The ISLAnds Project III: Variable Stars in Six Andromeda Dwarf Spheroidal Galaxies
Clara E. Martínez-Vázquez, Matteo Monelli, Edouard J. Bernard, Carme Gallart, Peter B. Stetson, Evan D. Skillman, Giuseppe Bono, Santi Cassisi, Giuliana Fiorentino, Kristen B. W. McQuinn, Andrew A. Cole, Alan W. McConnachie, Nicolas F. Martin, Andrew E. Dolphin, Michael Boylan-Kolchin, Antonio Aparicio, Sebastian L. Hidalgo, Daniel R. Weisz
aa r X i v : . [ a s t r o - ph . GA ] O c t Draft version October 26, 2017
Preprint typeset using L A TEX style AASTeX6 v. 1.0
THE ISLANDS PROJECT III: VARIABLE STARS IN SIX ANDROMEDA DWARF SPHEROIDAL GALAXIES ⋆ Clara E. Mart´ınez-V´azquez , Matteo Monelli , Edouard J. Bernard , Carme Gallart , Peter B.Stetson , Evan D. Skillman , Giuseppe Bono , Santi Cassisi , Giuliana Fiorentino , Kristen B. W. McQuinn ,Andrew A. Cole , Alan W. McConnachie , Nicolas F. Martin , Andrew E. Dolphin , MichaelBoylan-Kolchin , Antonio Aparicio , Sebastian L. Hidalgo , Daniel R. Weisz IAC-Instituto de Astrof´ısica de Canarias, Calle V´ıa Lactea s/n, E-38205 La Laguna, Tenerife, Spain Departmento de Astrof´ısica, Universidad de La Laguna, E-38206 La Laguna, Tenerife, Spain INAF-Osservatorio Astronomico di Bologna, Via Gobetti 93/3, I-40129 Bologna, Italy Universit´e Cˆote d’Azur, OCA, CNRS, Lagrange, France Dominion Astrophysical Observatory, Herzberg Institute of Astrophysics, National Research Council, 5071 West Saanich Road, Victoria,British Columbia V9E 2E7, Canada Minnesota Institute for Astrophysics, University of Minnesota, Minneapolis, MN, USA Department of Physics, Universit`a di Roma Tor Vergata, via della Ricerca Scientifica 1, I-00133 Roma, Italy INAF-Osservatorio Astronomico di Roma, via Frascati 33, I-00040 Monte Porzio Catone, Italy INAF-Osservatorio Astronomico di Teramo, Via M. Maggini, I-64100 Teramo Department of Astronomy, The University of Texas at Austin, 2515 Speedway, Stop C1400, Austin, TX 78712-1205, USA School of Physical Sciences, University of Tasmania, Hobart, Tasmania, Australia Observatoire astronomique de Strasbourg, Universit´e de Strasbourg, CNRS, UMR 7550, 11 rue de l’Universit´e, F-67000 Strasbourg, France Max-Planck-Institut f¨ur Astronomie, K¨onigstuhl 17, D-69117 Heidelberg, Germany Raytheon; 1151 E. Hermans Rd., Tucson, AZ 85706, USA Department of Astronomy, University of California Berkeley, Berkeley, CA 94720, USA
ABSTRACTWe present a census of variable stars in six M31 dwarf spheroidal satellites observed with the HubbleSpace Telescope. We detect 870 RR Lyrae (RRL) stars in the fields of And I (296), II (251), III(111), XV (117), XVI (8), XXVIII (87). We also detect a total of 15 Anomalous Cepheids, threeEclipsing Binaries, and seven field RRL stars compatible with being members of the M31 halo orthe Giant Stellar Stream. We derive robust and homogeneous distances to the six galaxies usingdifferent methods based on the properties of the RRL stars. Working with the up-to-date set ofPeriod-Wesenheit ( I , B – I ) relations published by Marconi et al., we obtain distance moduli of µ = [24.49, 24.16, 24.36, 24.42, 23.70, 24.43] mag (respectively), with systematic uncertainties of 0.08mag and statistical uncertainties < Keywords: binaries: eclipsing – galaxies: dwarf — galaxies: individual (And I, And II, And III,And XV, And XVI, And XXVIII) — stars: horizontal-branch — stars: variables: Cepheids— stars: variables: RR Lyrae ⋆ Based on observations made with the NASA/ESA HubbleSpace Telescope, obtained at the Space Telescope Science Insti- tute, which is operated by the Association of Universities for Re-search in Astronomy, Inc., under NASA contract NAS 5-26555.
Mart´ınez-V´azquez et al. INTRODUCTIONRR Lyrae variable stars (RRLs) are unambiguousstellar tracers of an old ( >
10 Gyr) stellar population.As such, they are a fossil record of the early stagesof galaxy evolution. Their pulsational properties andtheir position in the color-magnitude diagram (CMD)– on the horizontal branch (HB), ∼ . M V . –1.5;) dwarf spheroidal (dSph) galaxies (see,e.g., Baker & Willman 2015; Vivas et al. 2016 and ref-erences therein). In many brighter dSph galaxies (–13 . M V . –9), both satellites and isolated, the num-ber of RRL is greater than ≈ §
5) has led to a much bet-ter understanding of their relative distributions in dwarfgalaxies of different morphological type. The studyof variable stars in satellites of the Andromeda galaxy(And, M31) is largely incomplete. This has been longdue to two main reasons: i) their (relatively) faint appar-ent magnitude ( V ∼
25 mag), and ii) the stellar crowd-ing. The first successful attempt to identify RRL stars inthe M31 halo was achieved by Pritchet & van den Bergh(1987), using Canada-France-Hawaii telescope data.Saha & Hoessel (1990) and Saha et al. (1990) detectedcandidate RRL stars in the dwarf elliptical M31 satel-
These observations are associated with programs lites NGC185 and NGC147. Nevertheless, with the ad-vent of the
Hubble Space Telescope (HST) it was pos-sible to reach well below the HB. This allowed thefirst determination of the properties of RRL stars inthe M31 field and its satellites. Based on WFPC2data, the discovery of RRL stars was reported in And I(Da Costa et al. 1996), And II (Da Costa et al. 2000)and And III (Da Costa et al. 2002). The population ofvariable stars detected in the three galaxies were lateranalyzed in detail by Pritzl et al. (2004, And II) andPritzl et al. (2005, And I and And III). Additionally,And VI was studied by Pritzl et al. (2002) on the basisof data of comparable quality. Since then, the number ofknown satellites of M31 has increased dramatically, pri-marily due to the PAndAS survey (McConnachie et al.2009). With a few exceptions (And XI, And XIII,Yang & Sarajedini 2012; And XIX; Cusano et al. 2013;And XXI; Cusano et al. 2015; And XXV; Cusano et al.2016) most of them have not been investigated for stellarvariability. Moreover, the knowledge of the propertiesof RRL stars in M31 itself and in the largest satellites(M32, M33) is limited to a few ACS fields and is farfrom being complete.Under the ISLAndS project (based on very deep,multi-epoch HST ACS and WFC3 data), six M31 dSphsatellite companion galaxies were observed: And I,And II, And III, And XV, And XVI and And XXVIII.The main goal of this project is to determine whether thestar formation histories (SFHs) of the M31 dSph satel-lites show notable differences from those of the MW. Theproject is described in more detail in the project presen-tation paper (Skillman et al. 2017) while the first resultsconcerning the SFH of And II and And XVI were pre-sented in Weisz et al. (2014) and Monelli et al. (2016).In order to complement these previous studies, this pa-per focuses on the study of variable stars –mainly RRLs,but also Anomalous Cepheids (ACs)– present in the sixISLAndS galaxies. The data obtained within the frame-work of this project have allowed us to increase by afactor 2–3.4 the number of known variable stars andthe quality of the light curves in And I, And II, andAnd III compared to previous studies (Pritzl et al. 2004,2005). On the other hand, this project provides thefirst discoveries of variable stars in And XV, And XVI,and And XXVIII, although an analysis of the RRL inAndXVI within the context of its SFH has been pre-sented in Monelli et al. (2016); for homogeneity with therest of the observed ISLAndS galaxies, in this work wereanalyzed the And XVI variable stars from scratch ob-taining slightly refined values. Initial Star formation and Lifetimes of Andromeda Satellites ariable stars in ISLAndS galaxies § § § §
5. RRL starsare used in § § § OBSERVATIONS AND DATA REDUCTIONTable 1 presents a compilation of updated values forthe position of the center (RA and Dec, column 2 and 3,respectively), absolute M V magnitude (column 4), red-dening (E( B - V ), column 5) and structural parameters–ellipticity ( ǫ , column 6), position angle (PA, column 7),half-light radius (r h , column 8) and tidal radius (r t , col-umn 9)– for each of the six observed galaxies under theISLAndS project (hereafter called ISLAndS galaxies ).The data for these six ISLAndS galaxies have beenobtained under proposals GO-13028 and GO-13739, fora total of 111 HST orbits. They consist of one ACSpointing on the central region and a WFC3 parallel field(at 6 ′ from the ACS center) for each galaxy. For furtherdetails about the ACS and WFC3 field location, thereader is referred to Figure 4 by Skillman et al. (2017),where the strategy and the description of the ISLAndSproject is explained in depth.For both cameras, the F W and F W passbandswere chosen. The observing strategy was designed inorder to optimize the phase coverage of short periodvariables (between 0.3 and 1.2 d), specifically RRL andAC stars. In particular, the observations were spreadover a few days (from two to five), and the visits wereplanned to avoid accumulation of data around the sametime of day, in order to avoid aliasing problems around0.5 or 1 day periods. An overview of the observing runsis provided in Table 2, which specifies, for each galaxy(column 1), the beginning and ending dates (column 2),and the number of orbits obtained (column 3). For anoptimal sampling of the light curves, each orbit was splitinto one F W and one F W exposures, yieldingthe same number of epochs in each band for each galaxy.Detailed observing logs are presented in the Appendix A(Tables A1, A2, A3, A4, A5, and A6).The photometry has been homogeneously performedwith the DAOPHOT/ALLFRAME suite of routines, following the prescriptions described in Monelli et al.(2010), for both the ACS and parallel WFC3 fields.The photometric catalogs have been calibrated to theVEGAMAG photometric systems adopting the updatedzero points from the instrument web page. VARIABLE STARS IDENTIFICATIONCandidate RRL stars and ACs were searched for in arectangular region of the CMD with a width that cov-ers the full color range of the HB, and with height be-tween ∼ . We visually in-spected the LCs of all the stars in this region, withoutany cut on a variability index. The number of candidatesranged from 201 in And XVI to 7414 in And I. Theperiodogram was calculated between 0.2 and 10 days,a range which encompasses all the possible periods ofRRL stars and ACs. Pulsational parameters were de-rived for the confirmed variables sources. Using widgetbased software, we first estimate the period of candidatevariables through the Fourier analysis of the time series,following the prescription of Horne & Baliunas (1986).The analysis is refined by visual inspection of the LCsin both bands simultaneously in order to fine-tune theperiod. The intensity-averaged magnitudes and ampli-tudes of the mono-periodic light curves were obtained byfitting the LCs with a set of templates partly based onthe set of Layden et al. (1999) following the proceduredescribed in Bernard et al. (2009). We expect that thecompleteness of both the RRL star and AC samples are100% within each pointing for the following reasons: i) the search for candidates, described above, insures thatany star showing brightness variations has been visualinspected; ii) according to the artificial star tests pre-sented in Skillman et al. (2017), the photometric com-pleteness at the magnitude of the HB (and above) isabout 100%; and iii) the amplitude of the RRLs andACs pulsations are significantly larger than the magni-tude uncertainty in the region of the HB and above.The classification of variable stars was based on theirpulsational properties (period and amplitude), LCs, andpositions on the CMD. Table 3 summarizes the totalnumber of different types of variable stars detected.Most of them are RRL stars (870 in the dwarfs + 7field stars), while a few are ACs (15) and EBs (3). Eachvariable type will be described in more detail in the sub-sequent sections.The individual F W and F W measurements Other types of variable stars, such as long period RGB/AGBstars or very short period such as δ -Scuti could not be detectednor properly characterized with the current data set, so we focuson the core helium-burning ones only. Mart´ınez-V´azquez et al.
Table 1 . Positions and structural parameters for the ISLAndS galaxies.
Galaxy RA Dec M V E(B–V) ǫ = 1 − b/a PA r h r t References(name) (hh mm ss) ( o ′ ′′ ) (mag) (1-b/a) ( o ) ( ′ ) ( ′ )And I 00:45:39.7 38:02:15.0 –11.2 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± +0 . − . ± ±
15 1.3 ± ∼ ± ± ± ± ∼ ± ± ± ± ∼ Table 2 . Summary of HST observation.
Galaxy Obs. Dates OrbitsAnd I September 1-6, 2015 22And II October 4-6, 2013 17And III November 24-28, 2014 22And XV September 17-20, 2014 17And XVI November 20-22, 2013 13And XXVIII January 20-25, 2015 20 (time-series) for all the detected variables are avail-able in Appendix B (Tables B7, B8, B9, B10, B11,and B12). The typical photometric uncertainties onindividual measurements are of the order of 0.07 magand 0.06 mag in F W and F W , respectively, forthe most distant galaxy (And I), while for the near-est galaxy (And XVI) it is of the order of 0.04 mag inboth passbands. The variable stars were named witha prefix which refers to the galaxy, followed by “V”,indicating that the star is a variable (e.g., “AndI-V”)and a number which increases with increasing right as-cension. Interestingly, we note that no variable starswere detected in the parallel fields (WFC3) of And XV,And XVI, and And XXVIII, in agreement with the vi-sual appearance of the CMD that does not show anyobvious evolutionary sequence (HB, RGB, nor the morepopulous main sequence turn-off). We also note thatthe RRL stars of And XVI were already presented inMonelli et al. (2016), but are included in this work aswell for completeness. As some of our target galaxieshave been previously investigated for variability (And I:Pritzl et al. 2005; And II: Pritzl et al. 2004; And III:Pritzl et al. 2005), a detailed comparison is presented inAppendix C.The derived values of the pulsational properties (pe-riod, amplitudes, mean magnitudes) for the variablestars detected in the different galaxies are presentedin Appendix D (Tables D16, D17, D18, D19, D20,and D21). These tables include the star name, posi-tion (RA, Dec), period, mean magnitude and ampli-tude in the F W , F W , B , V and I passbands,and the classification. We note that the HST magni-tudes in the VEGAMAG system were transformed to the Johnson system using the calibration provided byBernard et al. (2009). The main purpose of this conver-sion from F W and F W magnitudes to Johnson BV I is not only to allow comparison with observationsof variable stars in globular clusters (GCs) and othergalaxies reported in the literature (see §
5) but also forusing the period-luminosity relations (for example to ob-tain distances, as we do in §
6) or the Bailey (period-amplitude) diagram (see §
4) that are most commonlyused in the V band.We display in Figure 1 the CMDs of the ACS fields ofthe six galaxies highlighting in them the different typesof variable stars detected: RRL stars (blue stars symbolsfor those dSph members and green open circles for fieldRRL stars), ACs (red circles) and EBs (magenta trian-gles). Table 3 displays the number of detected variablesof each type in the ACS fields. The CMD of And Ishows clearly the contamination of the M31 Giant Stel-lar Stream (GSS, Ibata et al. 2001; Ferguson et al. 2002;McConnachie et al. 2003) as shown by the presence of asecond, redder RGB and red clump visible in the CMD.In particular, we have found 5 RRL stars with propertiescompatible with membership in the GSS (see § ) and as a many as 296 (in And I).The striking difference in the number of RRL betweenAnd XVI and And XV, despite having a similar mass,can be explained as a consequence of their differentSFHs: the mass fraction already in place at old ages(10 Gyr ago) was only about 50% in And XVI, whileit was 90% in And XV (see Figure 7 in Skillman et al.2017).A few (3-4) ACs are present in And II, III, XV, andXXVIII, but none have been detected in And I nor inAnd XVI. This is not surprising in the case of the lat-ter, due to its low mass . The lack of ACs is however Excluding the RRL star AndXVI-V001 (V0 in Monelli et al.2016) as it is a candidate M31 halo field star not belonging toAnd XVI. The initial estimate of its luminosity (M V = –9.2 magIbata et al. 2007) suggested a relatively bright object. However,more recent estimates (Martin et al. 2016) revised this value to a ariable stars in ISLAndS galaxies F W
261 RRLs5 GSS RRLs1 EB And I 217 RRLs4 ACs1 EBs And II 2826242220 F W
108 RRLs1 M31 RRL4 ACs And III 117 RRLs4 ACs And XV-0.5 0.0 0.5 1.0 1.5 2.0 2.5F475W-F814W2826242220 F W Figure 1 . CMDs of the ACS fields for each ISLAndS galaxy. The And I CMD shows a significant contamination from M31Giant Stellar Stream (Ibata et al. 2001; Ferguson et al. 2002; McConnachie et al. 2003). Variable stars are overplotted. Bluestars represent the RRL stars. Red circles are the ACs. Green open circles are RRL stars tentatively associated with the fieldof M31. Magenta triangles are the probable eclipsing binaries.
Mart´ınez-V´azquez et al.
Table 3 . Variable star detections.
And I And II And III And XV And XVI And XXVIII TotalACS 261 217 108 117 8 87 a
296 251 111 117 8 87 870
ACS 0 4 4 4 0 3 15AC WFC3 0 0 0 0 0 0 0total
ACS 1 1 0 0 0 0 2EB WFC3 0 1 0 0 0 0 1total
ACS 5 b c c TOTAL
ACS
267 222 113 121 9 90 822TOTAL
WFC
35 35 3 0 0 0 73TOTAL
302 257 116 121 9 90 895 a Includes two stars with rather noisy light curves. Based on their position in the CMD, we assume they are RRL stars. b RRL (3 RRab + 2 RRc) stars compatible with being field stars of the giant stellar stream (GSS) of M31. c RRab star compatible with a candidate field star from M31. -0.5 0.0 0.5 1.0 1.5 2.0 2.5F475W-F814W2826242220 F W
35 RRLs And I-0.5 0.0 0.5 1.0 1.5 2.0 2.5F475W-F814W 34 RRLs1 EB And II-0.5 0.0 0.5 1.0 1.5 2.0 2.5F475W-F814W 3 RRLs And III
Figure 2 . CMDs of the parallel WFC3 fields for the three ISLAndS galaxy where there is still a relevant stellar population.Variable stars are overplotted. As in Figure 1, blue stars represent the RRL stars, and magenta triangles are the probableeclipsing binaries. In the case of the And I CMD, the contamination from M31 Giant Stellar Stream (Ibata et al. 2001;Ferguson et al. 2002; McConnachie et al. 2003) is still present. ariable stars in ISLAndS galaxies Figure 3 . Spatial distribution of the variable stars found inthe observed ACS+WFC3 fields for And I, II and III. Non-variable stars are represented by gray dots. Variables areshown with the same symbol and color code as in Figure 1.The black ellipses represent the half-light radius (r h ) for eachgalaxy (column 6 in Table 1). remarkable in the case of And I, as no other massivedSph presents such a dearth of ACs (see § RR LYRAE STARS4.1.
Mean properties and Bailey diagrams
RRL stars are low-mass ( ∼ ⊙ ) and radi-ally pulsating variable stars with periods ranging from0.2 to 1.0 d and V amplitudes from 0.2 to . >
10 Gyr) stellar population (Walker 1989; Smith 1995;Catelan & Smith 2015). A total of 870 RRL stars weredetected and characterized in the six ISLAndS galax-ies. Table 4 summarizes, for each galaxy, the number offundamental (RRab), first overtone (RRc) and double-mode (RRd) pulsators in both the ACS and WFC3 fieldsof view. Different types of RRL stars are usually easyto classify on the basis of a visual inspection of thelight curve and the period. RRab stars are character-ized by longer periods ( ∼ ∼ V . F W and F W passbands, respectively. Opensymbols are used to indicate outlier measurements thathave not been taken into account in deriving the pul-sational properties. We emphasize that the whole set Mart´ınez-V´azquez et al.
Figure 4 . Spatial distribution of the variable stars found in the observed ACS fields for And XV, XVI and XXVIII. Non-variablestars are represented by gray dots. Variables are shown with the same symbol and color code as in Figure 1. The black ellipsesrepresent the half-light radius (r h ) for each galaxy (column 6 in Table 1). The WFC3 fields are not shown here because theCMDs of these three fields do not have any evidence of a satellite stellar population. Table 4 . RRL star subgroups.
And I And II And III And XV And XVI And XXVIII TotalACS 203 160 83 80 3 35 562RRab WFC3 26 27 1 0 0 0 53total
229 187 84 80 3 35 615
ACS 42 42 13 24 5 35 158RRc WFC3 6 6 2 0 0 0 14total
48 48 15 24 5 34 172
ACS 16 15 12 13 0 15 69RRd WFC3 3 1 0 0 0 0 4total
19 16 12 13 0 15 73
TOTAL
ACS
261 217 108 117 8 85 a WFC
35 34 3 0 0 0 72TOTAL
296 251 111 117 8 85 a a We have identified 2 additional RRL star candidates with noisy light curves. We do not include them here because of the uncertainty in theirclassification.
Table 5 . Mean properties of the RRL stars.
Galaxy h P ab i h P c i f c f cd % Oo-I % Oo-II h m V i And I 0.597 ± σ =0.07) 0.343 ± σ =0.03) 0.17 0.23 80 20 25.13And II 0.601 ± σ =0.07) 0.332 ± σ =0.04) 0.20 0.25 80 20 24.78And III 0.622 ± σ =0.03) 0.344 ± σ =0.04) 0.15 0.24 89 11 25.04And XV 0.608 ± σ =0.05) 0.360 ± σ =0.04) 0.23 0.32 78 22 25.07And XVI 0.636 ± σ =0.02) 0.356 ± σ =0.04) — — 67 33 24.34And XXVIII 0.624 ± σ =0.07) 0.359 ± σ =0.04) 0.50 0.59 49 51 25.14Notes.-Mean periods are given in days.The definition of f c is NcNab + Nc while f cd is defined as Nc + NdNab + Nc + Nd ariable stars in ISLAndS galaxies F W , F W AndI-V001 p=0.569 F W , F W AndII-V001 p=0.332 F W , F W AndIII-V001 p=0.400 F W , F W AndXV-V001 p=0.503 F W , F W AndXVI-V002 p=0.358 F W , F W AndXXVIII-V001 p=0.642
Figure 5 . Examples of light curves of member RRL starsfrom each of the six ISLAndS galaxies in the F W (black)and F W (gray) bands. Periods (in days) are given in thelower-right corner, while the name of the variable is displayedin the left-hand side of each panel. Open symbols show thedata-points for which the uncertainties are larger than 3- σ above the mean error of a given star; these data were notused in periods and mean magnitudes calculations. All RRLlight curves are available in the electronic edition of TheAstrophysical Journal . of light curves is available in the electronic edition ofthis paper. Additionally, the properties of the individ-ual variable stars can be found in Appendix D.Figure 6 presents the period-amplitude (Bailey) dia-gram for the six galaxies (see caption for details). Theplot shows the two different relations for the Oosterhofftypes, represented in the plot by the dashed lines (Oost-erhoff I and II, or Oo-I and Oo-II Cacciari et al. 2005).As long known (Oosterhoff 1939, 1944), the properties ofRRab stars divide Galactic GCs into two groups, calledOosterhoff I and Oosteroff II. The mean period of funda-mental pulsators of the former group is shorter (P ∼ ∼ Oosterhoff gap betweenthe two Oostherhoof groups. For this reason, they havebeen often considered as
Oosterhoff-intermediate types(see e.g., Kuehn et al. 2008; Bernard et al. 2009, 2010;Garofalo et al. 2013; Stetson et al. 2014; Cusano et al.2015; Ordo˜nez & Sarajedini 2016). A V HASP
OoI OoIIOoIIAnd I
HASP
OoI OoIIOoIIAnd II 0.00.51.01.5 A V HASP
OoI OoIIOoIIAnd III
HASP
OoI OoIIOoIIAnd XV-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0log P0.00.51.01.5 A V HASP
OoI OoIIOoIIAnd XVI -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0log P
HASP
OoI OoIIOoIIAnd XXVIII
Figure 6 . Period-amplitude or Bailey diagrams for the RRLsamples. Stars and circles represent RRab and RRc stars (re-spectively) found in the ACS field (black) and in the WFC3(red). Blue squares display the five RRLs which are prob-able M31 field stars. The dashed gray lines are the rela-tions for RRab stars in Oo-I and Oo-II clusters obtainedby Cacciari et al. (2005) while the dotted gray line delimitsthe middle position between the last two. The gray solidcurve is derived from the M22 (Oo-II cluster) RRc stars byKunder et al. (2013). Grey vertical lines mark the HASPlimit defined by Fiorentino et al. (2015) (see text for furtherdetails). For the sake of clarity, RRd stars are not plotted.
Table 5 summarizes the mean pulsational propertiesfor the galaxies in our sample: the mean periods of RRab( < P ab > ) and RRc ( < P c > ) type stars, the fractionof RRc ( f c ) and of RRc+RRd ( f cd ) stars, the fractionof Oo-I-like and Oo-II-like stars (defined below in thissection), and the apparent mean magnitude in V -band(which will be used in § Oosterhoff-intermediate , since they have < P ab > ∼ Mart´ınez-V´azquez et al. ley diagram between the two typical Oosterhoff lines.Figure 6 clearly shows that stars tend to clump aroundeach Oosterhoff group locus, and with a predominance of
Oo-I like stars. In fact, if we split the sample using thedotted, intermediate line, and classify stars as
Oo-I like or Oo-II like according to their relative position with re-spect of this separator, four galaxies (And I, II, III, andXV) present a majority ( ∼ Oo-I like stars (seeTable 6). In the case of And I and II, the same resultwas found for the variable stars in the parallel WFC3.And XXVIII is the exception, with a fraction of
Oo-I like stars close to 50%. Moreover, the distributionof RRLs in the Bailey diagram is also different fromthe other And dSphs; the RRab stars show a broadspread and are not concentrated on either Oosterhoffline. And XXVIII is also peculiar for the large frac-tion of RRcd type stars, which represent ∼
58% of thetotal. In the LG, if we exclude low-mass galaxies withvery small samples of RRLs ( <
15, such as e.g., BootesI and And XVI, see § V band larger than 0.75 mag. These stars are interpretedas the metal-rich tail of the metallicity distribution ofRRL stars ([Fe/H] > –1.5), and have been found only insystems that were dense or massive enough to enrichto this metallicity before 10 Gyr ago (Fiorentino et al.2017). We confirm this trend with the six ISLAndSgalaxies, as HASPs have only been detected in the twomost massive satellite galaxies: And I (3 ) and And II(2). A detailed analysis of the chemical properties ofRRL stars will be discussed in a forthcoming paper.It is worth noting that a few stars with HASPproperties were already identified in the catalogs byPritzl et al. (2004) and Pritzl et al. (2005) for And IIand And I, respectively. In the case of And I, we con-firm the HASP nature of 3 out of the 7 stars, while theperiod was likely underestimated for the other 4, possi-bly due to aliasing (see Appendix C). However, we donot confirm any of the 8 HASP stars in And II (see the The other two most likely belong to the M31 GSS, see § Appendix C for a detailed comparison with literaturevalues). Nevertheless, we discovered 2 new HASPs inAnd I and 2 in And II, which are all located outside theWFPC2 field studied by Pritzl et al. (2004, 2005).4.2.
Five detected RR Lyrae stars from M31 GSS
Five RRLs in And I have mean magnitudes that area few tenths of a magnitude fainter than the HB (threeRRab: AndI-V053, AndI-V110 and AndI-V113; and twoRRc: AndI-V257 and AndI-V280). We exclude the pos-sibility that sampling problems of the light curve may becausing a bias toward fainter magnitudes. Possible ex-planations are: i) a significantly higher metal content, or ii) a distance effect. Assuming they are at the distanceof And I, in order to explain such faint luminosity (0.45mag fainter) a super solar metallicity is required. Thisvalue appears to be unlikely given the morphology ofthe CMD and the star formation history (Skillman et al.2017).On the other hand, as indicated in the previous sec-tion, the CMD of And I shows that a significant con-tamination by the GSS is present along the line of sightto And I. In particular, And I is projected on the GSS“Field 3” studied by McConnachie et al. (2003), whichis located at 860 ±
20 kpc according to the TRGB deter-mination. To verify whether the faint RRL stars can beassociated to the GSS, we first note that two of the threeRRab are HASP RRL stars. This suggests that theirmetallicity is likely to be higher than –1.5 dex. Assum-ing [Fe/H]=–1.5 and using the period-Wesenheit rela-tion described in § µ =24.86 mag (sys=0.08; rand=0.11), for thefive stars, corresponding to 937 kpc (sys=34; rand=47).This means that they are likely located ∼
140 kpc beyondAnd I (d ⊙ ∼
800 kpc, see § PROPERTIES OF THE OLD POPULATION INTHE M31 AND MW SATELLITES SYSTEM5.1.
Comparing the HB morphologies of the MW andM31 satellites
Pioneering works by Da Costa et al. (1996, 2000,2002) based on shallower WFPC2 data disclosed the firsthint that the M31 satellites are characterized by redderHB morphology with respect to MW dwarfs. A similarconclusion was reached by Martin et al. (2017), based onACS data for 20 M31 galaxies. The analysis was basedon a morphological index accounting for the number ofblue and red HB stars. In this section we apply a similarapproach, and taking advantage of the known number of ariable stars in ISLAndS galaxies Table 6 . Properties of the set of RRL stars in a sample of 41 LG dwarf galaxies of different morphological type within ∼ RRab h P ab i h P cd i Galaxy h [Fe/H] i ∗ RRab %OoI %OoII RRcd f cd Median Mean Median Mean ReferencesMW dwarf satellitesUrsa Major I -2.18 5 60 40 2 0.29 0.600 0.628 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ∗ Mean metallicity for each galaxy obtained from McConnachie (2012). Mart´ınez-V´azquez et al. -14 -12 -10 -8M V -1.0-0.50.00.5 R H B Sculptor DracoFornax LeoII CarinaAnd IIIAnd XV And XVIAnd XXVIIIAnd IAnd II
MW: < R
HB, 1rh > = -0.37+/-0.06
M31: < R
HB, 1rh > = -0.58+/-0.07
Figure 7 . R HB index versus the luminosity of the hostgalaxy, M V , for the ISLAndS targets (orange filled stars)and a sample of MW satellites (blue open diamonds). Thevalues have been calculated within 1 r h , except for And I andII (red stars) since the field of view of the ACS is not largeenough. The mean value for M31 satellites support redderHB morphology than for MW satellites. -3 -2 -1 0[Fe/H]0.550.600.650.700.75 < P > ab MWM31
Figure 8 . Left and Middle -
Mean period of the RRab starsof the sample of MW (blue) and M31 (orange) satellites ver-sus the mean metallicity and the percentage of Oo-I typestars in the system. There is no obvious between the twosubgroups.
Right -
Mean period distribution of the sampleof MW (blue histogram) and M31 dwarf galaxies (orange his-togram). The peaks of the two distribution are very close toeach other.
RRL stars, we can compare the morphology index R HB R HB =(B-R)/(B+V+R) where B and R are the numbers ofHB stars bluer and redder than the IS, respectively, and V is the -3 -2 -1 0[Fe/H]0.550.600.650.70 < P > ab LG dwarfsGCs
Figure 9 . Left - < P ab > for a sample of 41 dwarf galaxiesreported in Table 6 (black dots) as a function of [Fe/H], com-pared to that of GCs (purple bowties). Right -
Period distri-bution of the sample of dwarf galaxies and GCs. The peak ofthe former occurs at a period typical of the Oo-intermediatesystem, while the latter peaks in the short period regime,populated by Oo-I systems, which is devoid of galaxies. of the six ISLAndS galaxies and of a sample of MWsatellites. The latter consists of revised data for Carina,Fornax, Sculptor, Draco, and Leo II from the updatedcatalogs available in P. B. Stetson’s database (Stetson2017, priv. comm.). These studies are part of an ongo-ing series of papers on variable stars in globular clustersand dwarf galaxies by ourselves and our collaborators(Stetson et al. 2014; Braga et al. 2015; Coppola et al.2015; Mart´ınez-V´azquez et al. 2016b; Braga et al. 2016;Fiorentino et al. 2017).The value of the R HB index was calculated in a ho-mogeneous way considering only stars within 1 half-lightradius, r h . This was possible for all of the galaxies exceptfor And I and II, since the ACS only covers a fraction ofsuch area (see Figure 3). In the case of the MW satel-lites, we estimated and subtracted the Galactic field-starcontribution using a proper control field in the outskirtsof each object. The exact limits in color and magni-tude for the selection of HB stars for the R HB indexwere defined on a per-galaxy basis because of the varietyof CMD morphology, filter bandpasses, and foregroundcontamination. However, these were carefully chosen tolimit contamination from any RGB, AGB, RC, and bluestraggler populations present, while also avoiding biases.Figure 7 shows, as a function of the host galaxy ab-solute M V magnitude, the R HB index calculated inside1 r h for the MW (open diamonds) satellites and for theISLAndS (stars) galaxies. And I and II are calculatedover the full ACS area, which is smaller than 1 r h . The number of RRL stars (Lee 1990). ariable stars in ISLAndS galaxies σ . In fact, within 1 r h the meanvalue of R HB is more negative in the case of M31 galax-ies (R HB,M = –0.58 ± HB,MW = –0.37 ± i) the better photometric precision at the HB level, and thefilter combination providing better color discriminatingpower, allows us to clearly separate the red HB fromthe blue edge of the RGB, even in the case of And I; ii) the larger field of view of ACS compared to WFPC2provided a larger sample; iii) the up-to-date, wide field,homogeneous data available for the MW companions al-lowed us to perform the comparison in a more homoge-neous manner; iv) the better phase coverage allowed usto derive better defined mean colors of RRL stars.The current data do not allow us to fully explorewhether the HB morphology presents significant varia-tion as a function of galactocentric distance, i.e., dis-tance from the center of each galaxy. Nevertheless,when considering the parallel WFC3 field for And Iand And II, we derive larger values of the R HB index,and therefore an indication that the HB morphologygets bluer when moving to an external region. Thisis in agreement with what was found for other LGgalaxies (e.g., Harbeck et al. 2001; Tolstoy et al. 2004;Cole et al. 2017), and more in general with the pop-ulations gradients commonly found in dwarf galaxies(Hidalgo et al. 2013, and references therein). In fact,when considering the area within 2 r h , the six galaxiestend to have bluer HB. Unfortunately, a straight com-parison between the two satellite systems is complicatedby the fraction of area covered. This leaves open thequestion of whether the HB morphology remains differ-ent at larger galactocentric distances, or whether M31and MW satellites tend to be more similar when theirglobal properties are taken into account. More wide-field variability studies, particularly for the M31 satel-lites, would help solve this problem.5.2. Global properties of RRL stars In § < P ab > as a function of the meanmetallicity of the host galaxy (left panel), for 16 satel-lites of M31 (filled orange stars) and 15 MW dwarfs (blueopen diamonds). Galaxies with at least 5 known RRabstars have been included. The plot discloses that themean period of RRab type stars decreases for increas-ing mean metallicity of the host system (Sandage et al.1981), for both the MW and the M31 satellites. Thetrend presents some scatter, but interestingly a linearfit to the data provides very similar slope (0.040 ± ± present-day mean metallicity of the host galaxy.This suggests that galaxies that today are more massiveand more metal-rich on average also experienced fasterearly chemical evolution, which is imprinted in the prop-erties of their RRL stars. This implies that the mass-metallicity relation (e.g., Kirby et al. 2013) was in placeat early epoch (Mart´ınez-V´azquez et al. 2016b).The central panel of Figure 8 shows the mean period asa function of the fraction of Oo-I type stars, as defined in §
4. While there is no clear correlation for either satellitesystem, we find that the vast majority of galaxies host alarger fraction of Oo-I type stars, between 60 and 90% ofthe total amount of RRab stars. Nevertheless, the meanperiod of fundamental pulsators would classify them asOo-intermediate system. Again, this suggests that theRRL stars in complex systems such as galaxies are notproperly represented by a single parameter.4
Mart´ınez-V´azquez et al.
Finally, the right panel of Figure 8 shows the mean pe-riod distribution for the RRab in MW satellites (blue)and in M31 satellites (orange). Apparently, both ofthem are similar and their peaks agree within 1- σ .The former analysis reveals that, if we limit the com-parison to strictly old and well defined populations suchthat of RRL stars, there are no obvious differences be-tween the RRL populations of the satellite systems ofM31 and the MW.Figure 9 shows the behavior of h P ab i versus [Fe/H],but comparing a sample of 41 galaxies (black circles,including MW and M31 satellites, isolated dwarfs andtwo galaxies in the Sculptor group) with GCs (magentabowtie symbols). We use here the compilation fromCatelan (2009), including all the GCs with more than10 RRL stars. Galactic GCs, as well as clusters fromthe LMC and the Fornax dSph galaxy are shown. Theplot shows that a few Oo-intermediate clusters overlapwith galaxies in the Oosterhoff gap, but most off theOo-I clusters (i.e., with P ab < .
58) occupy a region ofthe parameter space where no galaxies are present – thisholds even if we restrict the GC sample to those with30 RRL or more. This is even more evident in the rightpanel of Figure 9, which shows the mean period distri-butions of the two samples. It clearly shows that thepeak for the galaxy distribution occurs at a period typ-ical of Oo-intermediate systems, while the peak of theGCs occurs in the Oo-I regime. DISTANCE MODULIIn the following, we use four independent methods toderive the distances to the six ISLAndS galaxies, thefirst three based on the properties of the RRL stars: i) the reddening-free period-Wesenheit relations (PWR,Marconi et al. 2015); ii) the luminosity-metallicity(M V versus [Fe/H]) relation (LMR, Bono et al. 2003;Clementini et al. 2003); iii) the first overtone blue edge(FOBE) relation (Caputo et al. 2000); these are supple-mented by iv) the tip of the RGB (TRGB) method.All the aforementioned relations require an assump-tion for the metal abundance. In particular, in the caseof the PWR, LMR, and FOBE relation, we need to as-sume a metallicity corresponding to the old population(representative of the RRL stars). On the other hand,the TRGB method uses the metallicity of the RGB starsto obtain the expected mean color value of the TRGB.In complex systems like dwarf galaxies, the metallicityof the global population may range over ∼ old stellar population. In the next section,we discuss in detail the choice of the metallicity in orderto determine the distance to the six galaxies. 6.1. The choice of the metallicity
The metallicity estimates available in the literaturefor the ISLAndS galaxies are all based on CaT spec-troscopy of bright RGB stars . As the RGB can bepopulated by stars of any age larger than ∼ ∼ σ of the metal-licity distribution (column 3), and the number of RGBstars (column 4) used in these studies (references in col-umn 5). Relatively low values were found for And III,And XV, And XVI, and And XXVIII, on average closeto [Fe/H] ∼ –1.8 or lower. On the other hand, in the caseof And I and And II, different authors (Kalirai et al.2010; Ho et al. 2012) agree on a much higher meanmetallicity ([Fe/H] ∼ –1.4), and a relatively large metal-licity spread ( σ AndI =0.37 dex, σ AndII =0.72 dex). Nev-ertheless, the small number of HASP stars (see §
4) sug-gests that, even if the tail of the RRL metallicity distri-bution reaches such relatively high values, the bulk ofthe RRL stars must have a lower metallicity ([Fe/H] < –1.5, Fiorentino et al. 2015). Therefore, as representativevalues of the metallicity for the RRL population, weadopted –in agreement with their SFHs (Skillman et al.2017)– [Fe/H]=–1.8 for And I and And II while, for therest of the galaxies, we assume that the metallicity of theold population must to be quite similar to that obtainedby the spectroscopic studies (see column 8).The adopted mean metallicities for each ISLAndSgalaxy are summarized in the second to last column ofTable 7. We note that the values have been homoge-nized to the scale of Carretta et al. (2009). Column 2reports the value in the original scale, which is specifiedin column 6. In those cases based on theoretical spectra,we applied a correction to take into account the differ-ent solar iron abundance (from log ǫ F e = 7.45 to 7.54),which translates to a distance modulus change between0.01 mag in the case of the FOBE and 0.04 in the caseof the LMR. In the case of And II, Kalirai et al. (2010) estimate both aphotometric and a spectroscopic metallicity, concluding that withthe data at hand the former is less dependent on the low S/N ofthe measurements. ariable stars in ISLAndS galaxies Table 7 . Metallicity studies with the largest samples of RGB stars.
RGB stars RRL starsGalaxy h [Fe/H] i σ h [ Fe/H ] i N stars References Metallicity scale a h [Fe/H] i bC [Fe/H] old pop. And I –1.45 ± And II –1.39 ± And III –1.78 ± And XV –1.80 ± − c
13 Letarte et al. (2009) C09 –1.80 –1.8
And XVI –2.00 ± − d
12 Collins et al. (2015) G07 –1.91 –2.0
And XXVIII –1.84 ± e
13 Slater et al. (2015) C09 –1.84 –1.8 a Metallicity scales: ZW84 = Zinn & West (1984), G07 = Grevesse et al. (2007), and C09 = Carretta et al. (2009). b We have either converted the metallicity to the C09 scale when it was possible, or shifted the metallicity value assuming the same Solar ironabundance (log ǫ (Fe)=7.54). The C09 scale was chosen as the homogeneous scale for being the most up-to-date. c Letarte et al. (2009) did not publish σ [ Fe/H ] . Instead they provide an interquartile range of 0.08, with a median metallicity of [Fe/H]=–1.58dex. d Collins et al. (2015) did not publish σ [ Fe/H ] . However, Letarte et al. (2009) published for And XVI an interquartile range of 0.12, with amedian of [Fe/H]=–2.23 dex. By stacking the spectra of the member stars (8 in this case), they found [Fe/H]=–2.1 with an uncertainty of ∼ e As this σ is obtained from a small number of individual measurements, it may not be representative of the actual distribution. The period-Wesenheit relations
PWRs are a powerful tool for distance determination,because they are reddening-free by construction and areonly marginally metallicity dependent. They are theo-retically described by: W ( X, X − Y ) = α + β log P + γ [ F e/H ] (1)where X and Y are magnitudes and W( X , X – Y ) de-notes the reddening free Wesenheit magnitude (Madore1982) obtained as W( X , X – Y )= X –R( X – Y ), where R isthe ratio of total-to-selective absorption, R=A X /E( X - Y ).An updated and very detailed analysis of the frame-work of the PWRs is provided by Marconi et al. (2015).Their Tables 7 and 8 give a broad range of optical,optical-NIR, and NIR PWRs, along with their corre-sponding uncertainties. In particular, in this work, weuse their PWR in the ( I , B − I ) filter combinations : W ( I, B − I ) = − . ± .
01) +( − . ± .
02) log P + (0 . ± . F e/H ] (2)which has an intrinsic dispersion of σ = 0.04 mag. Forthis relation, a metallicity change of 0.2 dex translatesinto a change in the distance of order 0.03 mag.The theoretical W( I , B – I ) was obtained from the in-dividual stars assuming a metallicity for the old popu-lation (see column 8 in Table 7 and discussion of § I , B – I )= I –0.78( B – I ). We report the According to the equations obtained by Bernard et al. (2009)to transform F W and F W to Johnson-Cousins BV I , both B and V are transformed from F W . For this reason we can-not apply the metal-independent PWR ( V , B − V ) published byMarconi et al. (2015), because B and V are correlated. distance moduli obtained by averaging individual esti-mates for the global sample (RRab + fundamentalizedRRc: log P fund = log P RRc + 0.127; Bono et al. 2001)in columnn 2 of Table 8. For comparison, if we useindependently the sample of RRab and RRc, the val-ues from the different determinations agree on averagewithin ± The luminosity-metallicity relation
The LMR is another simple, widely used approachto determine distances, in this case using the mean V magnitude of RRL stars. Despite the fact thatboth theoretical and empirical calibrations suggestthat the relation is not linear (being steeper in themore metal-rich regime (see e.g., Caputo et al. 2000;Sandage & Tammann 2006; Cassisi & Salaris 2013, andreferences therein), most examples in the literature useone of the different linear relations proposed.In the present work, we adopted the following rela-tions: h M V i = 0 . ± . . ± . F e/H ] (3)from Clementini et al. (2003), and h M V i = 0 . ± . . ± . F e/H ] , (4)from Bono et al. (2003) .The latter is valid only for metallicity lower than[Fe/H]=–1.6, which is appropriate for the six ISLAndSgalaxies (where the metallicity of the old population is The zero-point of this equation, as well as for the FOBE equa-tion 5 presented in next section, has been modified according withthe shift of +0.05 to correct for the electron-conduction opacities(Cassisi et al. 2007). Mart´ınez-V´azquez et al. considered to be [Fe/H]=–1.8 or lower). Columns 3 and4 of Table 8 show the true distance moduli obtainedusing the two relations. They are in excellent agreementwith each other and with those derived previously usingthe PWR. 6.4.
The FOBE method
Another method that can be used to estimate thedistance is based on the predicted period-luminosity-metallicity relation (PLMR) for pulsators located alongthe FOBE of the IS (see Caputo et al. 2000): M V,F OBE = − . − .
255 log( P F OBE ) − .
259 log(
M/M ⊙ ) + 0 .
058 log( Z ) (5)which has an intrinsic dispersion of σ V = 0.07 mag.This is considered a particularly robust technique forstellar systems with significant numbers of first-overtoneRRL (RRc) stars, especially if the blue side of the IS iswell populated. Thus, it can be applied safely to five ofour six galaxies (see Figure 6). The distance modulusis derived by matching the observed distribution of RRcstars to eq. 5. That is, for a given metallicity and a masscorresponding to the typical effective temperature forRRL stars, we shift the relation until the FOBE matchesthe observed distribution of RRc stars.For the adopted metallicity listed in Table 7, and usingthe evolutionary models from BaSTI (Pietrinferni et al.2004), we obtain masses at log T eff ≈ ∼ ⊙ . True distance moduli obtained for each galaxy us-ing this method are shown in column 5 of Table 8, andare in good agreement with those described in the pre-vious section. 6.5. The tip of the RGB
It is well established that the TRGB is a good stan-dard candle thanks to its weak dependence on age(Salaris et al. 2002) and, in the I band in particular,on the metallicity (at least for relatively metal-poorsystems, Da Costa & Armandroff 1990; Lee et al. 1993).The TRGB is frequently used to obtain reliable distanceestimates to galaxies of all morphological types, in theLG and beyond (e.g., Rizzi et al. 2007; Bellazzini et al.2011; Wu et al. 2014). However, determining the cutoffin the luminosity function at the bright end of the RGBis not straightforward in low-mass systems, becausemore than about 100 stars populating the top magnitudeof the RGB are required to reliably derive the locationof the tip (Madore & Freedman 1995; Bellazzini et al.2002; Bellazzini 2008). This condition is fulfilled only And XVI only has 5 RRc stars in And I ( > > ∼ F W luminosity functions witha Sobel kernel of the form [1,2,0,2,1]. From the fil-ter response function, we obtain the center of thepeak corresponding to the TRGB of each galaxy: F W ,AndI =20.45 ± F W ,AndI =20.05 ± F W ,AndIII =20.25 ± rms width of the peak of the Sobelfilter response.The distances were obtained from the TRGB magni-tudes using three calibrations: i) the empirical calibrations in the HST flight bandsfrom Rizzi et al. (2007, R07): M F WT RGB = − .
06 + 0 . F W − F W ) − .
74] ( σ = 0 .
10) (6) ii) the empirical calibration reported inBellazzini et al. (2011, B11), derived by Bellazzini(2008) from the original calibration as a function of[Fe/H] obtained in Bellazzini et al. (2001) and revisedin Bellazzini et al. (2004): M F WT RGB ≈ M IT RGB = 0 . V − I ) − . V − I ) − .
93 ( σ = 0 .
12) (7) iii) the theoretical calibration M F WT RGB , as a func-tion of the color ( F W – F W ) , obtained in thiswork by fitting the BaSTI predictions Pietrinferni et al.(2004, 2006) for the TRGB brightness for an old ( ∼ : M F WT RGB = − .
11 + 0 . F W − F W ) − . . F − F − . ( σ = 0 .
02) (8)In the case of the Rizzi et al. (2007) calibra-tion, we considered that F W – F W ∼ V – I . In fact, for both this calibration andthat of Bellazzini et al. (2011), we use the fol-lowing equation to determine the expected ( V – I ) We note that the zero-point of this theoretical calibration hasbeen corrected in order to account for the impact on the TRGBbrightness of more accurate conductive opacity evaluations. Fol-lowing the results obtained by Cassisi et al. (2007) we have cor-rected the M F WTRGB , by adding +0.08 mag. We do not have F W magnitudes for the ISLAndS dSphs,but the ( F W – F W ) color is very close to ( V – I ) ariable stars in ISLAndS galaxies V – I ) T RGB, =0.581[Fe/H] +2.472[Fe/H]+4.013(Bellazzini et al. 2001). Since this last equationis based on Zinn & West (1984, ZW84) scale, inorder to use it properly, we have to apply theconversion scales provided by Carretta et al. (2009):[Fe/H] ZW =([Fe/H] C –0.160)/1.105. Columns 6, 7,and 8 in Table 8 give the values of the true distancemoduli calculated using the previous relations for And I,And II, and And III. All three calibrations lead to dis-tances that are in good agreement with each other andwith the previously calculated RRL based distances.6.6. On the consistency of the different methods
As show in Table 8, all the distances obtained from thedifferent methods are in agreement within less than 1- σ with each other. The inclusion of the TRGB method inthis study was mainly for checking the distances we ob-tained using the properties of the RRL stars with thoseassessed with this method. In fact, it is worth men-tioning that since the TRGBs of these galaxies are notdensely populated (we have &
200 only for And I), thistechnique is secondary in our study, but it serves to showthat the metallicity we have assumed for the old popu-lation is robust. The good sampling of our LCs togetherwith the large amount of RRL stars in all these galax-ies (with exception of And XVI), make them the bestdistance indicators that we have in these galaxies so far.We adopt the distances obtained with the PWR aspreferred because: i) they are obtained with the RRLstars, ii) the PWR used for deriving them come from themost updated study (Marconi et al. 2015), and iii) thesystematic uncertainties are the smallest (see Table 8).Figure 10 summarizes the distance determinations de-rived in this work. In particular, the filled circles to-gether with the dotted line show the adopted final dis-tance measurement coming from the PWR ( § σ . Taking as reference the PWRdistance, some general trends can be noted between theresults of the different methods adopted. The distancederived using the LMR with the Bono et al. (2003) cal-ibration provides marginally larger distances with re-spect to both the Clementini et al. (2003) calibration (inagreement with the difference in the zero point), and alsowith the distance obtained from the PWR. The FOBEdistance is larger than the PWR distance in three cases(And III, XV, and XVI), and shorter for And II. Never-theless, this method is the most sensitive to the samplingof the IS, and in particular the lack of RRL close to theblue edge of the IS introduces a bias toward larger dis- µ And I µ And II µ And III µ And XV µ And XVI µ And XXVIII
PWR LMR
B03
LMR
C03
FOBE T
R07 T B11 T BaSTI
Figure 10 . Summary of our derived distances. Circles re-port values based on RRL stars while squares are based onthe TRGB (provided only for the most massive galaxies, forwhich the TRGB could be reliably estimated). The filledcircles and the dotted lines show the measurements basedon the PWR, which are the final adopted distances. Opensymbols show values obtained with the other methods, forcomparison. tances. The TRGB technique could only be applied tothe three most massive systems. Interestingly, in thecase of And II and And III the derived distance seemsto be, on average, marginally smaller, independent ofthe calibration adopted. We note that in the case ofAnd I the agreement between different indicators andmethods is remarkably good. This is possibly linked tothe fact that it presents the largest sample of RRL starsand the most populated TRGB region, thus suggestingthat statistical fluctuations have a minimal effect.6.7.
Comparison with previous works
Figure 11 displays a comparison with distance es-timates available in the literature and derived withdifferent techniques: RRL stars (open triangles:Pritzl et al. 2004, 2005), the HB luminosity (open dia-monds: Da Costa et al. 1996, 2000; Slater et al. 2015),and the TRGB (open stars: Mould & Kristian 1990;Koenig et al. 1993; McConnachie et al. 2004, 2005;Letarte et al. 2009; Conn et al. 2012). This figure showsan overall good agreement with our estimates, within theuncertainties. We note that the TRGB tends to providecloser distances than the RRL and the HB luminosity,though it is still compatible within 1.5- σ . A couple ofdiscrepant cases (And XV and XVI from Conn et al.8 Mart´ınez-V´azquez et al.
Table 8 . Summary of the different true distance moduli ( µ ) obtained using several methods. RRL stars RGB starsGalaxy PWR LMR B LMR C FOBE ∗ Tip R Tip B Tip
BaSTI
And I ± ± ± ± ± ± ± And II ± ± ± ± ± ± ± And III ± ± ± ± ± ± ± And XV ± ± ± ± And XVI ± ± ± ± And XXVIII ± ± ± ± ∗ FOBE method is based in only one RRc star, for this reason they do not have standard deviation. µ And I µ And II µ And III µ And XV µ And XVI µ And XXVIII
M90 K93 M04 M05 L09 C12 D96 D00 S15 P04 P05 This work
Figure 11 . Comparison of our adopted distance moduli(based on the PWR, filled circles and dotted lines) withthe literature values (open symbols). In particular, we re-port values based on the TRGB (stars: Mould & Kristian1990; Koenig et al. 1993; McConnachie et al. 2004, 2005;Letarte et al. 2009; Conn et al. 2012), HB luminosity (dia-mond: Da Costa et al. 1996, 2000; Slater et al. 2015), andRRL stars (triangles: Pritzl et al. 2004, 2005). OTHER VARIABLES7.1.
Anomalous Cepheid stars
AC stars are variables stars in the core He-burningevolutionary phase at luminosity higher than RRL stars.Their periods range from ∼ ∼ ⊙ . In or-der to have ACs with such masses, two different chan-nels are likely (Bono et al. 1997; Gallart et al. 2004;Cassisi & Salaris 2013). They can be the progeny of coalesced binary stars, thus evolved blue straggler stars(BSS) tracing the old population (Renzini et al. 1977;Hirshfeld 1980; Sills et al. 2009). Alternatively, theycan be an evolutionary stage of single stars with massbetween ∼ ∼ ⊙ and age between 1 and 6Gyr (Demarque & Hirshfeld 1975; Norris & Zinn 1975;Castellani & degl’Innocenti 1995; Caputo et al. 1999;Dolphin et al. 2002; Fiorentino et al. 2006). In bothscenarios, they trace the existence of a metal poor(Z < &
10 Gyr) nearby LG dwarfgalaxies host a few of them. However, they are very rarein GCs; so far, only one candidate has been confirmedin the metal-poor ([Fe/H] ∼ –2 dex) cluster NGC 5466(Zinn & Dahn 1976), and a few others have been sug-gested (Corwin et al. 1999; Arellano Ferro et al. 2008;Kuehn et al. 2011; Walker et al. 2017). On the otherhand, large samples of ACs have been collected in theLMC (141) and in the SMC (109) in the frameworkof the OGLE-IV project (Soszy´nski et al. 2015). Inthis context, it is worth mentioning that the work ofMateo et al. (1995) –updated by Fiorentino & Monelli(2012) collecting data from nearby dwarf galaxies, forwhich SFH is provided and ACs have been found (seetheir Figure 7)– noted a correlation between the fre-quency of ACs and the total luminosity of the hostgalaxy; the frequency of ACs decreases for increasingluminosity of the host galaxy. Another parameter thatseems to impact the ACs frequency is the SFH of thehost system, with primarily old systems, or fast galax-ies (as defined by Gallart et al. 2015) having a lowerspecific frequency of ACs compared to systems contain-ing an important amount of intermediate-age and youngpopulations ( slow galaxies ).Figure 1 shows the presence of a few variable stars lo-cated 1 to 2 mag above the HB. Given their pulsationproperties, their position in the period-Wesenheit dia-gram (see e.g., Figure 1 in Fiorentino & Monelli 2012),and the shape of their light curves, we classify them asACs. In particular, a total of 15 ACs have been de- ariable stars in ISLAndS galaxies : AndII-V083 (V14 by Pritzl et al. 2004),AndIII-V073, AndIII-V075, and AndIII-V105 (V01,V07, and V06 by Pritzl et al. 2005). The LCs of allthe ACs detected are shown in Figure 12. The differentshapes are an indication of different pulsational modes.However, the classification of the pulsation mode of ACsis not trivial and cannot be easily determined from themorphology of the LCs alone (Marconi et al. 2004).Figure 13 shows the period-Wesenheit plane for thereddening-free index W( I , B - I ) –which reduces the scat-ter due to the interstellar reddening and the intrinsicwidth of the IS – for the ACs of the LMC (open symbols)published by the OGLE collaboration (Soszy´nski et al.2015). This plot shows a clear separation between fun-damental (F, black dots) and first-overtone (FO, opencircles) pulsators for ACs. Therefore, ACs are defined bydifferent PL relations, and fundamental and first over-tone pulsation can also be distinguished in this way. Wetherefore overplot the 15 ACs found in this work withthe aim of checking their nature and identifying theirpulsation modes. The four ACs found in And II arerepresented by red circles, the four in And III by greensquares, the four in And XV by orange diamonds and thethree in And XXVIII by blue triangles. First, Figure 13 Contrary to Pritzl et al. (2005, theirV9), variable AndIII-V100 was classified as a RRL star as its location on the CMD isnot compatible with an AC. supports their classification as ACs. We note that onlyone star (AndXXVIII-V068) is somewhat distant fromthe bulk of the F mode ACs of the LMC. From an in-spection of the light curve of this star (see Figure 12),the lack of phase points close to the maximum light isevident. Therefore, the measurement of the mean mag-nitude of this star may be biased to brighter magnitude,as the fit with templates tends to overestimate the am-plitude. Figure 13 indicates that the majority of the ACs(12) are pulsating in the F mode, and only three of them(AndIII-V075, AndIII-V105 and AndXXVIII-V060) areFO pulsators.7.2.
Eclipsing binaries candidates
For the sake of completeness, we report the detectionof three EBs, one in And I and two in And II. Figure 14shows their LCs, which in all cases show a minimum. Forthe three candidates the minimum occurs at the samephase in the two bands. This feature, together with theflat bright part of the light curves, the periods, and theirposition in the CMD support the classification as EBs.The fact that only such a small number of candidateEBs was detected is due to both the relatively smallnumber of points per light curve taken, the non-optimaltime sampling for this kind of variable, and the limitedregion of the CMD that was searched for variables. SUMMARY AND FINAL REMARKSIn this paper we have analyzed multi-epoch HST datafor six dSphs satellites of M31 in order to study theirpopulation of variable stars. The main findings of thecurrent study are: • We have detected 895 variable stars in And I, II, III,XV, XVI, and XXVIII: 678 of them are new discoveries.In particular, we classified 870 RRL stars, 15 ACs, 3 EBand 7 field variable stars (5 of them probably belongingto the GSS of M31). Interestingly, no ACs were foundin And I despite being the second most massive dwarfin our sample, which we interpret as a hint of the fastchemical enrichment of this galaxy. • Pulsational properties (period, amplitude, mean mag-nitude) were derived for all detected variables. More-over, we provide all the light curves and time series pho-tometry. • Using the properties of RRL stars, we derived newhomogeneous distances to the six galaxies using threedifferent methods: the period-Wesenheit relation, themetallicity-luminosity relation and the first overtoneblue edge method. A fourth independent estimate wasderived using the tip of the RGB for the three mostpopulated systems. We find a satisfactory agreementboth between different methods and with most of theestimates available in the literature. It is worth noting0
Mart´ınez-V´azquez et al. F W , F W AndII-V083 p=0.881 F W , F W AndII-V099 p=1.340 F W , F W AndII-V158 p=1.663 F W , F W AndII-V189 p=1.206 F W , F W AndIII-V073 p=1.003 F W , F W AndIII-V075 p=0.428 F W , F W AndIII-V095 p=1.171 F W , F W AndIII-V105 p=0.818 F W , F W AndXV-V004p=1.136 F W , F W AndXV-V016p=0.868 F W , F W AndXV-V052p=1.356 F W , F W AndXV-V117p=2.186 F W , F W AndXXVIII-V049p=1.145 F W , F W AndXXVIII-V060p=0.559 F W , F W AndXXVIII-V068p=1.389
Figure 12 . Light curves of member AC stars for four of the six ISLAndS galaxies in the F W (black) and F W (gray)bands. Periods (in days) are given in the lower-right corner, while the name of the variable is displayed at the top of each panel.Open symbols show the data for which the uncertainties are larger than 3- σ above the mean error of a given star; these datawere not used in periods and mean magnitudes calculations. ariable stars in ISLAndS galaxies −0.4 −0.2 0.0 0.2 0.4 0.6log P−0.5−1.0−1.5−2.0−2.5−3.0−3.5 W ( I, V − I ) [ m ag ] F modeFO modeF LMC ACFO LMC ACAnd IIAnd IIIAnd XVAnd XXVIII
Figure 13 . Period-Wesenheit diagram for ACs. Black dots(F pulsators) and grey circles (FO pulsators) represent theACs of the LMC from the OGLE-IV release (Soszy´nski et al.2015). ACs discovered in our galaxies are represented by redcircles (And II), green squares (And III), orange diamonds(And XV), and blue triangles (And XXVIII). The solid anddashed lines are the empirical period-luminosity relations ob-tained by Soszy´nski et al. (2015) for ACs in the LMC for theF and FO mode, respectively. F W , F W AndI-V158p=0.815 F W , F W AndII-V010p=0.388 F W , F W AndII-V174p=1.057
Figure 14 . Light curves of the three EB candidates detectedin the field of And I and And II in the F W (black) and F W (gray) bands. Periods (in days) are given in thelower-right corner, while the name of the variable is displayedat the top of each panel. Open symbols show the data forwhich the uncertainties are larger than 3- σ above the meanerror of a given star; these data were not used in periods andmean magnitudes calculations. that those values obtained using the RRL stars are moreaccurate and precise. For these reasons, we adopted asfinal distance moduli those which are obtained throughthe period-Wesenheit relation, which are the most pre-cise values and based in the most updated relation forRRLs to date. • We have shown that, similar to MW satellites, themean period of RRab variables of the six ISLAndS is close to 0.6 day, a value that is typical of Oo-intermediate objects. On the other hand, the distri-bution of RRL stars in the Bailey diagram is such thatthe majority of stars ( ∼ • In spite of the slight difference in the HB morphologyparameter (R HB ), when we restrict the comparison be-tween M31 and MW systems to the properties of RRLstars only, we do not find significant differences betweenthe two groups of galaxies. In particular, based on asample of 16 satellites of M31 and 15 of MW, we finda similar trend between the mean period and the meanmetallicity. This suggests overall similar characteristicsof the oldest ( >
10 Gyr) population in the two systemsin agreement with what is discussed by Monelli et al.(2017) using the global period distributions of thousandsof RRL stars belonging to faint and bright satellites ofM31 and the MW.To date, none of the known Local Group dwarf galax-ies has a complete census of their entire population ofvariable stars. However, we are at the dawn of a newera for variability studies. Current and future surveysare about to bring an unprecedented amount of informa-tion on the variable stars populating the surroundingsof the MW and of the Local Group.
Gaia will bring thediscovery of thousands of new RRL stars (G . Mart´ınez-V´azquez et al. tional Aeronautics and Space Administration. Supportfor this work was provided by NASA through grantsGO-13028 and GO-13749 from the Space Telescope Sci-ence Institute, which is operated by AURA, Inc., un-der NASA contract NAS5-26555. This work has alsobeen supported by the Spanish Ministry of Economy andCompetitiveness (MINECO) under the grant (projectreference AYA2014-56795-P). EJB acknowledges sup-port from the CNES postdoctoral fellowship program.GF has been supported by the Futuro in Ricerca 2013(grant RBFR13J716). Support for D.R.W. is providedby NASA through Hubble Fellowship grants HST-HF-51331.01 awarded by the Space Telescope Science Insti-tute.
Facilities:
HST(ACS, WFC3)
Software:
IDL, DAOPHOT/ALLFRAME ariable stars in ISLAndS galaxies Table A1 . Observing log for And I.Image Name Date UT Start Filter Exp. timename (YY-MM-DD) (hh:mm:ss) name (s)
ACS (RA=00:45:42.8, Dec=+38:02:22.8) jcnb01lyq flc.fits 2015-09-01 09:56:05 F475W 1264jcnb01m1q flc.fits 2015-09-01 10:20:06 F814W 1002jcnb01m3q flc.fits 2015-09-01 11:24:32 F814W 1086jcnb01m7q flc.fits 2015-09-01 11:45:36 F475W 1372jcnb02mtq flc.fits 2015-09-01 16:17:43 F475W 1264... ... ... ... ...
WFC3 (RA=00:45:14.2, Dec=+38:00:03.0) icnb01lzq flc.fits 2015-09-01 09:55:10 F475W 1308icnb01m0q flc.fits 2015-09-01 10:19:26 F814W 1046icnb01m4q flc.fits 2015-09-01 11:24:27 F814W 1103icnb01m8q flc.fits 2015-09-01 11:45:24 F475W 1389icnb02muq flc.fits 2015-09-01 16:16:48 F475W 1308... ... ... ... ...
This table is a portion of its entirely form which will be available in the online journal.
APPENDIX A. OBSERVING LOGS FOR ISLANDS GALAXIESThis work is based on observations obtained with the ACS and WFC3 onboard the HST. These data were collectedin different runs for each galaxy over about 2 and 5.3 consecutive days between 2013 October 4 to 2015 September6 as part of a large HST proposal (GO-13028 and GO-13749, P.I.: E. Skillman). The observing sequence consistedof alternating ∼ http://archive.stsci.edu/ ),the date (column 2) and the UT start of each exposure (column 3) the filter used (column 4), and the exposure time(column 5).4 Mart´ınez-V´azquez et al.
Table A2 . Observing log for And II.Image Name Date UT Start Filter Exp. timename (YY-MM-DD) (hh:mm:ss) name (s)
ACS (RA=01:16:23.8, Dec=+33:26:05.5) jc1d01wfq flc.fits 2013-10-04 03:50:09 F475W 1280jc1d01whq flc.fits 2013-10-04 04:14:26 F814W 987jc1d01x5q flc.fits 2013-10-04 05:21:50 F814W 1100jc1d01x9q flc.fits 2013-10-04 05:43:08 F475W 1359jc1d02ycq flc.fits 2013-10-04 10:12:02 F475W 1280... ... ... ... ...
WFC3 (RA=01:16:04.4, Dec=+33:21:31.7) ic1d01wgq flc.fits 2013-10-04 03:49:14 F475W 1350ic1d01wiq flc.fits 2013-10-04 04:14:12 F814W 1122ic1d01x6q flc.fits 2013-10-04 05:21:45 F814W 1200ic1d01xbq flc.fits 2013-10-04 05:44:19 F475W 1409ic1d02ydq flc.fits 2013-10-04 10:11:07 F475W 1350... ... ... ... ...
This table is a portion of its entirely form which will be available in the online journal.
Table A3 . Observing log for And III.Image Name Date UT Start Filter Exp. timename (YY-MM-DD) (hh:mm:ss) name (s)
ACS (RA=00:35:30.7, Dec=+36:30:14.2) jcnb12c4q flc.fits 2014-11-24 05:33:55 F475W 1264jcnb12c7q flc.fits 2014-11-24 05:57:56 F814W 1002jcnb12c9q flc.fits 2014-11-24 06:56:08 F814W 1086jcnb12cdq flc.fits 2014-11-24 07:17:12 F475W 1372jcnb13cpq flc.fits 2014-11-24 10:20:31 F475W 1264... ... ... ... ...
WFC3 (RA= 00:35:51.5, Dec=+36:25:48.5) icnb12c5q flc.fits 2014-11-24 05:33:00 F475W 1308icnb12c6q flc.fits 2014-11-24 05:57:16 F814W 1046icnb12caq flc.fits 2014-11-24 06:56:03 F814W 1103icnb12ceq flc.fits 2014-11-24 07:17:00 F475W 1389icnb13cqq flc.fits 2014-11-24 10:19:36 F475W 1308... ... ... ... ...
This table is a portion of its entirely form which will be available in the online journal. ariable stars in ISLAndS galaxies Table A4 . Observing log for And XV.Image Name Date UT Start Filter Exp. timename (YY-MM-DD) (hh:mm:ss) name (s)
ACS (RA=01:14:18.7, Dec=+38:07:03.0) jcnb23w3q flc.fits 2014-09-17 11:23:38 F475W 1264jcnb23w6q flc.fits 2014-09-17 11:47:39 F814W 1002jcnb23w8q flc.fits 2014-09-17 12:53:44 F814W 1086jcnb23wcq flc.fits 2014-09-17 13:14:48 F475W 1372jcnb24azq flc.fits 2014-09-18 16:05:20 F475W 1264... ... ... ... ...
WFC3 (RA=01:13:50.3, Dec=+38:04:37.3) icnb23w4q flc.fits 2014-09-17 11:22:43 F475W 1308icnb23w5q flc.fits 2014-09-17 11:46:59 F814W 1046icnb23w9q flc.fits 2014-09-17 12:53:39 F814W 1103icnb23wdq flc.fits 2014-09-17 13:14:36 F475W 1389icnb24b0q flc.fits 2014-09-18 16:04:25 F475W 1308... ... ... ... ...
This table is a portion of its entirely form which will be available in the online journal.
Table A5 . Observing log for And XVI.Image Name Date UT Start Filter Exp. timename (YY-MM-DD) (hh:mm:ss) name (s)
ACS (RA=00:59:32.3, Dec=+32:23:38.9) jc1d09upq 1.fits 2013-11-20 12:46:13 F475W 1280jc1d09urq 1.fits 2013-11-20 13:10:30 F814W 987jc1d09uuq 1.fits 2013-11-20 14:13:37 F814W 1100jc1d09uyq 1.fits 2013-11-20 14:34:55 F475W 1359jc1d10wdq 1.fits 2013-11-20 23:55:40 F475W 1280... ... ... ... ...
WFC3 (RA=00:59:48.6, Dec=+32:18:37.2) ic1d09uqq flc.fits 2013-11-20 12:45:18 F475W 1350ic1d09usq flc.fits 2013-11-20 13:10:16 F814W 1122ic1d09uvq flc.fits 2013-11-20 14:13:32 F814W 1200ic1d09v0q flc.fits 2013-11-20 14:36:06 F475W 1409ic1d10weq flc.fits 2013-11-20 23:54:45 F475W 1350... ... ... ... ...
This table is a portion of its entirely form which will be available in the online journal. Mart´ınez-V´azquez et al.
Table A6 . Observing log for And XXVIII.Image Name Date UT Start Filter Exp. timename (YY-MM-DD) (hh:mm:ss) name (s)
ACS (RA=22:32:41.2, Dec=+31:12:58.2) jcnb31psq flc.fits 2015-01-20 23:57:41 F475W 1264jcnb31pvq flc.fits 2015-01-21 00:21:42 F814W 1002jcnb31qoq flc.fits 2015-01-21 01:19:28 F814W 1086jcnb31qsq flc.fits 2015-01-21 01:40:32 F475W 1372jcnb32rnq flc.fits 2015-01-21 04:44:14 F475W 1264jcnb32rqq flc.fits 2015-01-21 05:08:15 F814W 1002... ... ... ... ...
WFC3 (RA=22:33:09.6, Dec=+31:13:31.0) icnb31ptq flc.fits 2015-01-20 23:56:46 F475W 1308icnb31puq flc.fits 2015-01-21 00:21:02 F814W 1046icnb31qpq flc.fits 2015-01-21 01:19:23 F814W 1103icnb31qtq flc.fits 2015-01-21 01:40:20 F475W 1389icnb32roq flc.fits 2015-01-21 04:43:19 F475W 1308... ... ... ... ...
This table is a portion of its entirely form which will be available in the online journal. ariable stars in ISLAndS galaxies Table B7 . Photometry of the variable stars in And I dSph.
MHJD ∗ F W σ F W MHJD ∗ F W σ F W AndI-V00157266.425781 25.254 0.089 57266.441406 24.713 0.09057266.500000 25.523 0.056 57266.484375 24.417 0.11957266.691406 25.801 0.041 57266.707031 24.953 0.06757266.765625 25.917 0.064 57266.750000 25.028 0.09357267.417969 25.175 0.036 57267.433594 24.364 0.04357267.496094 25.118 0.036 57267.480469 24.549 0.06457267.683594 25.733 0.062 57267.699219 24.876 0.05457267.761719 25.863 0.052 57267.746094 25.007 0.12757268.281250 25.768 0.069 57268.296875 24.734 0.05357268.355469 25.832 0.063 57268.339844 24.781 0.08757268.890625 25.811 0.054 57268.824219 24.701 0.09557269.417969 25.788 0.064 57268.875000 24.767 0.08057269.339844 25.706 0.058 57269.355469 24.708 0.07957269.539062 25.700 0.072 57269.402344 24.811 0.06257268.808594 25.720 0.052 57269.554688 24.983 0.06257269.617188 26.001 0.065 57269.601562 24.958 0.11157270.464844 25.608 0.063 57270.480469 24.784 0.08957270.542969 25.681 0.088 57270.527344 24.797 0.08257270.664062 25.856 0.061 57270.679688 24.860 0.06257270.742188 26.026 0.062 57270.726562 25.066 0.10457271.660156 25.765 0.058 57271.675781 24.759 0.05557271.738281 25.904 0.033 57271.718750 24.856 0.078 ∗ Modified Heliocentric Julian Date of mid-exposure: HJD - 2,400,000(This table is a portion of its entirely form which will be available in the online journal.)
Table B8 . Photometry of the variable stars in And II dSph.
MHJD ∗ F W σ F W MHJD ∗ F W σ F W AndII-V00156569.171875 24.838 0.031 56569.187500 24.319 0.04756569.253906 24.822 0.051 56569.234375 24.196 0.04756569.437500 25.382 0.041 56569.453125 24.654 0.06156569.519531 24.785 0.069 56569.500000 24.326 0.05956570.101562 25.326 0.042 56570.117188 24.509 0.05856570.183594 24.775 0.048 56570.164062 24.268 0.04756570.500000 24.833 0.036 56570.515625 24.399 0.05156570.582031 24.849 0.050 56570.562500 24.332 0.03656570.699219 25.400 0.042 56570.714844 24.593 0.05156570.781250 25.339 0.062 56570.761719 24.677 0.08456570.898438 24.797 0.043 56570.914062 24.285 0.05156570.980469 25.153 0.048 56570.960938 24.457 0.04956571.097656 25.369 0.083 56571.113281 24.569 0.07956571.179688 24.772 0.052 56571.160156 24.300 0.06256571.562500 24.790 0.064 56571.578125 24.393 0.05556571.644531 25.160 0.055 56571.625000 24.396 0.05656571.695312 25.362 0.067 56571.710938 24.548 0.058 ∗ Modified Heliocentric Julian Date of mid-exposure: HJD - 2,400,000(This table is a portion of its entirely form which will be available in the online journal.) B. TIME SERIES OF VARIABLE STARS IN ISLANDS GALAXIESThe individual F475W and F814W measurements for all of the variables found in each galaxy of this work are listedin Tables B7, B8, B9, B10, B11, and B12, respectively.8
Mart´ınez-V´azquez et al.
Table B9 . Photometry of the variable stars in And III dSph.
MHJD ∗ F W σ F W MHJD ∗ F W σ F W AndIII-V00156985.244721 25.181 0.034 56985.259883 24.593 0.03656985.317071 25.254 0.021 56985.300786 24.459 0.03556985.443750 25.538 0.047 56985.458912 24.765 0.03856985.516204 25.525 0.048 56985.499919 24.701 0.06256985.642767 25.035 0.040 56985.657929 24.488 0.04856985.721413 24.881 0.149 56985.705128 24.540 0.03856986.239807 25.663 0.046 56986.254969 24.688 0.05956986.312643 25.630 0.029 56986.296358 24.256 0.17356987.168353 24.976 0.086 56987.183515 24.517 0.05356987.241710 25.183 0.023 56987.225425 24.404 0.04456987.499975 25.635 0.047 56987.515137 24.716 0.04656987.573493 25.028 0.034 56987.557209 24.472 0.03856988.229862 25.751 0.036 56988.245024 24.743 0.03456988.303346 25.523 0.029 56988.287061 24.663 0.02756988.428810 25.330 0.035 56988.443972 24.534 0.04356988.502386 25.340 0.038 56988.486102 24.496 0.04856989.158280 25.062 0.021 56989.173442 24.382 0.04756989.232169 25.009 0.072 56989.215885 24.380 0.05056989.357228 25.337 0.045 56989.372390 24.719 0.03056989.431186 25.660 0.044 56989.414902 24.680 0.05956989.630215 25.272 0.038 56989.613930 24.464 0.04656989.556175 25.420 0.063 56989.571337 24.501 0.065 ∗ Modified Heliocentric Julian Date of mid-exposure: HJD - 2,400,000(This table is a portion of its entirely form which will be available in the online journal.)
Table B10 . Photometry of the variable stars in And XV dSph.
MHJD ∗ F W σ F W MHJD ∗ F W σ F W AndXV-V00156917.486432 24.553 0.033 56917.501594 24.281 0.03656917.564258 25.095 0.039 56917.547973 24.378 0.03056918.682087 25.631 0.055 56918.697250 24.707 0.04856918.758837 25.772 0.053 56918.742552 24.872 0.06756918.881259 25.826 0.056 56918.896422 24.939 0.06456918.957951 25.254 0.098 56918.941666 24.810 0.05756919.476714 24.797 0.034 56919.491877 24.240 0.03856919.555350 24.883 0.033 56919.539065 24.342 0.03556919.678374 25.536 0.037 56919.693537 24.710 0.06156919.754476 25.851 0.040 56919.738191 24.777 0.05256919.877581 25.733 0.064 56919.892743 24.902 0.07456919.953659 25.653 0.045 56919.937374 25.007 0.05656920.541683 24.732 0.057 56920.556846 24.369 0.06356920.617634 25.256 0.039 56920.601349 24.489 0.05556920.740890 25.675 0.043 56920.756052 24.853 0.05656920.816829 25.835 0.069 56920.800544 24.877 0.05756920.883382 25.712 0.046 56920.867293 24.903 0.054 ∗ Modified Heliocentric Julian Date of mid-exposure: HJD - 2,400,000(This table is a portion of its entirely form which will be available in the online journal.) ariable stars in ISLAndS galaxies Table B11 . Photometry of the variable stars in And XVI dSph.
MHJD ∗ F W σ F W MHJD ∗ F W σ F W AndXVI-V00156616.545139 25.496 0.025 56616.560301 24.366 0.15356616.621076 25.675 0.034 56616.604792 24.668 0.04256617.010037 25.030 0.024 56617.025199 24.462 0.05656617.085986 25.392 0.022 56617.069701 24.566 0.02956617.408499 25.885 0.050 56617.423673 24.846 0.05456617.484483 25.923 0.063 56617.468199 24.823 0.05756617.674172 25.619 0.043 56617.689334 24.660 0.05456617.752522 25.330 0.034 56617.736660 24.650 0.04456618.006245 25.789 0.046 56618.021407 24.766 0.06956618.082218 25.945 0.049 56618.065933 24.916 0.05256618.471143 25.801 0.047 56618.485872 24.788 0.05756618.547139 25.841 0.067 56618.530854 24.764 0.06656618.598771 25.825 0.048 56618.615056 24.860 0.053 ∗ Modified Heliocentric Julian Date of mid-exposure: HJD - 2,400,000(This table is a portion of its entirely form which will be available in the online journal.)
Table B12 . Photometry of the variable stars in And XXVIII dSph.
MHJD ∗ F W σ F W MHJD ∗ F W σ F W AndXXVIII-V00157043.010749 25.951 0.110 57043.025911 24.779 0.06057043.082797 25.813 0.054 57043.066513 24.845 0.03157043.209741 25.053 0.062 57043.224903 24.318 0.05257043.281893 24.992 0.038 57043.265608 24.327 0.03957044.204663 25.908 0.053 57044.219825 24.515 0.07557044.277290 25.903 0.043 57044.261005 24.683 0.17557044.337289 25.878 0.077 57044.352451 24.875 0.04657044.412127 26.123 0.084 57044.395842 24.798 0.04657045.007156 25.852 0.091 57044.949321 24.892 0.06757044.934159 25.758 0.102 57044.990872 24.887 0.07057045.066797 26.149 0.065 57045.081959 24.925 0.07557045.139852 25.151 0.055 57045.123567 24.522 0.07557045.928885 — — 57045.944046 — —57046.002310 — — 57045.986025 — —57046.061499 25.748 0.054 57046.076661 24.674 0.05457046.134994 25.818 0.042 57046.118709 24.739 0.06357046.923494 25.931 0.072 57046.938656 24.971 0.06957046.997347 25.977 0.071 57046.981063 24.940 0.06757047.056108 25.098 0.043 57047.071270 24.255 0.07657047.130008 25.142 0.042 57047.113724 24.387 0.040 ∗ Modified Heliocentric Julian Date of mid-exposure: HJD - 2,400,000(This table is a portion of its entirely form which will be available in the online journal.) Mart´ınez-V´azquez et al.
Table C13 . Cross-identification with the Pritzl et al. catalog of variable stars in And I. ID Pritzl
Period
Pritzl ID This work
Period
This work
NotesV1 1.630 — — near to the BSS region; affected by near saturated field star; not variable in our dataV2 0.348 AndI-V185 0.349V3 0.412 AndI-V182 0.386V4 9.999 AndI-V172 0.607V5 0.654 AndI-V154 0.746V6 0.430 AndI-V186 0.429... ... ... ... ...V44 0.748 — — minor variation in F814; not variable in our dataV45 0.772 — — not variable in our data..... ... ... ... ...V89 0.523 — — in ACS gap... ... ... ... ...V98 0.625 — — in ACS gapV99 0.716 AndI-V237 0.630V100 0.782 — — not variable in our dataThe full table is available as Supporting Information with the online version of the paper.
Table C14 . Cross-identification with the Pritzl et al. catalog of variable stars in And II. ID Pritzl
Period
Pritzl ID This work
Period
This work
NotesV01 0.407 AndII-V080 0.370V02 0.546 AndII-V071 0.543V03 0.520 AndII-V098 0.516V04 0.540 AndII-V064 0.692V05 0.583 AndII-V081 0.580... ... ... ... ...V16 0.346 — — not variable in our data... ... ... ... ...V37 0.751 — — not variable in our data... ... ... ... ...V43 0.490 — — in ACS gap... ... ... ... ...V70 0.707 — — not variable in our dataV71 0.474 AndII-V153 0.580V72 0.469 AndII-V087 0.592V73 0.698 AndII-V118 0.538The full table is available as Supporting Information with the online version of the paper. C. COMPARISON WITH THE LITERATUREFigure C1 compares the periods of variable stars in common between our work and the published period for the threegalaxies already studied in the literature (And I, And II and And III Pritzl et al. 2004, 2005), according to the labelledsymbols. Pritzl’s data for each galaxy consist in two data-sets separated by 4-5 days. In each data-set, they startedobserving first all the F W and then all the F W images. The cadence of the data depends on the individualcase. For And I, the strategy was 3 × F W (4 × F W ) and 6 × F W (6 × F W ) in the first (second) set, with1 image per orbit. For And II, the data were collected as 3 × F W (4 × F W ) and 7 × F W (8 × F W ) in thefirst (second) set, with 2 images per orbit. For And III the strategy was similar to that for And II, but collecting4 × F W and 8 × F W in each data-set.A total of 94, 69 and 54 stars (out of 100, 73 and 56 in their catalogues) were recovered for And I, And II andAnd III, respectively. The cross-identifications made between our catalogues and Pritzl’s are shown in Tables C13,C14, C15. The small fraction of stars that were not recovered either appear as non variable in our data or fall in theACS gap (see the column “Notes” of Tables C13, C14, C15). We note that the matching was complicated by the factthat the coordinates listed in the Pritzl catalogs were significantly offset, with different offsets for each WFPC2 chip,in particular in the case of And III. ariable stars in ISLAndS galaxies Table C15 . Cross-identification with the Pritzl et al. catalog of variable stars in And III. ID Pritzl
Period
Pritzl ID This work
Period
This work
NotesV01 0.834 AndIII-V073 1.003V02 0.590 AndIII-V069 0.591V03 0.773 — — in ACS gapV04 0.629 AndIII-V065 0.559V05 0.650 AndIII-V067 0.632V06 0.678 AndIII-V105 0.818V07 0.480 AndIII-V075 0.428V08 1.510 — — not variable in our data... ... ... ... ...V53 0.534 AndIII-V039 0.533V54 0.623 AndIII-V041 0.625V55 0.599 AndIII-V031 0.596V56 0.496 AndIII-V046 0.640The full table is available as Supporting Information with the online version of the paper.
The comparison discloses a general good agreement (53% of the stars within a difference of 0.05 days and 80% within0.1 days) though with a few outliers for which the period is significantly discrepant (20% have a difference larger than0.1 day). However, this effect can be easily explained taking into account that HST observations may suffer froma strong aliasing introduced by the orbital time. By having more epochs and scheduling them to avoid redundantperiods, our program strongly suppressed the effects of aliasing.The aliasing effect is a very common error in signal treatment, as it happens whenever the original periodic signal(in our case, the light curve of the variable stars) is reconstructed using a discrete sampling. When a periodic signalof frequency f true is sampled with a frequency f sampling , the resulting number of cycles per sample is f true /f sampling (normalized frequency), and the samples are indistinguishable from those of another sinusoid (or periodic signal), calledan alias , whose normalized frequency differs from f true /f sampling by an integer (see e.g., VanderPlas 2017). Then, wecan express the aliases of frequency as f alias = | f true − N f sampling | , being N an integer. In this case, the calculationof the period can be affected by the HST orbital cadence of 96 minutes.The curves overplotted on Figure C1 represent how the true period is affected by a cadence of 96 minutes. Interest-ingly, most if not all the discrepant points are explained by this aliasing effect. Taking into account the limited numberof phase points in previous studies, and the optimized strategy of our observations, we suggest that has resulted inmore precise period determinations. Nevertheless, this comparison also supports the quality of the previous analysis,given the observational material available.2 Mart´ınez-V´azquez et al. P P r i t z l And IAnd IIAnd III0.2 0.4 0.6 0.8 1.0P
This work -0.3-0.2-0.10.00.10.20.3 D i ff e r en c e Figure C1 . Current period versus the period (top) and period difference (bottom) found by Pritzl et al. (2004, 2005) for the94, 69 and 54 stars matched in And I (black circles), And II (blue plusses) and And III (red open squares), respectively. Thedotted curves in the top panel are the aliasing lines (see text). We have taken the HST orbital period (96 minutes) as f sampling and | N | =[1,3,6,9,15,18] for obtaining these curves (aliasing lines). Note how the outliers follow the aliasing lines in most cases.This indicates the high probability that the offsets are due to aliasing. ariable stars in ISLAndS galaxies Table D16 . Parameters of the variable stars in And I dSph.
ID RA DEC Period h F W i A F W h F W i A F W h B i A B h V i A V h I i A I Typename (J2000) (J2000) (current)AndI-V001 0:45:09.233 +37:58:47.19 0.569 25.532 1.038 24.692 0.574 25.657 1.137 25.266 0.915 24.677 0.580 RRabAndI-V002 0:45:09.646 +37:59:48.86 0.567 25.505 0.590 24.737 0.358 25.613 0.669 25.251 0.436 24.719 0.360 RRabAndI-V003 0:45:09.819 +37:59:32.31 0.296 25.329 0.336 24.831 0.091 25.398 0.392 25.159 0.207 24.819 0.097 RRcAndI-V004 0:45:10.116 +37:58:44.43 0.598 25.322 1.390 24.607 0.699 25.415 1.605 25.103 0.949 24.599 0.669 RRabAndI-V005 0:45:10.429 +37:58:56.47 0.585 25.557 0.807 24.733 0.608 25.680 0.864 25.276 0.660 24.722 0.576 RRabAndI-V006 0:45:11.526 +37:58:45.53 0.349 25.442 1.054 24.806 0.508 25.531 1.120 25.228 0.875 24.790 0.518 RRdAndI-V007 0:45:12.020 +38:00:25.42 0.352 25.404 0.543 24.762 0.254 25.488 0.603 25.196 0.436 24.748 0.250 RRdAndI-V008 0:45:12.286 +37:58:45.04 0.791 25.338 0.304 24.450 0.243 25.478 0.321 25.023 0.263 24.421 0.273 RRabAndI-V009 0:45:13.494 +38:00:57.92 0.353 25.364 0.479 24.714 0.210 25.455 0.545 25.142 0.354 24.699 0.214 RRcAndI-V010 0:45:13.931 +37:59:21.84 0.581 25.273 1.393 24.614 0.658 25.357 1.550 25.096 0.999 24.611 0.617 RRabAndI-V011 0:45:14.260 +37:58:40.41 0.479 25.476 1.252 24.849 0.947 25.605 1.300 25.230 1.078 24.836 0.904 RRabAndI-V012 0:45:14.735 +38:00:35.23 0.623 25.172 1.365 24.434 0.656 25.302 1.483 24.921 1.069 24.421 0.647 RRabAndI-V013 0:45:15.057 +38:00:50.11 0.515 25.423 1.213 24.688 0.731 25.522 1.350 25.194 0.845 24.681 0.734 RRabAndI-V014 0:45:15.796 +37:59:31.21 0.359 25.326 0.484 24.685 0.235 25.420 0.530 25.106 0.381 24.670 0.234 RRcAndI-V015 0:45:16.130 +38:01:24.28 0.703 25.345 0.422 24.430 0.208 25.486 0.463 25.037 0.377 24.416 0.214 RRabAndI-V016 0:45:16.173 +37:59:27.10 0.644 25.458 0.576 24.553 0.329 25.606 0.634 25.139 0.478 24.538 0.338 RRabAndI-V017 0:45:16.178 +37:58:57.24 0.640 25.508 0.420 24.633 0.124 25.640 0.489 25.213 0.317 24.622 0.141 RRabAndI-V018 0:45:16.711 +37:59:41.10 0.301 25.213 0.212 24.726 0.135 25.290 0.222 25.033 0.174 24.718 0.139 RRcAndI-V019 0:45:16.813 +37:59:09.94 0.564 25.381 1.011 24.679 0.677 25.475 1.142 25.167 0.722 24.666 0.582 RRabAndI-V020 0:45:17.035 +38:00:02.48 0.574 25.453 1.113 24.692 0.536 25.560 1.257 25.216 0.836 24.685 0.525 RRabStars from “AndI-V001” to “AndI-V0038” were detected in the WFC3 field, while stars from “AndI-V039” to “AndI-V314” were detected inthe ACS field.Full version are available as Supporting Information with the online version of the paper. D. PULSATION PROPERTIES OF VARIABLE STARS IN ISLANDS GALAXIESThe properties of the variable stars found in this work for And I, II, III, XV, XVI, and XXVIII are detailed inTables D16, D17, D18, D19, D20, and D21, respectively. The first columns give the identification number and thenext two list the equatorial coordinates (J2000.0). Column 4 give the period of the variable in days, while columns 5to 14 list the intensity-averaged magnitudes and amplitude in the filters F F B , V , and I , respectively. Lastcolumn displays the variable type. REFERENCES Antoja, T., et al. 2015, MNRAS, 453, 541Arellano Ferro, A., Giridhar, S., Rojas L´opez, V., Figuera, R.,Bramich, D. M., & Rosenzweig, P. 2008, RMxAA, 44, 365Baker, M., & Willman, B. 2015, AJ, 150, 160Baldacci, L., Rizzi, L., Clementini, G., & Held, E. V. 2005,A&A, 431, 1189Bellazzini, M. 2008, Mem. Soc. Astron. Italiana, 79, 440Bellazzini, M., et al. 2011, A&A, 527, A58Bellazzini, M., Ferraro, F. R., Origlia, L., Pancino, E., Monaco,L., & Oliva, E. 2002, AJ, 124, 3222Bellazzini, M., Ferraro, F. R., & Pancino, E. 2001, ApJ, 556, 635Bellazzini, M., Ferraro, F. R., Sollima, A., Pancino, E., &Origlia, L. 2004, A&A, 424, 199Bernard, E. J., et al. 2008, ApJL, 678, L21Bernard, E. J., et al. 2010, ApJ, 712, 1259Bernard, E. J., et al. 2009, ApJ, 699, 1742Bernard, E. J., et al. 2013, MNRAS, 432, 3047Bono, G., Caputo, F., Cassisi, S., Incerpi, R., & Marconi, M.1997, ApJ, 483, 811Bono, G., Caputo, F., Castellani, V., Marconi, M., & Storm, J.2001, MNRAS, 326, 1183Bono, G., Caputo, F., Castellani, V., Marconi, M., Storm, J., &Degl’Innocenti, S. 2003, MNRAS, 344, 1097 Braga, V. F., et al. 2015, ApJ, 799, 165Braga, V. F., et al. 2016, ArXiv e-printsCacciari, C., Corwin, T. M., & Carney, B. W. 2005, AJ, 129, 267Caputo, F., Cassisi, S., Castellani, M., Marconi, G., &Santolamazza, P. 1999, AJ, 117, 2199Caputo, F., Castellani, V., Marconi, M., & Ripepi, V. 2000,MNRAS, 316, 819Carretta, E., Bragaglia, A., Gratton, R., D’Orazi, V., &Lucatello, S. 2009, A&A, 508, 695Cassisi, S., Potekhin, A. Y., Pietrinferni, A., Catelan, M., &Salaris, M. 2007, ApJ, 661, 1094Cassisi, S., & Salaris, M. 2013, Old Stellar Populations: How toStudy the Fossil Record of Galaxy FormationCastellani, V., & degl’Innocenti, S. 1995, A&A, 298, 827Catelan, M. 2009, Ap&SS, 320, 261Catelan, M., & Smith, H. A. 2015, Pulsating StarsClementini, G., Gratton, R., Bragaglia, A., Carretta, E., DiFabrizio, L., & Maio, M. 2003, AJ, 125, 1309Clementini, G., et al. 2016, A&A, 595, A133Cole, A. A., et al. 2017, ApJ, 837, 54Collins, M. L. M., et al. 2015, ApJL, 799, L13Conn, A. R., et al. 2012, ApJ, 758, 11Coppola, G., et al. 2015, ApJ, 814, 71 Mart´ınez-V´azquez et al.
Table D17 . Parameters of the variable stars in And II dSph.
ID RA DEC Period h F W i A F W h F W i A F W h B i A B h V i A V h I i A I Typename (J2000) (J2000) (current)AndII-V001 1:15:56.325 +33:21:19.95 0.332 25.061 0.686 24.441 0.336 25.149 0.747 24.848 0.563 24.425 0.330 RRcAndII-V002 1:15:57.224 +33:21:32.52 0.601 25.064 1.016 24.272 0.362 25.310 0.948 24.785 0.685 24.239 0.325 RRabAndII-V003 1:15:58.620 +33:21:29.91 0.625 25.115 0.766 24.243 0.375 25.257 0.886 24.811 0.613 24.231 0.383 RRabAndII-V004 1:15:58.951 +33:21:14.99 0.663 25.124 0.339 24.234 0.158 25.258 0.378 24.828 0.294 24.221 0.180 RRabAndII-V005 1:15:58.986 +33:21:05.31 0.769 24.971 0.561 24.102 0.297 25.109 0.620 24.667 0.470 24.088 0.308 RRabAndII-V006 1:15:59.033 +33:21:08.26 0.590 24.957 1.231 24.204 0.557 25.064 1.356 24.700 1.008 24.140 0.573 RRabAndII-V007 1:15:59.056 +33:20:45.97 0.606 24.927 1.551 24.194 0.791 25.033 1.692 24.670 1.306 24.180 0.792 RRabAndII-V008 1:15:59.377 +33:21:57.11 0.622 25.128 0.762 24.267 0.380 25.261 0.836 24.845 0.644 24.271 0.348 RRabAndII-V009 1:15:59.439 +33:21:26.87 0.345 24.990 0.655 24.378 0.339 25.080 0.705 24.768 0.554 24.363 0.336 RRcAndII-V010 1:16:00.025 +33:20:38.76 0.388 26.014 0.174 24.680 0.252 26.276 0.219 25.551 0.161 24.699 0.236 EBAndII-V011 1:16:01.326 +33:22:38.74 0.612 25.053 0.540 24.218 0.319 25.184 0.570 24.763 0.501 24.207 0.332 RRabAndII-V012 1:16:01.426 +33:20:39.01 0.640 24.976 0.540 24.125 0.267 25.108 0.593 24.686 0.431 24.109 0.273 RRabAndII-V013 1:16:02.447 +33:20:24.48 0.357 25.041 0.452 24.378 0.291 25.140 0.507 24.804 0.358 24.356 0.296 RRdAndII-V014 1:16:02.573 +33:22:07.27 0.621 25.124 0.521 24.366 0.259 25.245 0.578 24.844 0.417 24.337 0.239 RRabAndII-V015 1:16:02.657 +33:23:07.14 0.570 25.157 1.160 24.493 0.400 25.266 1.317 24.919 0.885 24.478 0.392 RRabAndII-V016 1:16:03.585 +33:22:15.65 0.556 24.914 1.441 24.419 0.683 24.993 1.554 24.757 1.123 24.412 0.642 RRabAndII-V017 1:16:04.235 +33:20:58.80 0.347 24.825 0.194 24.293 0.061 24.897 0.217 24.647 0.153 24.267 0.080 RRcAndII-V018 1:16:04.463 +33:22:25.54 0.572 25.082 1.090 24.397 0.599 25.176 1.185 24.857 0.895 24.457 0.388 RRabAndII-V019 1:16:04.471 +33:22:06.01 0.641 25.065 0.845 24.331 0.402 25.171 0.952 24.815 0.698 24.316 0.395 RRabAndII-V020 1:16:05.220 +33:19:59.10 0.751 25.048 0.546 24.139 0.296 25.187 0.600 24.743 0.459 24.126 0.303 RRabStars from “AndII-V001” to “AndII-V0035” were detected in the WFC3 field, while stars from “AndII-V036” to “AndII-V260” were detectedin the ACS field.Full version are available as Supporting Information with the online version of the paper.
Table D18 . Parameters of the variable stars in And III dSph.
ID RA DEC Period h F W i A F W h F W i A F W h B i A B h V i A V h I i A I Typename (J2000) (J2000) (current)AndIII-V001 0:35:22.101 +36:29:14.19 0.400 25.357 0.568 24.585 0.299 25.486 0.644 25.079 0.481 24.570 0.303 RRdAndIII-V002 0:35:22.192 +36:29:30.74 0.645 25.410 0.710 24.493 0.467 25.551 0.781 25.107 0.591 24.490 0.445 RRabAndIII-V003 0:35:22.642 +36:31:02.97 0.655 25.325 0.973 24.477 0.506 25.461 1.071 25.030 0.730 24.464 0.510 RRabAndIII-V004 0:35:22.827 +36:31:49.08 0.626 25.462 0.470 24.587 0.269 25.599 0.518 25.161 0.392 24.573 0.273 RRabAndIII-V005 0:35:23.254 +36:30:04.01 0.399 25.257 0.545 24.510 0.335 25.376 0.595 24.991 0.450 24.492 0.330 RRcAndIII-V006 0:35:23.687 +36:31:51.12 0.627 25.420 0.752 24.562 0.387 25.557 0.814 25.123 0.672 24.548 0.394 RRabAndIII-V007 0:35:23.985 +36:31:11.84 0.706 25.167 0.765 24.290 0.381 25.315 0.819 24.849 0.681 24.275 0.387 RRabAndIII-V008 0:35:24.109 +36:31:14.42 0.653 25.390 0.430 24.465 0.280 25.540 0.439 25.065 0.423 24.452 0.288 RRabAndIII-V009 0:35:24.208 +36:31:05.28 0.375 25.363 0.462 24.611 0.230 25.474 0.523 25.115 0.372 24.594 0.232 RRdAndIII-V010 0:35:24.240 +36:30:03.13 0.606 25.311 0.825 24.476 0.406 25.439 0.906 25.026 0.630 24.466 0.439 RRabAndIII-V011 0:35:24.529 +36:30:22.74 0.607 25.313 0.915 24.484 0.512 25.446 0.976 25.019 0.764 24.469 0.521 RRabAndIII-V012 0:35:25.360 +36:29:24.55 0.601 25.407 0.779 24.489 0.429 25.553 0.862 25.099 0.652 24.474 0.442 RRabAndIII-V013 0:35:25.445 +36:30:56.36 0.650 25.384 0.263 24.496 0.153 25.525 0.284 25.074 0.225 24.473 0.143 RRabAndIII-V014 0:35:25.888 +36:29:46.85 0.646 25.276 0.519 24.372 0.396 25.417 0.568 24.967 0.460 24.370 0.329 RRabAndIII-V015 0:35:26.072 +36:31:35.04 0.661 25.302 0.666 24.436 0.301 25.432 0.742 25.012 0.594 24.425 0.324 RRabAndIII-V016 0:35:26.112 +36:29:53.61 0.614 25.343 0.949 24.510 0.453 25.463 1.073 25.053 0.776 24.504 0.498 RRabAndIII-V017 0:35:26.231 +36:30:26.37 0.406 25.235 0.499 24.481 0.200 25.349 0.557 24.979 0.391 24.465 0.207 RRdAndIII-V018 0:35:26.302 +36:30:44.48 0.413 25.269 0.490 24.497 0.288 25.383 0.528 25.000 0.424 24.479 0.295 RRdAndIII-V019 0:35:26.384 +36:30:24.06 0.328 25.353 0.622 24.718 0.302 25.453 0.646 25.117 0.569 24.700 0.299 RRcAndIII-V020 0:35:26.533 +36:30:51.13 0.406 25.239 0.422 24.483 0.210 25.355 0.471 24.978 0.341 24.465 0.209 RRdStars from “AndIII-V001” to “AndIII-V114” were detected in the ACS field, while stars from “AndIII-V115” to “AndIII-V118” were detectedin the WFC3 field.Full version are available as Supporting Information with the online version of the paper. ariable stars in ISLAndS galaxies Table D19 . Parameters of the variable stars in And XV dSph.
ID RA DEC Period h F W i A F W h F W i A F W h B i A B h V i A V h I i A I Typename (J2000) (J2000) (current)AndXV-V001 1:14:10.037 +38:06:34.23 0.503 25.366 1.443 24.657 0.761 25.469 1.552 25.128 1.215 24.643 0.760 RRabAndXV-V002 1:14:10.426 +38:06:37.79 0.618 25.364 0.861 24.523 0.458 25.494 0.948 25.071 0.707 24.508 0.465 RRabAndXV-V003 1:14:10.438 +38:06:38.32 0.587 25.312 1.015 24.541 0.633 25.429 1.089 25.047 0.876 24.524 0.640 RRabAndXV-V004 1:14:11.397 +38:06:27.19 1.136 23.779 1.574 23.104 0.828 23.875 1.697 23.558 1.319 23.093 0.819 ACAndXV-V005 1:14:11.724 +38:06:47.28 0.547 25.335 1.217 24.546 0.808 25.440 1.359 25.083 0.996 24.530 0.833 RRabAndXV-V006 1:14:11.949 +38:06:51.41 0.377 25.235 0.602 24.599 0.433 25.329 0.665 25.014 0.488 24.581 0.434 RRcAndXV-V007 1:14:12.105 +38:07:17.08 0.629 25.320 0.671 24.470 0.385 25.449 0.736 25.032 0.565 24.455 0.393 RRabAndXV-V008 1:14:12.198 +38:06:54.71 0.329 25.246 0.605 24.632 0.393 25.332 0.664 25.038 0.482 24.610 0.390 RRcAndXV-V009 1:14:12.555 +38:06:08.23 0.365 25.305 0.493 24.659 0.240 25.404 0.526 25.074 0.416 24.641 0.239 RRcAndXV-V010 1:14:13.229 +38:07:23.41 0.576 25.356 0.452 24.537 0.351 25.475 0.482 25.083 0.404 24.521 0.355 RRabAndXV-V011 1:14:13.380 +38:06:34.38 0.609 25.360 1.069 24.522 0.558 25.490 1.164 25.074 0.905 24.513 0.544 RRabAndXV-V012 1:14:13.441 +38:07:03.68 0.543 25.256 0.949 24.537 0.415 25.360 1.032 25.006 0.722 24.521 0.422 RRabAndXV-V013 1:14:13.592 +38:07:44.73 0.608 25.244 1.430 24.481 0.668 25.356 1.613 24.983 1.140 24.468 0.672 RRabAndXV-V014 1:14:13.642 +38:06:04.53 0.518 25.359 1.305 24.685 0.669 25.457 1.428 25.135 1.039 24.673 0.665 RRabAndXV-V015 1:14:13.936 +38:06:04.72 0.621 25.385 0.631 24.559 0.347 25.512 0.698 25.101 0.515 24.544 0.354 RRabAndXV-V016 1:14:13.944 +38:07:35.31 0.868 24.331 1.406 23.554 0.776 24.436 1.550 24.084 1.146 23.536 0.734 ACAndXV-V017 1:14:14.008 +38:08:00.67 0.677 25.226 0.821 24.406 0.427 25.356 0.919 24.942 0.649 24.393 0.440 RRabAndXV-V018 1:14:14.094 +38:07:06.15 0.601 25.308 0.946 24.479 0.518 25.434 1.042 25.022 0.730 24.466 0.534 RRabAndXV-V019 1:14:14.115 +38:07:40.49 0.719 25.313 0.801 24.418 0.351 25.451 0.890 25.020 0.597 24.406 0.358 RRabAndXV-V020 1:14:14.371 +38:07:08.30 0.644 25.491 0.593 24.562 0.330 25.638 0.652 25.167 0.477 24.553 0.328 RRabAll stars were detected in the ACS field.Full version are available as Supporting Information with the online version of the paper.
Table D20 . Parameters of the variable stars in And XVI dSph.
ID RA DEC Period h F W i A F W h F W i A F W h B i A B h V i A V h I i A I Typename (J2000) (J2000) (current)AndXVI-V001 0:59:24.386 +32:22:33.16 0.622 25.459 1.142 24.642 0.555 25.584 1.281 25.173 0.912 24.627 0.571 M31 RRLAndXVI-V002 0:59:25.335 +32:22:16.10 0.358 24.563 0.624 23.877 0.402 24.671 0.692 24.306 0.484 23.857 0.394 RRcAndXVI-V003 0:59:27.972 +32:22:57.58 0.389 24.560 0.552 23.791 0.374 24.667 0.612 24.313 0.445 23.774 0.375 RRcAndXVI-V004 0:59:29.434 +32:22:25.90 0.350 24.580 0.520 23.900 0.346 24.682 0.547 24.346 0.467 23.882 0.347 RRcAndXVI-V005 0:59:30.846 +32:22:14.01 0.617 24.603 0.922 23.756 0.581 24.734 0.992 24.335 0.759 23.741 0.586 RRabAndXVI-V006 0:59:34.271 +32:21:59.44 0.640 24.594 1.199 23.746 0.657 24.725 1.296 24.280 0.894 23.730 0.671 RRabAndXVI-V007 0:59:36.072 +32:23:16.35 0.392 24.606 0.456 23.877 0.227 24.715 0.500 24.352 0.388 23.858 0.231 RRcAndXVI-V008 0:59:37.515 +32:22:10.10 0.289 24.669 0.298 24.163 0.198 24.738 0.312 24.495 0.264 24.147 0.195 RRcAndXVI-V009 0:59:38.101 +32:23:15.78 0.651 24.610 0.663 23.785 0.424 24.737 0.712 24.324 0.589 23.769 0.430 RRabAll stars were detected in the ACS field. Mart´ınez-V´azquez et al.
Table D21 . Parameters of the variable stars in And XXVIII dSph.
ID RA DEC Period h F W i A F W h F W i A F W h B i A B h V i A V h I i A I Typename (J2000) (J2000) (current)AndXXVIII-V001 22:32:32.098 +31:13:11.52 0.642 25.549 1.142 24.648 0.601 25.694 1.259 25.229 0.960 24.639 0.620 RRabAndXXVIII-V002 22:32:33.432 +31:13:07.56 0.608 25.392 0.697 24.659 0.228 25.508 0.590 25.141 0.392 24.615 0.203 RRL?AndXXVIII-V003 22:32:35.539 +31:12:15.41 0.407 25.414 0.460 24.603 0.217 25.536 0.517 25.131 0.366 24.587 0.222 RRcAndXXVIII-V004 22:32:35.594 +31:12:32.34 0.366 25.490 0.501 24.730 0.272 25.607 0.538 25.215 0.435 24.713 0.275 RRcAndXXVIII-V005 22:32:36.365 +31:12:22.37 0.565 25.529 0.974 24.647 0.566 25.653 1.102 25.244 0.769 24.635 0.582 RRabAndXXVIII-V006 22:32:36.682 +31:14:05.83 0.540 25.493 1.222 24.639 0.771 25.621 1.344 25.198 1.016 24.629 0.785 RRabAndXXVIII-V007 22:32:36.703 +31:13:45.74 0.681 25.363 0.720 24.456 0.345 25.502 0.802 25.055 0.606 24.445 0.355 RRabAndXXVIII-V008 22:32:36.958 +31:13:14.25 0.341 25.411 0.541 24.559 0.246 25.542 0.592 25.123 0.451 24.546 0.250 RRdAndXXVIII-V009 22:32:37.075 +31:12:45.87 0.362 25.453 0.480 24.715 0.396 25.567 0.508 25.196 0.459 24.698 0.403 RRdAndXXVIII-V010 22:32:37.332 +31:12:31.50 0.510 25.471 1.147 24.682 0.678 25.596 1.263 25.195 0.936 24.668 0.683 RRabAndXXVIII-V011 22:32:37.507 +31:12:31.46 0.366 25.465 0.493 24.722 0.334 25.572 0.554 25.213 0.390 24.706 0.331 RRdAndXXVIII-V012 22:32:37.510 +31:11:55.04 0.385 25.482 0.393 24.683 0.227 25.603 0.421 25.208 0.339 24.667 0.226 RRdAndXXVIII-V013 22:32:37.975 +31:13:40.09 0.366 25.394 0.541 24.613 0.341 25.508 0.583 25.134 0.478 24.596 0.346 RRcAndXXVIII-V014 22:32:37.980 +31:14:01.26 0.369 25.466 0.572 24.647 0.324 25.587 0.626 25.200 0.492 24.631 0.321 RRdAndXXVIII-V015 22:32:38.287 +31:14:04.83 0.646 25.412 0.790 24.488 0.301 25.565 0.910 25.092 0.524 24.403 0.483 RRabAndXXVIII-V016 22:32:38.614 +31:13:12.52 0.397 25.178 0.574 24.534 0.257 25.271 0.625 24.960 0.491 24.517 0.257 RRcAndXXVIII-V017 22:32:38.635 +31:13:34.34 0.558 25.351 1.367 24.601 0.771 25.461 1.476 25.100 1.138 24.585 0.778 RRabAndXXVIII-V018 22:32:38.690 +31:14:14.99 0.412 25.353 0.569 24.603 0.251 25.463 0.634 25.097 0.454 24.588 0.257 RRcAndXXVIII-V019 22:32:38.837 +31:13:12.22 0.524 25.455 1.076 24.678 0.608 25.571 1.156 25.170 0.809 24.662 0.710 RRabAndXXVIII-V020 22:32:39.029 +31:12:51.00 0.651 25.427 0.432 24.543 0.189 25.566 0.493 25.125 0.370 24.528 0.199 RRabAll stars were detected in the ACS field.Full version are available as Supporting Information with the online version of the paper.
Corwin, T. M., Carney, B. W., & Nifong, B. G. 1999, AJ, 118,2875Cusano, F., et al. 2013, ApJ, 779, 7Cusano, F., et al. 2015, ApJ, 806, 200Cusano, F., et al. 2016, ArXiv e-printsDa Costa, G. S., & Armandroff, T. E. 1990, AJ, 100, 162Da Costa, G. S., Armandroff, T. E., & Caldwell, N. 2002, AJ,124, 332Da Costa, G. S., Armandroff, T. E., Caldwell, N., & Seitzer, P.1996, AJ, 112, 2576Da Costa, G. S., Armandroff, T. E., Caldwell, N., & Seitzer, P.2000, AJ, 119, 705Da Costa, G. S., Rejkuba, M., Jerjen, H., & Grebel, E. K. 2010,ApJL, 708, L121Demarque, P., & Hirshfeld, A. W. 1975, ApJ, 202, 346Dolphin, A. E., et al. 2002, AJ, 123, 3154Ferguson, A. M. N., Irwin, M. J., Ibata, R. A., Lewis, G. F., &Tanvir, N. R. 2002, AJ, 124, 1452Fiorentino, G., et al. 2015, ApJL, 798, L12Fiorentino, G., Contreras Ramos, R., Tolstoy, E., Clementini,G., & Saha, A. 2012, A&A, 539, A138Fiorentino, G., Limongi, M., Caputo, F., & Marconi, M. 2006,A&A, 460, 155Fiorentino, G., & Monelli, M. 2012, A&A, 540, A102Fiorentino, G., et al. 2017, A&A, 599, A125Gallart, C., Aparicio, A., Freedman, W. L., Madore, B. F.,Mart´ınez-Delgado, D., & Stetson, P. B. 2004, AJ, 127, 1486Gallart, C., et al. 2015, ApJL, 811, L18Garofalo, A., et al. 2013, ApJ, 767, 62Gilbert, K. M., et al. 2009, ApJ, 705, 1275Grevesse, N., Asplund, M., & Sauval, A. J. 2007, SSRv, 130, 105Harbeck, D., et al. 2001, AJ, 122, 3092Hidalgo, S. L., et al. 2013, ApJ, 778, 103Hirshfeld, A. W. 1980, ApJ, 241, 111Ho, N., et al. 2012, ApJ, 758, 124 Horne, J. H., & Baliunas, S. L. 1986, ApJ, 302, 757Ibata, R., Irwin, M., Lewis, G., Ferguson, A. M. N., & Tanvir, N.2001, Nature, 412, 49Ibata, R., Martin, N. F., Irwin, M., Chapman, S., Ferguson,A. M. N., Lewis, G. F., & McConnachie, A. W. 2007, ApJ,671, 1591Jeffery, E. J., et al. 2011, AJ, 141, 171Kalirai, J. S., et al. 2010, ApJ, 711, 671Kinemuchi, K., Harris, H. C., Smith, H. A., Silbermann, N. A.,Snyder, L. A., La Cluyz´e, A. P., & Clark, C. L. 2008, AJ, 136,1921Kirby, E. N., Cohen, J. G., Guhathakurta, P., Cheng, L.,Bullock, J. S., & Gallazzi, A. 2013, ApJ, 779, 102Koenig, C. H. B., Nemec, J. M., Mould, J. R., & Fahlman, G. G.1993, AJ, 106, 1819Kuehn, C., et al. 2008, ApJL, 674, L81Kuehn, C. A., Smith, H. A., Catelan, M., Pritzl, B. J., De Lee,N., & Borissova, J. 2011, AJ, 142, 107Kunder, A., et al. 2013, AJ, 146, 119Layden, A. C., Ritter, L. A., Welch, D. L., & Webb, T. M. A.1999, AJ, 117, 1313Lee, M. G., Freedman, W. L., & Madore, B. F. 1993, ApJ, 417,553Lee, Y.-W. 1990, ApJ, 363, 159Letarte, B., et al. 2009, MNRAS, 400, 1472Madore, B. F. 1982, ApJ, 253, 575Madore, B. F., & Freedman, W. L. 1995, AJ, 109, 1645Marconi, M., et al. 2015, ApJ, 808, 50Marconi, M., Fiorentino, G., & Caputo, F. 2004, A&A, 417, 1101Martin, N. F., et al. 2016, ApJ, 833, 167Martin, N. F., et al. 2017, ArXiv e-printsMart´ınez-V´azquez, C. E., et al. 2015, MNRAS, 454, 1509Mart´ınez-V´azquez, C. E., et al. 2016a, MNRAS, 461, L41Mart´ınez-V´azquez, C. E., et al. 2016b, MNRAS, 462, 4349Mateo, M., Fischer, P., & Krzeminski, W. 1995, AJ, 110, 2166 ariable stars in ISLAndS galaxies37