 7Department of Physics & Astronomy, Siena College, NY 12211, USA 8Departamento de Astronomia, Universidad de Chile, Camino del Observatorio 1515, Las Condes, Santiago, Chile 9Department of Physics and Astronomy, University of Wyoming, Laramie, WY 82071, USA 10Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA ABSTRACT Stars that escape globular clusters form tidal tails that are predominantly shaped by the global distri- bution of mass in the Galaxy, but also preserve a historical record of small-scale perturbations. Using deep grz photometry from DECaLS, we present highly probable members of the tidal tails associated with the disrupting globular cluster Palomar 5. These data yield the cleanest view of a stellar stream beyond ∼ 20 kpc and reveal: (1) a wide, low surface-brightness extension of the leading tail; (2) signif- icant density variations along the stream; and (3) sharp changes in the direction of both the leading and the trailing tail. In the fiducial Milky Way model, a rotating bar perturbs the Palomar 5 tails and can produce streams with similar width and density profiles to those observed. However, the deviations of the stream track in this simple model do not match those observed in the Palomar 5 trailing tail, indicating the need for an additional source of perturbation. These discoveries open up the possibility of measuring the population of perturbers in the Milky Way, including dark-matter subhalos, with an ensemble of stellar streams and deep photometry alone. Keywords: Galaxy: halo — dark matter — Galaxy: kinematics and dynamics 1. INTRODUCTION halo (e.g., Koposov et al. 2010; Küpper et al. 2015; Bovy Direct N-body simulations of globular clusters or- et al. 2016). biting in a static galactic potential predict that the However, streams are long-lived and witness the host clusters continually lose stars through evaporation and galaxy evolve, including its gradual increase in mass, its tidal stripping (e.g., Baumgardt & Makino 2003). Stars rotating bar, and orbiting dark matter subhalos. When escape the cluster with a small relative velocity and simulated in more realistic environments that feature some of these events, the resulting streams are no longer arXiv:1910.00592v1 [astro-ph.GA] 1 Oct 2019 thus form thin, kinematically cold streams (e.g., Combes et al. 1999). As a result, globular cluster streams are ex- thin, coherent structures (e.g., Bonaca et al. 2014; Ngan cellent tracers of the underlying tidal field, and under et al. 2015; Price-Whelan et al. 2016b). Recently, Price- the assumption of a static potential, they constrain the Whelan & Bonaca(2018) detected gaps and off-the- enclosed mass within their current location (Bonaca & stream features in the GD-1 stellar stream that could be Hogg 2018). Nearby stellar streams have already been signatures of perturbation (Bonaca et al. 2019b). This used to measure the mass and shape of the Milky Way discovery establishes stellar streams as a cosmological probe of dark matter on small scales. However, streams can also be affected by baryonic perturbers, such as gi- [email protected] ant molecular clouds (Amorisco et al. 2016), the Galac- Corresponding author: Ana Bonaca tic bar (Pearson et al. 2017), and spiral arms (Banik & Bovy 2019). Streams can even naturally develop fea- 2 bonaca et al. tures in their density profile during cluster disruption To select likely Pal 5 stars, we first queried the (e.g., Küpper et al. 2008; Just et al. 2009). Additionally, DECaLS DR8 sweep catalogs2 for point sources. The stream debris can spread out rapidly in phase space if catalog was constructed using The Tractor forward- their progenitors are evolving on non-regular orbits (e.g., modeling code for source extraction3 and a source was Pearson et al. 2015; Fardal et al. 2015; Price-Whelan classified as ‘PSF’ if the PSF model was preferred to the et al. 2016a). To infer the abundance of dark-matter round exponential model used to represent galaxies4. We subhalos from stream perturbations, we need to confirm removed spurious sources by requiring allmask_g==0, the origin of stream perturbations first. allmask_r==0 and brightstarinblob==0. With this In this paper, we revisit the tidal tails of the Palo- clean stellar catalog, we identified stars on Pal 5’s main mar 5 (Pal 5) globular cluster (Odenkirchen et al. 2001; sequence in the dereddened color-magnitude diagram Rockosi et al. 2002). The Pal 5 stream features density (using the re-calibrated SFD dust map; Schlegel et al. variations not reproduced in a static model of the Milky 1998; Schlafly & Finkbeiner 2011). Specifically, we se- Way, but neither their significance nor their origin have lected stars following an 11.5 Gyr MIST isochrone with been established (Carlberg et al. 2012; Bernard et al. [Fe=H] = −1:3 (Choi et al. 2016) between 20 < g < 23:7. 2016; Ibata et al. 2016; Erkal et al. 2017). Arguably, In the top panel of Figure1 we present the sky distri- Pal 5 provides the best opportunity for disentangling bution of likely Pal 5 main sequence stars. The (φ1; φ2) different mechanisms that shape the stream as it has coordinate frame is oriented along the great circle that a surviving progenitor. Such an endeavor, however, re- best-fits the Pal 5 stream, while keeping the cluster at ◦ ◦ quires a robust map of the entire tidal debris. Pal 5 is the origin (φ1 = 0 ; φ2 = 0 ) and its motion in the di- located too far from the Sun to enable efficient mem- rection of positive φ1 (Price-Whelan 2017). There is a bership selection based on Gaia proper motions, while distance gradient along the Pal 5 stream (Ibata et al. accurate mapping using the existing photometry is lim- 2016), so to increase its contrast against the field Milky ited because the catalogs are either wide, but shallow Way stars, we applied the isochrone selection at two dis- ◦ ◦ (Bernard et al. 2016), or deep, but narrow (Ibata et al. tances: 23 kpc for φ1 <= 0 and 19 kpc for φ1 > 0 . 2016). To confidently identify Pal 5 members over a wide With our map, Pal 5 is continuously detected between ◦ ◦ area, we use deep, wide-field photometry from the DE- φ1 = −15 and 7 . Cam Legacy Survey (§2). In §3 we use the updated map The color-magnitude diagrams (CMDs) of stars in of Pal 5 to quantify how the stream track, width and different regions of the Pal 5 stream are shown in the density vary along the stream. We then explore how bottom of Figure1. Each panel contains stars from a 3◦ these properties change across Pal 5 models simulated long and 0:8◦ wide area marked in the top of Figure1 in a range of Galactic potentials (§4) and conclude with (see red boxes). The Pal 5 main-sequence turn-off at a discussion of perturbers that jointly could have caused g ∼ 20:5 stands out in all fields, but the depth to which the observed Pal 5 features (§5). the main sequence is detected varies from g ∼ 24 close to the cluster to g ∼ 22 in the leading tail. The non-uniform 2. DATA detection depth is partly due to variable photometric We study the Palomar 5 system in the photometric depth (the coverage is shallower in the leading tail), and catalog of DECam Legacy Survey (DECaLS, part of the partly due to contamination from the Sagittarius stellar DESI Legacy Imaging Surveys, Dey et al. 2019)1. The ◦ stream (main-sequence turn-off at g ∼ 22 for φ1 & −5 ) survey was designed to provide deep grz imaging at high and faint galaxies (g & 23). Despite these challenges, galactic latitudes, with the targeted 5 σ depth of g = 24, the Pal 5 main-sequence is evident even in field 9 (9◦ < ◦ r = 23:4, and z = 22:5. In addition to data obtained as a φ1 < 12 ), beyond the apparent leading tail truncation ◦ part of the survey, DECaLS also includes imaging from at φ1 ∼ 7 . This indicates that Pal 5 tails may be longer publicly available DECam data in the survey footprint. than previously thought (Bernard et al. 2016). We conducted a targeted survey of Pal 5 whose data To improve our selection of Pal 5 members, we first products are now a part of DECaLS (NOAO Proposal employ the z-band to distinguish between faint stars and ID nos. 2014A-0321, PI: Geha; 2014A-0611, PI: Munoz; galaxies more efficiently. In DECam filters, the g−z color 2015A-0620, PI: Bonaca). The median (minimum) 5 σ of stars bluer than g − r . 1:2 is approximately linear PSF depth in the Pal 5 region is g = 25:6(25:3), r = 25:1(24:8), z = 24:1(23:4), which makes DECaLS the deepest and largest-area survey of Pal 5. 2 http://legacysurvey.org/dr8/files/ #sweep-catalogs-region-sweep 3 https://github.com/dstndstn/tractor 4 1 see also www.legacysurvey.org https://github.com/legacysurvey/legacypipe pal 5’s biggest fan 3 4 9 1 8 ] 2 2 7 g 3 6 e 4 5 d [ 0 2 -2 -4 -20 -15 -10 -5 0 5 10 15 1 [deg] 16 1 2 3 DM = 0.10 DM = 0.08 DM = 0.06 18 d = 1.1 kpc d = 0.9 kpc d = 0.6 kpc ] g a m [ 20 0 g 22 24 16 4 5 6 DM = 0.04 DM = 0.02 DM = 0.00 18 d = 0.4 kpc d = 0.2 kpc d = 0.0 kpc ] g a m [ 20 0 g 22 24 16 7 8 9 DM = -0.14 DM = -0.37 DM = -0.59 18 d = -1.5 kpc d = -3.6 kpc d = -5.5 kpc ] g a m [ 20 0 g 22 24 -0.5 0.0 0.5 1.0-0.5 0.0 0.5 1.0-0.5 0.0 0.5 1.0 (g - r)0 [mag] (g - r)0 [mag] (g - r)0 [mag] Figure 1.