Overview of the Umbrella Galaxy NGC 4651 with its surrounding stellar substructures. This colour composite image was produced by combining luminance-filter data from the 0.5 m BlackBird Remote Observatory telescope with g ′ , r ′ , and [O iii ] images from 

Context in source publication

Context 1

... classically pictured as detached “island universes”, galaxies are now appreciated as actively connected to their environments, and growing continuously through the infall of smaller galaxies. This view is supported both by discoveries of stellar substructures in their outer regions or “haloes” (e.g. Searle & Zinn 1978; Malin et al. 1983) and by the paradigm of hierarchical assembly in a cold dark matter (CDM) cosmology (e.g. White & Rees 1978; Cooper et al. 2011). Consequently, the study of halo substructure has become a major industry (e.g. Ibata et al. 2001b; Johnston et al. 2008; Martell & Grebel 2010; Helmi et al. 2011; Mouhcine et al. 2011; Xue et al. 2011; Bate et al. 2014), both as a route to understanding the assembly histories of galaxies, and as a proving ground for CDM via quantitative tests of substructure predictions. These tests to date have turned up a number of prob- lems that are variations of the longstanding small-scale substructure “crisis” (Moore 1994; Klypin et al. 1999; see re- view in Weinberg et al. 2014). These include deficits of satellites around the Milky Way and other galaxies, and shal- low dark matter cores rather than central cusps in galaxies over a wide range of luminosities (e.g. de Blok et al. 2008; Herrmann & Ciardullo 2009; Agnello & Evans 2012; Newman et al. 2013a). These puzzles might be resolved through better modelling of baryon physics, or by recourse to CDM alternatives – but distinguishing between these two general options could be an insurmountable obstacle for analyses that focus only on the central regions of galaxies. A different, and arguably more robust, line of inquiry is to study the large-scale galaxy properties that are minimally affected by baryon physics and so may be more uniquely compared to theoretical predictions. These properties include the orbits and total (virial) masses of satellites. Recent work along these lines include comparisons of the orbital coherence of satellite groups to predictions for cosmological infall (e.g. Pawlowski et al. 2012; Wang et al. 2013), and searches for the impact signatures of small dark matter sub- haloes on visible stellar streams (Carlberg 2013). The investigations of substructure have so far been dominated by studies of the two main galaxies in the Local Group (M31 and the Milky Way), but it is critically important to extend them to more distant galaxies – both to build up better statistics, and to understand how the properties of substructure depend systematically on galaxy type and environment. Although progress has been made along these lines (see Atkinson et al. 2013 and references therein), it has been almost exclusively based on photometry, while missing the dimension of velocity that is obtained directly through spectroscopy. Such information is indis- pensable for inferring the three-dimensional (3D) anatomy of substructures, the timescales of orbital decay, the progenitor properties including total mass, and the dynamical interplay between satellite and host galaxies. While these aspects can sometimes be probed through gas dynamics (e.g. Iodice et al. 2003), most infalling satellites are gas poor, in a reflection of the morphology–density rela- tion that is probably caused by quenching during the accretion process (e.g. Einasto et al. 1974; Mayer et al. 2006; Grcevich & Putman 2009; Geha et al. 2012; Slater & Bell 2013; Sánchez-Janssen et al. 2013). The paucity of gas in halo substructures motivates pursuing stellar-light spectroscopy, where the challenge is the extremely low surface brightnesses of typically ∼ 27 mag arcsec − 2 or fainter. An alternative in such cases is to exploit bright, discrete tracer objects: globular clusters (GCs) and planetary nebulae (PNe). This approach is now used extensively for chemodynamical mapping of the faint outer regions of early-type galaxies (e.g. Coccato et al. 2009; Usher et al. 2012; Pota et al. 2013), but so far has had very limited application to visible substructures. Pioneering examples include PN- and GC-based spectroscopic studies of tidal debris around M51 (Durrell et al. 2003), of an outer halo stream in M87 (Romanowsky et al. 2012), of substructures around M31 (Veljanoski et al. 2013), and of a tidal stream linking NGC 4365 and NGC 4342 (Blom et al. 2014). Other substructures have been found serendipitously in large kinematic datasets (Merrett et al. 2003; McNeil et al. 2010; Shih & Méndez 2010; Ventimiglia et al. 2011; Cortesi et al. 2011), but such cases are hard to model without photometric counterparts for reference. The time is ripe for a systematic spectroscopic survey of distant halo substructures, if it can be demonstrated that the observations are feasible, and that the subsequent dynamical modelling and interpretation is tractable. To this end, here we present a pilot study of the ‘Umbrella Galaxy’, NGC 4651, using multi wavelength photometry, spectroscopy of GCs, PNe, and H ii regions, and preliminary dynamical modelling of its substructure. NGC 4651 is a spiral galaxy (Sc) that hosts one of the brightest and most spectacular systems of halo substructures in the local Universe, as seen in Fig. 1. A dramatic umbrella-like feature extends far beyond the disc to the East side of the galaxy, and consists of a straight, col- limated stream (henceforth, stick) terminating in a sharp, perpendicular plume (henceforth, shell). On the West side there are several broad plume- and shell-like features, whose morphologies are more difficult to distinguish owing to their lower surface brightnesses and overlapping locations. The umbrella feature was first reported by Zwicky (1956, where it was interpreted as a dwarf companion), and was discussed in various subsequent catalogues including Arp (1966). The main galaxy disc appears relatively undis- turbed, with a normal star formation rate (Kennicutt 1998). The outer disc shows mild kinematic disturbances in ionized gas (Rubin et al. 1999) and a very lopsided H i distribution, with a tail-like feature to the West that is coincident with a broad optical plume, and no detectable gas to the East in the umbrella region (Chung et al. 2009). In a prelude to the current paper, Mart ́ ınez-Delgado et al. (2010) acquired deep optical imaging of NGC 4651 as part of a survey of eight nearby Milky Way-like spiral galaxies with suspected halo substructures, using a small robotic telescope. They described the morphology of the umbrella and counter-shell, and noted similarities to simulations of stellar halo accretion events from Johnston et al. (2008). They inferred that the observed features could plausibly have formed in the tidal disruption of a low-mass galaxy on a near radial orbit around 6 to 10 Gyr ago. The initial goals of our follow-up project are to quantify the luminosity, colour, and stellar mass of the substructure; to measure its kinematical properties; to derive an orbital model that accounts for the essential features of the observations; and to interpret the results in the context of cosmological accretion predictions. Various properties of NGC 4651 are summarized in Table 1. Of special note is the distance, which was not well- established in the literature, and which we estimate as 19 Mpc based on the planetary nebula luminosity function (see Section 3.1). The implied angular scale is 92 pc arcsec − 1 . This paper is organized as follows: in Section 2, we present the new imaging and spectroscopic data. Section 3 presents our analysis, while dynamical modelling is un- dertaken in Section 4. Some implications are discussed in Section 5, and our conclusions can be found in Section ...