A tidally disrupting dwarf galaxy in the halo of NGC 253
Elisa Toloba, David Sand, Kristine Spekkens, Denija Crnojevic, Joshua Simon, Puragra Guhathakurta, Jay Strader, Nelson Caldwell, Brian McLeod, Anil Seth
aa r X i v : . [ a s t r o - ph . GA ] D ec Draft version September 17, 2018
Preprint typeset using L A TEX style emulateapj v. 5/2/11
A TIDALLY DISRUPTING DWARF GALAXY IN THE HALO OF NGC 253
Elisa Toloba , David J. Sand , Kristine Spekkens , Denija Crnojevi´c , Joshua D. Simon ,Puragra Guhathakurta , Jay Strader , Nelson Caldwell , Brian McLeod , and Anil C. Seth Texas Tech University, Physics Department, Box 41051, Lubbock, TX 79409-1051, USA UCO/Lick Observatory, University of California, Santa Cruz, 1156 High Street, Santa Cruz, CA 95064, USA Department of Physics, Royal Military College of Canada, P.O. Box 17000, Station Forces, Kingston, Ontario, K7L 7B4, Canada Carnegie Observatories, 813 Santa Barbara Street, Pasadena, CA 91101, USA Department of Physics and Astronomy, Michigan State University, East Lansing, MI 48824, USA Harvard-Smithsonian Center for Astrophysics, Cambridge, MA 02138, USA and Department of Physics and Astronomy, University of Utah, Salt Lake City, UT 84112, USA
Draft version September 17, 2018
ABSTRACTWe report the discovery of Scl-MM-Dw2, a new dwarf galaxy at a projected separation of ∼
50 kpcfrom NGC 253, as part of the PISCeS (Panoramic Imaging Survey of Centaurus and Sculptor) project.We measure a tip of the red giant branch distance of 3 . ± .
30 Mpc, suggesting that Scl-MM-Dw2is likely a satellite of NGC 253. We qualitatively compare the distribution of red giant branch (RGB)stars in the color-magnitude diagram with theoretical isochrones and find that it is consistent withan old, ∼
12 Gyr, and metal poor, − . < [Fe/H] < − .
1, stellar population. We also detect a smallnumber of asymptotic giant branch stars consistent with a metal poor 2 − M HI /L V < . M ⊙ /L ⊙ . The stellar and gaseous properties of Scl-MM-Dw2 suggestthat it is a dwarf spheroidal galaxy. Scl-MM-Dw2 has a luminosity of M V = − . ± . r h = 2 . ± .
46 kpc which makes it moderately larger than dwarf galaxies inthe Local Group of the same luminosity. However, Scl-MM-Dw2 is very elongated ( ǫ = 0 . ± . µ ,V = 26 . ± . − ). Its elongation anddiffuseness make it an outlier in the ellipticity-luminosity and surface brightness-luminosity scalingrelations. These properties suggest that this dwarf is being tidally disrupted by NGC 253. Subject headings: galaxies: individual (NGC 253) — galaxies: dwarf — galaxies: stellar content —galaxies: halos — galaxies: photometry INTRODUCTION
The Λ+Cold Dark Matter (ΛCDM) model for struc-ture formation is very successful on large scales ( & M V ≈−
10. Likewise, a search for dwarf companions tothe dwarf galaxy NGC 3109 has yielded at least one newsatellite with M V = − ∼
50 kpc in pro-jection from NGC 253, which we dub Scl-MM-Dw2, in [email protected] accordance with our prior work in the Sculptor group.The data were taken as part of the Panoramic ImagingSurvey of Centaurus and Sculptor (PISCeS; Sand et al.2014; Crnojevi´c et al. 2014), which aims to study thefaint satellites and stellar halo substructure of NGC 253and NGC 5128 in resolved stellar light. Due to its highellipticity and faint hints of other nearby substructure,Scl-MM-Dw2 is likely being disrupted, allowing for a di-rect view of the continuing buildup of NGC 253’s halo. InSection 2, we describe our data and reduction methods,along with the procedure followed to discover the dwarfgalaxy. In Section 3, we measure the physical propertiesof the new dwarf, estimate its distance, and study its lo-cation in the scaling relations of other dwarf galaxies. InSection 4, we discuss our findings, summarize the mainproperties of Scl-MM-Dw2, and conclude.For reference, 1 ′ = 1 . OBSERVATIONS AND DISCOVERY OF SCL-MM-DW2
The data presented here are part of the largerPISCeS survey, which utilizes the Megacam instrument(McLeod et al. 2015) located at the f / ′ × ′ field of view anda binned pixel scale of 0 . ′′
16. Here we focus on the fourMegacam fields in the vicinity of Scl-MM-Dw2. Figure 1shows their location in an extended Digitized Sky Surveyimage centered on NGC 253. Toloba et al.
RA (deg) −27−26−25−24−23 D e c ( d e g ) NE Fig. 1.—
DSS image centered on NGC 253. The blue circleindicates the planned extent of the PISCeS survey (R=150 kpc).The purple box indicates the region discussed here. The green starindicates the position of Scl-MM-Dw2.
PISCeS typically observes each field for 6 ×
300 s in the g and r band to achieve image depths of ∼ . ′′
65 and 1 . ′′
0. Initial data reduc-tion, which consists of image detrending, astrometricmatching, and stacking of the multiple individual ex-posures, is performed by the Smithsonian AstrophysicalObservatory Telescope Data Center using the pipelinedesigned for Megacam by M. Conroy, J. Roll, and B.McLeod. We perform point spread function (PSF) fit-ting photometry on the final stacked images using theDAOPHOT and ALLSTAR software suite (Stetson 1987,1994), matching the g and r band catalogs with DAO-MATCH/DAOMASTER (Stetson 1993). The deepestcatalogs possible are obtained by running ALLFRAME(Stetson 1994) to perform simultaneous photometry forobjects detected in the individual g and r band stacks.We culled our catalogs of outliers in χ versus magni-tude, magnitude error versus magnitude, and sharpnessversus magnitude space to remove objects that were notpoint sources.To flux calibrate the data we employ a dual approach.First, we observe equatorial fields from the Sloan Digi-tal Sky Survey (SDSS; Alam et al. 2015) at varying air-masses on photometric nights. We transform instru-mental magnitudes measured in our Megacam imagesinto SDSS magnitudes with our derived zeropoints, colorterms and extinction coefficients derived from the SDSSimaging. For nights that are not photometric, we takeadvantage of the tile overlaps between Megacam point-ings (typically ∼ g − r ∼ .
5, unresolved background galaxies ( g − r ∼ . Hubble Space Telescope (HST) Cycle23 observations may shed light on this (PID: 14259; PICrnojevi´c). PROPERTIES OF SCL-MM-DW2
Stellar Populations and Distance
We study the stellar content of the newly discoveredScl-MM-Dw2 by superposing old, metal-poor isochrones(Marigo et al. 2008; Girardi et al. 2010, hereafter Padovaisochrones) over the CMD of the galaxy. Figure 2 showsthe density CMD (a.k.a. Hess diagram) of an ellipticalarea within one half light radius (see § r band luminosity function tolook for this transition and restrict the color range to0 . < g − r < . r = 24 . ± .
18. Using M T RGBr = − . ± .
10 (as determined for the SDSS r band in Sand et al. 2014) we obtain a distance mod-ulus for Scl-MM-Dw2 of m − M = 27 . ± .
21. Thisdistance is consistent with that of NGC 253 at the ∼ σ level (assuming a distance modulus of 27.70 ± −1.0 −0.5 0.0 0.5 1.0 1.5 2.0 2.5 g-r r −1.0 −0.5 0.0 0.5 1.0 1.5 2.0 2.5 g-r −1.0 −0.5 0.0 0.5 1.0 1.5 2.0 2.5 g-r s t a r s / p i x e l Fig. 2.—
De-reddened color-magnitude Hess diagrams (density CMD). Left panel: CMD showing the stars within the r h of the newlydiscovered dwarf galaxy. Middle panel: CMD showing the average stars within two background field regions, one in each of the Megacamfields where the galaxy lands, with the same area as in the left panel to account for possible small scale structure and radial variationsin NGC 253’s halo. Right panel: CMD showing the difference between the left and middle panels. Statistically, this CMD shows onlystars within the newly discovered galaxy. The orange and red lines indicate the Padova isochrones for 12 Gyr and [Fe/H] = − . − . We use this distance modulus to place the theoreti-cal Padova isochrones at the distance of Scl-MM-Dw2.Figure 2 shows two isochrones that straddle the RGBridgeline stars seen in the CMD, which correspond to anage of 12 Gyr and metallicities of [Fe/H] = − . − . . ± .
04 within the half-light radius, r h , and0 . ± .
02 within 2 r h ). This is consistent with a centralyoung (2 − − . < [Fe/H] < − .
1) stellar population superposed on the overall old( ∼
12 Gyr) and metal poor population. We also es-timate the total stellar mass of the galaxy assuming astellar mass-to-light ratio of 1.6 (Kirby et al. 2013, seeTable 1). The stellar mass obtained is consistent withthe value estimated by comparing the number of brightRGB stars with a modelled population of 10 M ⊙ . HI gas limits
We investigated the possibility of HI gas associatedwith Scl-MM-Dw2. Indeed, the HI Parkes All Sky Surveyspectrum (HIPASS; Barnes et al. 2001) exhibits a ∼ . σ HI emission peak along the line of sight to Scl-MM-Dw2with a heliocentric radial velocity of ∼
275 km s − sug-gesting a possible association with the dwarf. Seeking toconfirm this tentative detection, we therefore obtainedmuch deeper position-switched HI observations using Di-rector’s Discretionary time (program AGBT 15B 349) onthe Green Bank Telescope (GBT; PI: Spekkens). The re-sulting spectrum has an rms noise σ GBT = 0 .
75 mJy ata spectral resolution of 10 km s − , and is over an or-der of magnitude more sensitive than the corresponding HIPASS spectrum.We do not find any HI emission at −
700 km s − ≤ V hel ≤ −
100 km s − and 100 km s − ≤ V hel ≤ − , whereas the range −
100 km s − ≤ V hel ≤
100 km s − is contaminated by Milky Way HI disk emis-sion. The faint HIPASS feature is therefore a statisticalfluke or an artifact of the bright HI emission from thenearby NGC 253 (Lucero et al. 2015). Our non-detectionimplies that unless the radial velocity of Scl-MM-Dw2puts it in the contaminated region of our GBT spectrum,any HI counterpart to Scl-MM-Dw2 has a 5 σ , 15 km s − HI mass upper limit of M limHI = 1 . × M ⊙ , and thus agas fraction of M HI /L V < . M ⊙ /L ⊙ . Scl-MM-Dw2is therefore gas-poor, similar to other dwarf spheroidalgalaxies in the Local Volume (e.g. Grcevich & Putman2009; Spekkens et al. 2014). Structure
Figure 3 shows the spatial distribution of candidateRGB and AGB stars. To the Northwest of Scl-MM-Dw2,with coordinates close to RA ∼ . ◦ and Dec ∼ − . ◦ there is a second ellipse away from the dwarf’s main bodythat highlights a possible faint overdensity of stars. Thispossible overdensity coincides with the overlap region be-tween different Megacam fields, and is a noisier area.Deeper photometry is necessary to confirm if these starsare true RGB stars and if they are related to Scl-MM-Dw2. The candidate RGB stellar map clearly shows thatScl-MM-Dw2 is a very elongated galaxy that points to-wards the halo of NGC 253, which appears in the South-west of the map.We measure the structural parameters of Scl-MM-Dw2using the maximum likelihood technique described byMartin et al. (2008) and following the implementationof Sand et al. (2012). We select all candidate RGB stars(utilizing the purple selection box in Figure 2) withina spatial box of ∼ ′ × ′ approximately centered onScl-MM-Dw2. We fit the candidate RGB stars with anexponential profile plus a constant background with the Toloba et al. RA (deg) −25.1−25.0−24.9−24.8−24.7−24.6−24.5−24.4 D e c ( d e g ) Candidate RGB stars 02468101214161820 s t a r s / p i x e l RA (deg) −24.85−24.80−24.75−24.70−24.65 D e c ( d e g ) Candidate RGB and AGB stars
Fig. 3.—
Candidate RGB and AGB stars maps. These candidate stars are selected using the purple and blue boxes shown in Figure 2.Left panel: the black dashed ellipse in the overdensity region of RGB stars indicates the half-light radius and ellipticity of Scl-MM-Dw2.Theblack dashed ellipse to the Northwest of the galaxy indicates the position of a possible small overdensity of RGB stars. Right panel: zoom-inof Scl-MM-Dw2, with the spatial distribution of candidate RGB stars in black, and AGB stars in red. The ellipse is the same as in the leftpanel. The blue dashed line extends the major axis of the red ellipse to 2.5 r h . While Scl-MM-Dw2 is very prominent in the candidateRGB stars it appears to have only a few candidate AGB stars centrally concentrated. This suggests that the bulk of the stellar populationis metal poor and old with a younger component in the center. following free parameters: the central coordinates of thegalaxy (RA ,DEC ), position angle (PA), ellipticity ( ǫ ),half-light radius ( r h ), and the background surface den-sity (Σ b ). The uncertainties in these parameters are es-timated using 1000 bootstrap resamples of the data, andwe report the 68% confidence intervals. The measuredstructural parameters are presented in Table 1.The surface brightness within the half light radius h µ h,V i is calculated from the total luminosity (see be-low, § r h , and ellipticity. The central surfacebrightness, µ ,V , is calculated using the conversion from h µ h,V i for an exponential profile from Graham & Driver(2005). Scl-MM-Dw2 is large, elongated, and verydiffuse, with r h ∼ ǫ = 0 .
66, and h µ h,V i =28 . − . The measured ellipticity is similar tothe disrupting Sagittarius dwarf around the Milky Way(MW; McConnachie 2012). Along with the orientationand proximity of Scl-MM-Dw2 to NGC 253, this sug-gests that it too is being tidally disrupted, which we willdiscuss further in § Luminosity
We estimate the total luminosity of Scl-MM-Dw2 viadirect aperture photometry using an elliptical aperturewith semimajor axis equal to the r h of the galaxy, semi-minor axis equal to r h × (1 − ǫ ), and the appropriateposition angle. We estimate the flux within this aper-ture and subtract the estimated background by placingthe same elliptical aperture randomly at different posi-tions in the image. The resulting flux corresponds to theflux within the r h of Scl-MM-Dw2, which we then multi-ply by a factor of two to account for the total flux of thegalaxy. The uncertainty was calculated based on the di-rect Poisson uncertainty in the Scl-MM-Dw2 aperture, aswell as the scatter in measurements from the backgroundapertures.The final apparent magnitudes are r = 15 . ± . g = 15 . ± .
5; after applying our measured distance mod-ulus and the filter transformations of Jester et al. (2005)for stars with R − I < .
15, we derive M V = − . ± . TABLE 1Properties of Scl-MM-Dw2
Parameter ValueRA (hh:mm:ss) 00:50:17.06 ± . ′′ DEC (dd:mm:ss) -24:44:58.58 ± . ′′ m − M (mag) 27 . ± . D (Mpc) 3 . ± . M V (mag) − . ± . r h (arcmin) 3 . ± . r h (kpc) 2 . ± . ǫ . ± . . ± . µ ,V (mag arcsec − ) 26 . ± . h µ h,V i (mag arcsec − ) 28 . ± . ∗ (M ⊙ ) 1 . ± . × M HI (M ⊙ ) < . × M HI /L V ( M ⊙ /L ⊙ ) < . Note . — RA and DEC indicate the central coordinates ofthe galaxy in J2000. M V is measured in the AB system. µ ,V and h µ h,V i are the central and effective surface brightness. Theseparameters are described in Section 3.3. (see Table 1). Scaling Relations
Figure 4 shows the size-luminosity, ellipticity-luminosity, and surface brightness-luminosity scalingrelations for Scl-MM-Dw2 in comparison with thedSphs in the Local Group (McConnachie 2012;Martin et al. 2013a,b; The DES Collaboration et al.2015; Kim & Jerjen 2015; Kim et al. 2015; Laevens et al.2015b,a; Martin et al. 2015), our other 3 recent dis-coveries in the PISCeS survey (Sand et al. 2014;Crnojevi´c et al. 2014), and the ultra diffuse galaxiesfound in the Virgo and Coma clusters (Mihos et al.2015; van Dokkum et al. 2015).In the size-luminosity relation, Scl-MM-Dw2 is a slight tidally disrupting dwarf galaxy in the halo of NGC 253 5outlier, being larger than most dwarf galaxies for itsluminosity. Meanwhile, it is amongst the most elon-gated faint dwarfs, with ǫ =0.66 ± ∼ − lower than Local Groupdwarfs at a similar luminosity. Its very low surfacebrightness is only comparable to the slightly brighter ul-tra diffuse galaxies recently found in the Virgo cluster(Mihos et al. 2015). DISCUSSION AND CONCLUSIONS
We report the discovery of Scl-MM-Dw2, a dwarfgalaxy with a luminosity of M V = − . ∼ ′ ( ∼
50 kpc) in projection from NGC 253’s cen-ter ( ∼ . × r NGC253 h ; de Vaucouleurs et al. 1991). Weuse the TRGB method to find a distance modulus of m − M = 27 . ± .
21 which places this galaxy at3 . ± .
30 Mpc, consistent with that of NGC 253 withinthe uncertainty (Radburn-Smith et al. 2011). Thus, it islikely that Scl-MM-Dw2 is a satellite of NGC 253.We superpose Padova isochrones on the color-magnitude diagram centered in the galaxy and find thatits stellar distribution can be explained with an old, ∼
12 Gyr, and metal poor, − . < [Fe/H] < − .
1, stel-lar population. In addition, we detect some AGB stars inthe center of the galaxy that are consistent with a popu-lation of 2 − − . < [Fe/H] − .
1. Thus, thisdwarf has a central metal poor intermediate-age popula-tion superposed on the overall old stellar population. Wedo not detect any HI gas in emission in our GBT spec-trum which constrains the gas mass to an upper limitof M limHI = 1 . × M ⊙ and therefore a gas fraction M HI /L V < . M ⊙ /L ⊙ . These properties place thisnew galaxy in the category of gas poor dwarf spheroidalgalaxy.We estimate the structural parameters of Scl-MM-Dw2and compare them with those of other dSphs in the Lo-cal Group as well as from our own PISCeS program, and with the ultra diffuse galaxies found in the Virgo andComa clusters. We find that Scl-MM-Dw2 is an outlierin all scaling relations. Its size is slightly larger than ex-pected and it is extremely elongated and diffuse for its lu-minosity. The only other outlier on these relations is theSagittarius dwarf galaxy in the Milky Way halo. How-ever, while Sagittarius has a central star cluster (M54;Sarajedini & Layden 1995), Scl-MM-Dw2 does not showany nucleus. Moreover, the major axis of Scl-MM-Dw2points towards the center of NGC 253 suggesting that itmay have undergone a very close pericenter passage. Inaddition, we detect an overdensity of RGB stars on theNorthwest and Southeast of the galaxy, which could bea tidal tail of Scl-MM-Dw2.Klimentowski et al. (2009) simulated a dwarf galaxyunder the gravitational influence of a Milky Way analogand find that two kinds of tidal tails form: (1) the densesttidal tails, formed in the vicinity of the dwarf, are ori-ented towards the center of the host galaxy; and (2) themore diffuse tidal tails, formed on a much larger spatialscale and less likely to be detected due to their extremelylow luminosity, are oriented along the orbit of the dwarfgalaxy. Sand et al. (2012) investigate the dependence ofthe angle between all Milky Way dSphs and the GalacticCenter and find no clear correlation between their ori-entation and their elongation. Similar results have beenfound for M31 dSphs (Salomon et al. 2015).The angle between the major axis of Scl-MM-Dw2 andthe center of NGC 253 is only ∼ ◦ . This measure-ment is affected by projection effects, thus, the orienta-tion of the galaxy alone is not enough evidence of thisprocess. However, all these properties summarized aboveseem to point towards Scl-MM-Dw2 being disrupted byNGC 253.E.T., D.J.S, J.D.S, and P.G. acknowledge the NSFgrant AST-1412504. E.T. and P.G. are also supportedby NSF grant AST-1010039 and D.J.S. by AST-1517649.K.S. acknowledges support from the Natural Sciencesand Engineering Research Council of Canada (NSERC). REFERENCESAlam, S., Albareti, F. D., Allende Prieto, C., et al. 2015, ApJS,219, 12Barnes, D. G., Staveley-Smith, L., de Blok, W. J. G., et al. 2001,MNRAS, 322, 486Chiboucas, K., Jacobs, B. A., Tully, R. B., & Karachentsev, I. D.2013, AJ, 146, 126Chiboucas, K., Karachentsev, I. D., & Tully, R. B. 2009, AJ, 137,3009Cooper, A. P., Cole, S., Frenk, C. S., et al. 2010, MNRAS, 406,744Crnojevi´c, D., Sand, D. J., Caldwell, N., et al. 2014, ApJ, 795,L35de Vaucouleurs, G., de Vaucouleurs, A., Corwin, Jr., H. G., et al.1991, Third Reference Catalogue of Bright Galaxies. Volume I:Explanations and references. Volume II: Data for galaxiesbetween 0 h and 12 h . Volume III: Data for galaxies between 12 h and 24 h .Girardi, L., Williams, B. F., Gilbert, K. M., et al. 2010, ApJ, 724,1030Graham, A. W. & Driver, S. P. 2005, PASP, 22, 118Grcevich, J. & Putman, M. E. 2009, ApJ, 696, 385Jester, S., Schneider, D. P., Richards, G. T., et al. 2005, AJ, 130,873 Johnston, K. V., Bullock, J. S., Sharma, S., et al. 2008, ApJ, 689,936Kim, D. & Jerjen, H. 2015, ApJ, 808, L39Kim, D., Jerjen, H., Mackey, D., Da Costa, G. S., & Milone,A. P. 2015, ApJ, 804, L44Kirby, E. N., Cohen, J. G., Guhathakurta, P., et al. 2013, ApJ,779, 102Klimentowski, J., Lokas, E. L., Kazantzidis, S., et al. 2009,MNRAS, 400, 2162Laevens, B. P. M., Martin, N. F., Bernard, E. J., et al. 2015a,ApJ, 813, 44Laevens, B. P. M., Martin, N. F., Ibata, R. A., et al. 2015b, ApJ,802, L18Lee, M. G., Freedman, W. L., & Madore, B. F. 1993, ApJ, 417,553Lucero, D. M., Carignan, C., Elson, E. C., et al. 2015, MNRAS,450, 3935Marigo, P., Girardi, L., Bressan, A., et al. 2008, A&A, 482, 883Martin, N. F., de Jong, J. T. A., & Rix, H.-W. 2008, ApJ, 684,1075Martin, N. F., Nidever, D. L., Besla, G., et al. 2015, ApJ, 804, L5Martin, N. F., Schlafly, E. F., Slater, C. T., et al. 2013a, ApJ,779, L10 Toloba et al.
Log(r h ) (pc) −15−10−50 M V ( m a g ) SgrHercUMaIUMaII
MW dwarfsM31 dwarfsComa UDGsVirgo UDGsCe A dwarfsScl-MM-Dw1Scl-MM-Dw2 −15−10−50 M V (mag) ǫ SgrHercUMaIUMaII −15−10−50 M V (mag) µ , V ( m a g a r c s e c − ) SgrHercUMaIUMaII