Satellites as probes of dark matter and gravitational theories

The Milky Way hosts about 150 globular clusters, and at least 17 dwarf spheroidal galaxies. These satellites experience a constantly changing gravitational field on their orbits. Close encounters with the Galactic bulge and passages through the Galactic disk enhance the effect of the constantly changing tidal field. As a consequence satellite member stars can leave their host's gravitational potential. For globular clusters, internal mechanisms, such as 2-body relaxation are also resulting in a loss of stars. Hence, the globular clusters are constantly losing stars and are being dissolved. In this thesis I investigate 17 globular cluster for signs of dissolution. I.e., we are studying the two-dimensional distribution of (potential) cluster member stars on the sky using photometric data from the Sloan Digital Sky Survey. We use a color-magnitude weighted counting algorithm to count the stars around the globular clusters. We detect the known tidal tails of Pal 5 and NGC 5466. Further, we also confirm some previous finding of possible tidal features for NGC 5053 and NGC 6341. For NGC 4147, we observe for the first time complex two-dimensional features, resembling a multiple-arm morphology. For almost all clusters in our sample we observe a halo of extra tidal stars. We observe no new large scale tidal features for our sample of clusters containing stars brighter than ~22.5 mag. The lack of large scale tidal tails is compatible with theoretical predictions of the destruction timescales for the clusters in our sample.
We also observe the two-dimensional distribution of stars around three dwarf spheroidal galaxies: Sextans, Leo II, and Ursa Minor. Each galaxy reveals a unique structure. The main, luminous body of Sextans is not filling the tidal radius. We observe an off-center peak of highest stellar density. For Leo II, we observe an almost symmetric structure, compatible with the theory that Leo II has never come close to the Milky Way. We detect the complex structure of Ursa Minor, with two off-center peaks. We observe no large scale structure emanating from this dwarf galaxy. We further investigate the possibility of a line-of-sight depth of Sextans and Ursa Minor. We study the thickness of the blue horizontal branch. For Sextans, we observe an increasing thickness with increasing radius, comparable with the photometric error. Only detailed modeling will be able to show the significance of this varying thickness. For Ursa Minor, the increase in horizontal branch thickness is negligible, compared to the photometric error. Hence, Ursa Minor shows no sign of a significant line-of-sight depth. The distribution of red and blue horizontal stars was investigated for Sextans. The ''red'' population is much more concentrated. The peak of the density of the two populations does not coincide.
Further, we investigated one globular cluster in particular, Pal 14. This cluster is sparse and at a remote location in the Galaxy. We aim to answer the question whether Pal 14 is governed by classical or modified Newtonian dynamics. We measured the radial velocity of 17 red giant branch stars and (probable) AGB stars with UVES@VLT and the Keck I telescope. The resulting line-of-sight velocity dispersion is comparable to the theoretical predictions for the case of classical dynamics. The predicted value for modified dynamics is about twice as large as the observed value. With HST images we derived the cluster's mass function and computed its total mass. The main sequence mass function slope is flatter than the canonical value, the cluster seems to be depleted in lower mass stars. N-body simulations predict for a given mass of the cluster its line-of-sight velocity dispersion in modified dynamics. The measured mass for Pal 14 is requiring a much larger velocity dispersion in modified Newtonian dynamics than we have measured. This leads to the conclusion that if Pal 14 is on a circular orbit, modified dynamics cannot explain the low velocity dispersion and the measured mass simultaneously.