4.1 Recent Observational Techniques and Results
The early velocity dispersion results were based on only a handful of stars and on observations with appreciable measurement errors. More recent programs have concentrated on obtaining velocities for larger numbers of stars in each galaxy and in reducing the errors in the velocity measurements. In fact, two distinct strategies have been adopted. The first is to obtain velocities for a large number of stars using spectrographic observations, typically of the Ca II triplet. Velocity dispersions are obtained to an accuracy of around 5 km s-1, which is comparable to the expected velocity dispersions of these systems. However, the use of fibers allows not only a large number of stars to be obtained with a single exposure, but also makes repeat observations a practical possibility. This is important to reject high apparent velocity dispersions caused by binary stars and pulsating variables. So far, this technique has been applied to Sculptor and Sextans, but all the local dwarf spheroidals could be studied in this way.
The second strategy is to obtain much higher resolution velocities. Echelle observations have achieved velocities accurate to 1-2 km s-1 in Carina, Ursa Minor, Fornax and Draco. These observations can go somewhat fainter than the fiber technique, so that potentially more stars are available for observation.
The results of such studies suggest large amounts of DM in most of the nearby
dwarf spheroidals, as well as extremely high central DM densities.
Mateo et al. (1991)
used Echelle spectroscopy to obtain velocities for 44 stars and 4
globular clusters around Fornax. Importantly,
Mateo et al. (1991)
studied two different fields, one near the center of Fornax and another displaced
appreciably along the major axis. In both fields, large velocity dispersions
were obtained, suggesting the presence of DM with a more extended spatial
distribution than the stellar component of the galaxy. Moreover, the central
DM density is high with a value of 0.07 ± 0.03
M
pc-3. This is
an order of magnitude higher than typically found in bright spirals (see
Section 6 below).
The repeat observations that
Mateo et al. (1991)
have
for some stars indicate that there are few binaries or pulsating variables
contained in the sample.
Recent results have been obtained for Carina
(Mateo et al. 1992a)
and Sextans
(Suntzeff et
al. 1992).
Large central mass-to-light ratios
are obtained in both cases, implying dark-to-luminous mass ratios around 10.
Central DM densities are also high, with values around 0.1
M pc-3.
Since few repeat measurements are available for these galaxies, it is
conceivable that binary stars or pulsating variables
may be artificially inflating the velocity dispersions.
However, if Fornax is typical, this possibility seems unlikely.
Da Costa and Armandroff (1992) have made two or more observations of some of their sample of stars in Sculptor. They have observed stars both inside and beyond the core radius and find high velocity dispersions in both fields, indicative of an extended dark halo. The central mass-to-light ratio is not extreme, although the central DM density is once again high.
The two faintest local dwarf spheroidals, Ursa Minor and Draco, appear to have
the highest DM fractions. Aaranson and Olszewski
(1987;
1988) and
Olszewski and
Aaronson (1992)
report a series of observations
of stars in both galaxies, in which a number of repeat measurements have been
made. This study has revealed
extremely high central M / LV with values approaching
100 and central densities around 1
M
pc-3 (see however Section 4.2 below).
If the dark halos in Draco and Ursa Minor are more extended than the light,
as is typically the case in other galaxies, the total
dark-to-luminous mass ratio
in these objects is at least 100. Similar observations of Carina
and Sextans give central M / LV values
around 50
(Olszewski and
Aaronson 1992).
There is some indication from the results summarized above that the dwarf spheroidals closer to the Galactic center have the highest mass-to-light ratio and lowest luminosities. The observations are also consistent with the view that the masses of all the dwarf spheroidals are similar, despite an appreciable range in luminosity. However, observational and modelling uncertainties make these conclusions tentative at this stage.
Further details of the observational techniques and strategies in this field are given in the excellent review by Pryor (1992).