STELLAR VELOCITIES IN THE CARINA, FORNAX, SCULPTOR, AND SEXTANS dSph GALAXIES: DATA FROM THE MAGELLAN/MMFS SURVEY^*

Dwarf spheroidal (dSph) galaxies are the smallest stellar systems thought to contain dark matter. Because they represent the lower observational extreme of both the galaxy-luminosity and halo-mass functions, dSphs are important objects with which to compare the models of galaxy formation. As dSphs typically lack neutral hydrogen, their pressure-supported stellar components provide the best available kinematic tracers. The dSph satellites of the Milky Way (MW) are sufficiently nearby that one can obtain high-resolution spectra of individual stars, enabling line-of-sight velocity measurements that resolve narrow (typically ∼5–10 km s⁻¹) dSph velocity dispersions. The first such study used velocity measurements of just three stars to argue that the Draco dSph, if in virial equilibrium, has a mass-to-light ratio M/L ⩾ 30 (solar units), indicative of a dominant dark matter component (Aaronson 1983).

Subsequent observations of Draco and other MW satellites supported the notion that dSph kinematics are dominated by dark matter. Stellar velocity samples containing measurements for tens of stars per galaxy showed that the most luminous dSphs (L_V ∼ 10^5–7L_V,⊙) all have central velocity dispersions of ∼10 km s⁻¹ (e.g., Aaronson & Olszewski 1987; Mateo et al. 1991, 1993; Suntzeff et al. 1993; Hargreaves et al. 1994; Armandroff et al. 1995; Hargreaves et al. 1996; Olszewski et al. 1995; Queloz et al. 1995; Vogt et al. 1995; Mateo et al. 1998). Given such data, simple kinematic models, which assume spherical symmetry, dynamic equilibrium, velocity isotropy, and radially constant M/L (i.e., mass follows light), imply that dSphs have masses of ∼10⁷ M_☉ and M/L ∼ 10¹–10². The absence of a correlation between dynamical mass and luminosity suggests that the large variation in M/L is solely a reflection of the variation in baryon content (Mateo et al. 1993, 1998; Gilmore et al. 2007; Walker et al. 2007b; Strigari et al. 2008).

The advent of high-resolution, multiobject spectrographs at large telescopes now makes it possible to gather spectra for hundreds of stars during a single night. With samples reaching hundreds of stars per dSph, velocity dispersion profiles are now available for all of the brighter MW dSphs (Kleyna et al. 2002, 2004; Wilkinson et al. 2004; Muñoz et al. 2005, 2006; Sohn et al. 2007; Walker et al. 2006a, 2006b, 2007b; Battaglia et al. 2006; Mateo et al. 2008; Koch et al. 2007b, 2007a). For the most luminous dSphs, it is now possible to build spectroscopic data sets for thousands of stars. With such large data sets, one can measure higher moments (e.g., kurtosis) of the velocity distribution, and thereby place observational constraints on orbital anisotropy (Łokas et al. 2005). The large data sets also provide information about the velocity distribution in two dimensions, and thus are capable of uncovering kinematic evidence of substructure (Kleyna et al. 2003; Walker et al. 2006b) as well as tidal streaming (Muñoz et al. 2006; Mateo et al. 2008). The ability to measure spectral line strengths from these low S/N spectra provides an extra dimension of information that can be used to study stellar metallicity distributions (e.g., Koch et al. 2006, 2007b) to identify correlations between kinematics and metallicity (Tolstoy et al. 2004; Battaglia et al. 2006), and to help clean samples of contaminating foreground stars.

In a previous paper (Walker et al. 2007a, Paper I hereafter), we introduced a spectroscopic survey of individual dSph stars, undertaken using the Michigan/MIKE Fiber System (MMFS) at the Magellan 6.5 m Clay Telescope at Las Campanas Observatory. Our MMFS spectra have resolution R∼ 20,000 and sample the region 5140–5180 Å, which contains the prominent magnesium-triplet absorption feature. Paper I describes MMFS as well as our procedures for target selection, observation, and data reduction. As of 2008 August, we have used MMFS to obtain 8855 spectra from 7103 stars in four dSphs: Carina, Fornax, Sculptor, and Sextans. From each spectrum, we measure the line-of-sight velocity and spectral indices that quantify the strengths of the iron and magnesium absorption lines present in the spectra (Paper I).

Here, we present the entire MMFS data set. In a companion paper (Walker et al. 2009, Paper III hereafter), we develop a statistical algorithm that uses the available velocity and magnesium data, as well as the stellar positions, to evaluate for each star the probability of dSph membership. Adding these probabilities, we find that the MMFS sample contains more than 5000 dSph members. In forthcoming papers, we use the MMFS data to provide detailed analyses of dSph kinematics and chemodynamic substructure.

We refer the reader to Paper I for a description of MMFS and the details of our methodology regarding target selection, observing procedure, and data reduction. After the publication of Paper I, we obtained new data during observing runs in 2007 January, 2007 September, 2008 April, and 2008 August. In these runs, we used MMFS to observe nine additional Carina fields (including five distinct sets of targets in the densely populated central field), three Sextans fields, eight Sculptor fields, and six Fornax fields. Table 1 logs these new observations (see Paper I for a log of all previous MMFS observations). The first two columns identify the galaxy and field number (identified in the maps of Figure 1). Columns 3–7 list the heliocentric Julian date (HJD) at the midpoint of the first exposure, UT date at the midpoint of the first exposure, total exposure time, number of red-giant candidates to which we assigned fibers, and the number of these for which we obtained an acceptable velocity measurement. The fraction of observed stars with acceptable measurements suffered in both 2007 runs due to harsh observing conditions. Figure 1 maps all dSph fields observed with MMFS as of 2008 August.

Zoom In Zoom Out Reset image size

Tables 2, 3, 4, and 5 present the MMFS spectroscopic data from each individual observation of targets in Carina, Fornax, Sculptor, and Sextans, respectively. For stars with multiple velocity measurements, results from repeat observations are listed directly beneath the first measurement. Otherwise, results are listed chronologically by time of observation. Column 1 identifies each target by galaxy and identification number. Column 2 gives the field number and spectrograph channel ("B" for blue, "R" for red) in which the star was observed. Column 3 lists the Heliocentric Julian Date of the observation. Columns 4–5 list equatorial coordinates (J2000.0). Columns 6–7 give the apparent V magnitude and V–I color, respectively⁴. Column 8 lists the measured velocity in the heliocentric rest frame (HRF). Columns 9–10 list composite spectral indices ΣFe and ΣMg, respectively.

Table 2. MMFS Spectroscopic Data–Carina

Target	Field	HJD −2.45 × 106	α2000 (hh:mm:ss)	δ2000 (dd:mm:ss)	V (mag)	V-I (mag)	Vhelio (km s−1)	ΣFe (Å)	ΣMg (Å)	$\hat{P}_{M}$	〈V〉helio (km s−1)	〈ΣMg〉 (Å)
Car-0001	18B	3087.533	06:42:17.94	−50:53:58.4	19.90	1.14	218.8 ± 2.0	0.42 ± 0.05	0.48 ± 0.10	1.000	219.0 ± 2.0	0.49 ± 0.10
	1R	4554.518					223.0 ± 10.3	0.21 ± 0.28	0.53 ± 0.30
Car-0002	18B	3087.533	06:42:18.60	−50:54:27.6	18.87	1.14	235.4 ± 0.6	0.38 ± 0.02	0.45 ± 0.05	1.000	235.5 ± 0.6	0.44 ± 0.04
	1R	3088.572					237.3 ± 5.8	0.43 ± 0.02	0.46 ± 0.09
	25R	3411.554					236.0 ± 6.2	0.37 ± 0.07	0.24 ± 0.15
	1R	4554.518					241.3 ± 4.0	0.43 ± 0.09	0.46 ± 0.16
Car-0003	18B	3087.533	06:42:25.74	−50:52:12.6	18.84	1.10	39.6 ± 0.9	0.39 ± 0.04	1.12 ± 0.03	0.000
Car-0004	18B	3087.533	06:42:27.17	−50:53:07.4	19.53	0.99	57.1 ± 1.9	0.43 ± 0.05	1.02 ± 0.06	0.000
Car-0005	18B	3087.533	06:42:26.13	−50:53:29.1	17.99	1.35	4.6 ± 2.1	0.40 ± 0.02	0.95 ± 0.03	0.000	6.0 ± 1.8	0.95 ± 0.03
	1R	4122.587					8.8 ± 3.1	...	...
Car-0006	18B	3087.533	06:42:27.95	−50:53:37.4	20.41	0.92	218.0 ± 3.4	...	0.87 ± 0.22	0.964
Car-0007	18B	3087.533	06:42:25.17	−50:54:01.8	20.42	1.02	216.2 ± 5.0	0.30 ± 0.23	0.57 ± 0.25	0.997
Car-0008	18B	3087.533	06:42:21.59	−51:00:21.4	20.47	0.84	225.1 ± 5.7	0.42 ± 0.08	0.31 ± 0.20	1.000	225.4 ± 3.1	0.24 ± 0.10
	13B	3410.547					225.6 ± 3.8	0.38 ± 0.05	0.21 ± 0.12
Car-0009	18B	3087.533	06:42:25.55	−50:59:35.8	19.43	1.05	232.1 ± 1.9	0.42 ± 0.04	0.41 ± 0.09	1.000	232.1 ± 1.9	0.41 ± 0.09
	1R	4123.555					232.9 ± 9.2	...	...
Car-0010	18B	3087.533	06:42:24.18	−50:56:12.6	18.48	1.20	223.2 ± 0.9	0.40 ± 0.02	0.39 ± 0.04	1.000	223.1 ± 0.8	0.39 ± 0.04
	25R	3411.554					222.5 ± 1.6	0.34 ± 0.04	0.38 ± 0.10
Car-0011	18B	3087.533	06:42:27.59	−50:55:54.0	20.50	0.80	214.5 ± 3.4	0.37 ± 0.07	0.35 ± 0.19	1.000	217.0 ± 2.3	0.44 ± 0.09
	13B	3410.547					219.1 ± 3.2	0.35 ± 0.06	0.47 ± 0.11
Car-0012	18B	3087.533	06:42:11.82	−50:57:13.1	19.94	1.00	229.5 ± 4.0	0.33 ± 0.05	0.39 ± 0.12	1.000
Car-0013	18B	3087.533	06:42:16.01	−50:56:54.8	19.65	0.96	219.1 ± 1.8	0.40 ± 0.03	0.35 ± 0.09	1.000	219.3 ± 1.7	0.35 ± 0.09
	1R	4122.587					226.7 ± 11.8	...	...
Car-0014	18B	3087.533	06:42:14.01	−50:56:58.3	20.62	0.95	232.6 ± 2.5	0.47 ± 0.07	0.42 ± 0.16	1.000
Car-0015	18B	3087.533	06:42:14.77	−50:56:21.5	20.13	1.08	220.3 ± 2.2	0.27 ± 0.07	0.52 ± 0.12	0.999	220.5 ± 2.2	0.52 ± 0.12
	1R	4557.518					224.8 ± 11.0	...	...
...

Only a portion of this table is shown here to demonstrate its form and content. Machine-readable and Virtual Observatory (VO) versions of the full table are available.

Download table as:  Machine-readable (MRT)Virtual Observatory (VOT)Typeset image

Table 3. MMFS Spectroscopic Data–Fornax

Target	Field	HJD −2.45 × 106	α2000 (hh:mm:ss)	δ2000 (dd:mm:ss)	V (mag)	V-I (mag)	Vhelio (km s−1)	ΣFe (Å)	ΣMg (Å)	$\hat{P}_{M}$	〈V〉helio (km s−1)	〈ΣMg〉 (Å)
For-0001	15B	3287.826	02:39:53.25	−34:34:35.8	19.27	1.35	51.4 ± 0.4	0.51 ± 0.02	0.65 ± 0.05	0.996
For-0002	15B	3287.826	02:39:56.10	−34:35:43.1	19.23	1.27	56.8 ± 0.5	0.51 ± 0.02	0.59 ± 0.05	0.997
For-0003	15B	3287.826	02:39:55.28	−34:37:27.2	19.06	1.24	53.4 ± 0.4	0.50 ± 0.03	0.59 ± 0.07	0.995
For-0004	15B	3287.826	02:39:55.69	−34:38:21.8	19.22	1.28	64.3 ± 0.6	0.56 ± 0.03	0.82 ± 0.05	0.985
For-0005	15B	3287.826	02:39:58.47	−34:38:51.5	18.97	1.17	54.9 ± 0.5	0.49 ± 0.02	0.55 ± 0.05	0.995
For-0006	15B	3287.826	02:39:55.36	−34:38:56.7	19.33	1.13	49.9 ± 0.8	0.49 ± 0.05	0.77 ± 0.08	0.992
For-0007	15B	3287.826	02:39:59.50	−34:33:25.7	19.24	1.25	70.4 ± 0.5	0.57 ± 0.03	0.64 ± 0.06	0.994
For-0008	15B	3287.826	02:40:01.52	−34:35:09.8	19.30	1.32	63.2 ± 0.5	0.62 ± 0.03	0.97 ± 0.05	0.926
For-0009	15B	3287.826	02:40:00.13	−34:35:16.3	19.38	1.19	51.6 ± 0.5	0.54 ± 0.03	0.70 ± 0.07	0.996
For-0010	15B	3287.826	02:39:58.41	−34:36:00.1	19.03	1.28	41.1 ± 0.4	0.53 ± 0.02	0.78 ± 0.04	0.985
For-0011	15B	3287.826	02:40:00.18	−34:36:55.3	19.18	1.17	71.0 ± 1.6	0.49 ± 0.05	0.50 ± 0.15	0.986
For-0012	15B	3287.826	02:40:07.09	−34:37:03.0	19.24	1.08	32.9 ± 1.2	0.35 ± 0.05	0.64 ± 0.10	0.966
For-0013	15B	3287.826	02:40:07.35	−34:37:56.7	19.06	1.26	54.5 ± 0.6	0.58 ± 0.03	0.65 ± 0.09	0.995
For-0014	15B	3287.826	02:40:05.29	−34:38:12.8	19.14	1.31	43.6 ± 1.0	⋅⋅⋅	⋅⋅⋅	0.983
For-0015	15B	3287.826	02:40:11.40	−34:42:50.8	19.22	1.13	79.0 ± 0.8	0.43 ± 0.04	0.43 ± 0.09	0.949	79.2 ± 0.8	0.42 ± 0.08
	592B	3666.797					83.6 ± 3.2	0.38 ± 0.08	0.38 ± 0.19
For-0016	15B	3287.826	02:40:14.65	−34:39:25.3	19.31	1.21	60.1 ± 0.8	0.47 ± 0.04	0.42 ± 0.09	0.990
For-0017	15B	3287.826	02:40:12.08	−34:38:47.8	19.25	1.34	59.4 ± 0.5	0.45 ± 0.03	0.57 ± 0.06	0.995
For-0018	15B	3287.826	02:40:23.40	−34:38:46.0	19.18	1.23	48.3 ± 0.7	0.43 ± 0.03	0.63 ± 0.06	0.995
For-0019	15B	3287.826	02:40:25.26	−34:38:14.1	19.20	1.04	146.1 ± 5.7	⋅⋅⋅	⋅⋅⋅	0.000
For-0020	15B	3287.826	02:40:04.61	−34:40:07.9	19.32	1.36	65.3 ± 0.6	0.50 ± 0.04	0.57 ± 0.08	0.994
For-0021	15B	3287.826	02:39:56.33	−34:39:49.7	19.11	1.30	57.4 ± 0.7	0.50 ± 0.03	0.58 ± 0.07	0.995
For-0022	15B	3287.826	02:40:01.03	−34:39:16.9	18.92	1.13	70.2 ± 0.7	0.44 ± 0.03	0.54 ± 0.06	0.990
For-0023	15B	3287.826	02:40:09.10	−34:39:13.9	19.24	1.16	58.0 ± 0.6	0.50 ± 0.03	0.61 ± 0.07	0.995
For-0024	15B	3287.826	02:40:04.67	−34:38:41.0	19.22	1.15	65.6 ± 0.6	0.54 ± 0.04	0.57 ± 0.08	0.993
For-0025	15B	3287.826	02:40:02.44	−34:38:26.2	19.17	1.18	36.5 ± 0.6	0.51 ± 0.03	0.62 ± 0.07	0.982
...

Only a portion of this table is shown here to demonstrate its form and content. Machine-readable and Virtual Observatory (VO) versions of the full table are available.

Download table as:  Machine-readable (MRT)Virtual Observatory (VOT)Typeset image

Table 4. MMFS Spectroscopic Data–Sculptor

Target	Field	HJD −2.45 × 106	α2000 (hh:mm:ss)	δ2000 (dd:mm:ss)	V (mag)	V-I (mag)	Vhelio (km s−1)	ΣFe (Å)	ΣMg (Å)	$\hat{P}_{M}$	〈V〉helio (km s−1)	〈ΣMg〉 (Å)
Scl-0001	20B	3286.629	01:00:46.17	−33:39:19.6	19.32	1.13	114.1 ± 1.9	0.36 ± 0.02	0.43 ± 0.04	0.998	112.6 ± 1.2	0.56 ± 0.03
	4B	3288.571					111.5 ± 1.6	0.34 ± 0.02	0.69 ± 0.04
Scl-0002	20B	3286.629	01:00:45.44	−33:39:59.9	18.72	1.15	110.1 ± 0.9	0.32 ± 0.01	0.39 ± 0.03	1.000	110.2 ± 0.9	0.40 ± 0.03
	4R	3288.571					111.5 ± 2.6	0.27 ± 0.03	0.42 ± 0.08
Scl-0003	20B	3286.629	01:00:44.46	−33:41:54.1	19.91	1.14	118.4 ± 2.8	0.42 ± 0.04	0.53 ± 0.06	1.000	116.4 ± 1.9	0.50 ± 0.05
	1B	4017.555					114.8 ± 2.5	0.32 ± 0.05	0.39 ± 0.11
Scl-0004	20B	3286.629	01:00:46.96	−33:43:19.9	19.47	1.15	95.6 ± 2.2	0.38 ± 0.02	0.34 ± 0.05	1.000	103.8 ± 0.7	0.34 ± 0.03
	4B	3288.571					104.5 ± 2.0	0.30 ± 0.02	0.32 ± 0.05
	1B	4017.555					104.6 ± 0.8	0.36 ± 0.03	0.36 ± 0.07
Scl-0005	20B	3286.629	01:00:42.44	−33:43:26.0	19.11	1.18	115.5 ± 2.1	0.43 ± 0.02	0.33 ± 0.04	1.000	117.7 ± 0.9	0.40 ± 0.03
	4R	3288.571					115.5 ± 7.9	0.39 ± 0.04	0.52 ± 0.10
	5R	3289.593					114.6 ± 3.3	0.40 ± 0.03	0.53 ± 0.09
	1B	4017.555					118.7 ± 1.1	0.38 ± 0.04	0.50 ± 0.08
Scl-0006	20B	3286.629	01:00:44.88	−33:43:57.7	19.17	1.11	93.0 ± 2.3	0.38 ± 0.02	0.31 ± 0.05	0.999	93.4 ± 2.2	0.32 ± 0.05
	4R	3288.571					100.7 ± 10.1	0.15 ± 0.08	0.37 ± 0.14
Scl-0007	20B	3286.629	01:00:42.07	−33:44:02.7	19.26	1.18	118.6 ± 2.0	0.29 ± 0.04	0.63 ± 0.05	0.991
Scl-0008	20B	3286.629	01:00:51.01	−33:38:20.5	18.97	1.13	113.8 ± 4.9	0.30 ± 0.02	0.63 ± 0.03	0.984	103.6 ± 1.4	0.60 ± 0.03
	1B	4017.555					102.7 ± 1.5	0.33 ± 0.04	0.39 ± 0.08
Scl-0009	20B	3286.629	01:00:50.99	−33:40:02.4	18.50	1.17	95.7 ± 0.8	0.39 ± 0.01	0.42 ± 0.03	0.999	95.7 ± 0.8	0.41 ± 0.03
	4R	3288.571					97.0 ± 4.1	0.33 ± 0.03	0.32 ± 0.09
Scl-0010	20B	3286.629	01:00:49.48	−33:40:37.8	18.98	1.16	101.6 ± 1.6	0.33 ± 0.02	0.39 ± 0.03	1.000	101.6 ± 1.6	0.41 ± 0.03
	4R	3288.571					101.9 ± 5.1	0.40 ± 0.03	0.59 ± 0.09
Scl-0011	20B	3286.629	01:00:51.15	−33:47:17.3	19.39	1.04	112.6 ± 4.8	0.34 ± 0.03	0.51 ± 0.05	1.000	114.9 ± 1.1	0.49 ± 0.04
	1B	4017.555					115.0 ± 1.1	0.37 ± 0.04	0.43 ± 0.09
Scl-0012	20B	3286.629	01:00:51.95	−33:46:24.0	19.12	1.20	111.8 ± 0.8	0.36 ± 0.02	0.46 ± 0.04	1.000	111.3 ± 0.4	0.45 ± 0.03
	5R	3289.593					112.4 ± 11.2	0.30 ± 0.04	0.48 ± 0.10
	1B	4017.555					111.0 ± 0.5	0.39 ± 0.03	0.39 ± 0.06
...

Only a portion of this table is shown here to demonstrate its form and content. Machine-readable and Virtual Observatory (VO) versions of the full table are available.

Download table as:  Machine-readable (MRT)Virtual Observatory (VOT)Typeset image

Table 5. MMFS Spectroscopic Data–Sextans

Target	Field	HJD −2.45 × 106	α2000 (hh:mm:ss)	δ2000 (dd:mm:ss)	V (mag)	V-I (mag)	Vhelio (km s−1)	ΣFe (Å)	ΣMg (Å)	$\hat{P}_{M}$	〈V〉helio (km s−1)	〈ΣMg〉 (Å)
Sex-0001	8B	3086.796	10:14:25.29	−01:42:53.7	19.70	0.62	269.4 ± 2.5	0.35 ± 0.03	0.45 ± 0.07	0.000	270.2 ± 1.7	0.32 ± 0.06
	15B	3417.761					270.9 ± 2.2	0.31 ± 0.03	0.14 ± 0.09
Sex-0002	8B	3086.796	10:14:31.87	−01:39:06.6	⋅⋅⋅	⋅⋅⋅	47.3 ± 2.5	0.36 ± 0.02	0.99 ± 0.03	0.000	47.1 ± 1.4	0.92 ± 0.02
	15B	3417.761					47.1 ± 1.8	0.35 ± 0.02	0.84 ± 0.03
Sex-0003	8B	3086.796	10:14:27.43	−01:37:24.8	⋅⋅⋅	⋅⋅⋅	123.8 ± 3.2	0.28 ± 0.03	1.05 ± 0.04	0.000	123.3 ± 2.0	0.92 ± 0.03
	15B	3417.761					122.9 ± 2.6	0.34 ± 0.02	0.80 ± 0.04
Sex-0004	8B	3086.796	10:14:18.28	−01:44:08.4	18.49	1.33	29.8 ± 2.5	0.30 ± 0.05	1.00 ± 0.06	0.000	30.1 ± 1.8	0.85 ± 0.03
	15B	3417.761					30.4 ± 2.5	0.33 ± 0.02	0.80 ± 0.04
Sex-0005	8B	3086.796	10:14:18.50	−01:39:59.3	18.06	1.25	31.9 ± 1.9	0.38 ± 0.02	0.98 ± 0.02	0.000	32.4 ± 1.3	0.94 ± 0.01
	15B	3417.761					32.8 ± 1.8	0.37 ± 0.01	0.89 ± 0.02
Sex-0006	8B	3086.796	10:14:20.25	−01:39:23.0	19.77	0.73	273.5 ± 4.0	0.28 ± 0.05	0.48 ± 0.12	0.000	271.6 ± 1.8	0.62 ± 0.05
	15B	3417.761					271.1 ± 2.1	0.35 ± 0.03	0.64 ± 0.05
Sex-0007	8B	3086.796	10:14:17.73	−01:38:28.7	19.42	0.58	147.6 ± 3.0	0.24 ± 0.04	0.48 ± 0.08	0.000	145.9 ± 2.0	0.37 ± 0.05
	15B	3417.761					144.5 ± 2.8	0.25 ± 0.03	0.30 ± 0.06
Sex-0008	8B	3086.796	10:14:06.42	−01:34:50.0	17.55	1.26	−9.0 ± 1.7	0.35 ± 0.01	0.87 ± 0.02	0.000
Sex-0009	8B	3086.796	10:14:08.40	−01:35:49.1	19.52	0.90	223.6 ± 4.9	0.31 ± 0.03	0.35 ± 0.07	0.998	226.1 ± 1.7	0.31 ± 0.05
	15B	3417.761					226.4 ± 1.8	0.26 ± 0.03	0.28 ± 0.06
Sex-0010	8B	3086.796	10:14:17.31	−01:35:55.0	⋅⋅⋅	⋅⋅⋅	−39.6 ± 2.6	0.35 ± 0.02	0.77 ± 0.04	0.000	−39.8 ± 1.6	0.81 ± 0.02
	15B	3417.761					−40.0 ± 2.1	0.37 ± 0.02	0.84 ± 0.03
Sex-0011	8B	3086.796	10:14:15.20	−01:36:16.9	19.77	1.02	72.8 ± 3.0	0.31 ± 0.04	0.93 ± 0.06	0.000	64.6 ± 0.8	0.84 ± 0.04
	15B	3417.761					64.0 ± 0.8	0.37 ± 0.04	0.73 ± 0.07
Sex-0012	8B	3086.796	10:14:21.75	−01:31:45.0	18.12	0.82	317.0 ± 1.5	0.33 ± 0.01	0.59 ± 0.02	0.000
Sex-0013	8B	3086.796	10:14:16.38	−01:44:00.5	20.36	1.20	110.9 ± 3.5	0.43 ± 0.06	0.56 ± 0.13	0.000
Sex-0014	8B	3086.796	10:14:01.64	−01:45:32.9	18.77	1.17	16.9 ± 1.8	0.41 ± 0.03	0.81 ± 0.04	0.000	16.8 ± 1.4	0.80 ± 0.03
	15B	3417.761					16.5 ± 2.1	0.36 ± 0.03	0.79 ± 0.04
Sex-0015	8B	3086.796	10:13:55.24	−01:33:20.5	19.80	0.95	246.2 ± 4.4	0.34 ± 0.04	0.38 ± 0.09	0.997	223.9 ± 1.4	0.40 ± 0.06
	7B	3090.704					233.6 ± 5.1	0.30 ± 0.10	0.07 ± 0.25
	15B	3417.761					220.2 ± 1.6	0.34 ± 0.04	0.44 ± 0.07
...

Only a portion of this table is shown here to demonstrate its form and content. Machine-readable and Virtual Observatory (VO) versions of the full table are available.

Download table as:  Machine-readable (MRT)Virtual Observatory (VOT)Typeset image

We use repeat measurements of stars measured on both the blue and the red channels of the MIKE spectrograph to quantify any systematic difference that may arise due to channel-dependent dispersion characteristics (Bernstein et al. 2003; Paper I). For 647 stars, we obtained at least one acceptable velocity measurement on both channels. For these stars, the left-hand panel of Figure 2 plots the velocity measured on the blue channel against that measured on the red channel. The best-fitting line through these data points has slope 0.998 ± 0.002 and intercept −0.33 ± 0.52 km s⁻¹, indicating there is no systematic difference between velocities measured with the blue and the red channels.

Zoom In Zoom Out Reset image size

We find evidence for a slight channel dependence in measuring magnesium strength. For 476 stars, we obtained at least one acceptable measurement of ΣMg on both channels. For these stars, the right-hand panel in Figure 2 plots ΣMg_blue against ΣMg_red. The best-fitting straight line is given by

$$\begin{equation} \Sigma \mathrm{Mg}_{\rm blue}=(0.936 \pm 0.048)\Sigma \mathrm{Mg}_{\rm red} - (0.007 \pm 0.048)\ \mathrm{\AA}. \end{equation} \tag{ 1 }$$

For subsequent analysis, we apply Equation (1) to place values of ΣMg measured with the red channel on the blue-channel scale. The values of ΣMg listed in Tables 2–5 for red-channel spectra are those obtained after applying Equation (1). We do not apply a similar correction, however, for the values of ΣFe measured with the red channel, because the red-channel's values of ΣFe are calculated using fewer absorption lines than the blue channel's values (the red channel has low throughput at the blue end of our spectral range; see Paper I). Blue- and red-channel measurements of ΣFe must, therefore, be considered separately.

2.1. Metallicity Calibration

In principle the ΣFe and ΣMg indices contain information about metal abundances. In order to calibrate these values onto a metallicity scale, we obtained MMFS spectra of individual stars in globular clusters spanning the metallicity range −2 ⩽ [Fe/H] ⩽ −0.5. Left-hand panels in Figure 3 plot ΣFe (top) and ΣMg (bottom) against V − V_HB, where V_HB is the apparent magnitude of the cluster's horizontal branch. The empirical relationship is approximately linear for each cluster, as is the case for more conventional indices derived from the calcium triplet ("CaT," e.g., Armandroff & Da Costa 1991). The slope reflects the dependence of opacity on surface gravity and temperature, both of which correlate with V − V_HB as an evolved star ascends the red giant branch. Following the procedure of Rutledge et al. (1997; see also Koch et al. 2006), we assume a common slope and fit straight lines to the data from each cluster. The intercepts, known as "reduced" indices, are given by

$$\begin{equation} \Sigma \mathrm{Fe_{\rm blue}}^{\prime }=\Sigma \mathrm{Fe_{\rm blue}}+(0.040\pm 0.001)(V-V_{\rm HB}); \end{equation} \tag{ 2 }$$

$$\begin{equation} \Sigma \mathrm{Mg_{\rm blue}}^{\prime }=\Sigma \mathrm{Mg_{\rm blue}}+(0.079\pm 0.002)(V-V_{\rm HB}) \end{equation} \tag{ 3 }$$

and provide a measure of metal abundance that, insofar as the linear model is valid, is independent of surface gravity and temperature.

Zoom In Zoom Out Reset image size

Right-hand panels in Figure 3 plot cluster metallicity, on the standard scales of Zinn & West (1984, "ZW84") and Carretta & Gratton (1997, "CG97"), against the mean-reduced indices for each globular cluster. The best-fitting straight lines are given by

$$\begin{equation} \mathrm{[Fe/H]_{ZW84}}=(7.05\pm 2.10)\Sigma \mathrm{Fe} -3.97\pm 2.03 [2.19]; \end{equation} \tag{ 4 }$$

$$\begin{equation} \mathrm{[Fe/H]_{ZW84}}=(1.93\pm 0.16)\Sigma \mathrm{Mg}^{\prime } -2.34\pm 0.15 [0.24]; \end{equation} \tag{ 5 }$$

$$\begin{equation} \mathrm{[Fe/H]_{CG97}}=(6.81\pm 1.87)\Sigma \mathrm{Fe}^{\prime } -3.73\pm 0.48 [0.89]; \end{equation} \tag{ 6 }$$

$$\begin{equation} \mathrm{[Fe/H]_{CG97}}=(1.76\pm 0.16)\Sigma \mathrm{Mg}^{\prime } -2.11\pm 0.10 [0.20], \end{equation} \tag{ 7 }$$

where the errors include measurement errors as well as residuals from the fit, and the value in the square brackets is the median metallicity error from dSph stars in the MMFS sample. The metallicities obtained from the iron indices carry large errors due to the relative weakness of the iron features in the MMFS spectra. The magnesium index thus provides our best measure of metallicity, yielding median errors of 0.20 dex on the scale of CG97 (0.24 dex on the scale of ZW84). In subsequent discussion, we consider only the [Fe/H] values on the CG97 scale, obtained using Equation (7) (results are qualitatively unchanged if we use instead the values on the ZW84 scale, obtained from Equation (5)).

We apply Equations (2)–(7) to obtain [Fe/H] for each dSph star in the MMFS sample. Figure 4 displays scatter plots of [Fe/H]_CG97 against distance from the dSph center, as well as histograms of the global [Fe/H] distributions (results are qualitatively the same if we use the metallicity scale of ZW84). We find that the resulting metallicity distributions are significantly narrower and biased toward high metallicity with respect to distributions previously derived from CaT spectroscopy. For example, the Carina sample of Koch et al. (2006) and the Fornax sample of Battaglia et al. (2006) both contain significant numbers of stars with metallicities as low as [Fe/H] ∼−2.5 dex, and as high as ∼−0.5 dex; for both galaxies the MMFS distributions fall to zero short of these extremes. While the MMFS sample ranks dSphs by mean [Fe/H] in the same sequence as previously published studies, it displays a narrower spread. These comparisons lead us to conclude that either the slope we have determined in our metallicity calibration (Equations (4)–(7)) is simply too shallow, which we may remedy with observation of more calibrating clusters, or there are problems with one or both of the methods by which which Ca and Mg is used to measure iron abundance. For example, Kirby et al. (2008) have recently shown that a spectral synthesis method that fits model spectra directly to iron lines implies wider metallicity distributions than are indicated by the standard CaT method.

Zoom In Zoom Out Reset image size

For the analysis of the MMFS metallicity data we, therefore, recommend against using the absolute values of [Fe/H] that result from calibration using Equations (4)–(7). Instead, we recommend using the reduced indices ΣFe and ΣMg, which are obtained by applying Equations (2) and (3) to the raw indices listed in Tables 2–5, as measures of relative metallicity. In this way the MMFS sample can be divided into metal-rich and metal-poor subcomponents, as has been done in previous analyses of these and other dSphs (e.g., Tolstoy et al. 2004; Battaglia et al. 2006), and can be used to search for evidence of metallicity gradients (M. G. Walker et al. 2009, in preparation).

2.2. Repeat Measurements

The MMFS data set contains repeat velocity measurements for 1363 dSph target stars. There are 3115 independent measurements of these stars, including up to five measurements for some stars. For subsequent analyses, we replace the n_j measurements of the jth star with the weighted mean

$$\begin{equation} \langle V\rangle _{j}=\frac{\sum _{i=1}^{n_j}(w_{ij}V_{ij})}{\sum _{i=1}^{n_j}w_{ij}}, \end{equation} \tag{ 8 }$$

where the weights, w_ij = σ⁻²_V,ij, are defined by the velocity errors $\sigma _{V_{ij}}$. The weighted mean provides an unbiased estimate of the true velocity, and has variance $\sigma _{\bar{V},j}^2=\big(\!\sum _{i=1}^{n_j}w_{ij}\big)^{-1}$. We again use the weighted mean to combine repeat measurements (of which there are 2408 for 1089 stars) of the composite index ΣMg. For all stars with n_j > 1 independent velocity measurements, Column (12) of Tables 2–5 lists the weighted mean velocity. Column (13) gives the weighted mean value of ΣMg for stars with multiple measurements.

2.3. dSph Membership

We have presented all spectroscopic observations of dSph stars taken with MMFS as of 2008 August. The MMFS sample more than doubles the number of velocity measurements for dSph stars, dating from Aaronson's (1983) initial kinematic study to the most recent published data from surveys using Keck and the VLT (Muñoz et al. 2006; Koch et al. 2006, 2007b, 2007a; Battaglia et al. 2006; Battaglia et al. 2008). In the forthcoming work, we use the data presented here, as well as the membership probabilities, to analyze the kinematics, chemodynamics, and potential substructure in these dSphs.

Data tables that present the complete MMFS data sets are provided in the electronic version of this article, and we welcome their use. We thank the staff at Las Campanas Observatory for generous and expert support. M.G.W. and M.M. thank the Horace H. Rackham Graduate School at the University of Michigan for generous support, including funds for travel to Magellan. M.M. acknowledges support from NSF grants AST-0206081, 0507453, and 0808043. E.W.O. acknowledges support from NSF Grants AST-0205790, 0505711, and 0807498. M.G.W. acknowledges support from the STFC-funded Galaxy Formation and Evolution programme at the Institute of Astronomy, University of Cambridge.