Stellar surface structures as a practical limitation to ultra-high-precision optical astrometry (Eriksson, Lindegren):

In the near future, high-precision astrometric observations may become important for discovering exoplanets around nearby stars, complementing other techniques such as Doppler and transit observations. Brightness irregularities on the surfaces of these stars (spots, plages, granulation, etc) will however introduce noise in the observed positions, possibly hiding the astrometric signatures of small exoplanets. We use observed fluctuations of the magnitudes and radial velocities for stars of different types to estimate the likely size of their "astrometric jitter". It is concluded that small (Earth-like) planets may indeed be detected by the astrometric method, but only around stars that are unusually stable, similar to our Sun.
Main reference: Eriksson & Lindegren (A&A 476, 1389, 2007).

Bayesian estimation of stellar ages and star formation histories (Jørgensen, Lindegren):

The ages of stars can sometimes be determined from their positions in the observational HR diagram, using theoretical isochrones. We have developed a technique to derive probability density functions for the ages of stars, based on their observed characteristics. Critically, the method depends on the absolute magnitudes of the stars, which are obtained by means of their trigonometric parallaxes. The method is essentially an application of Bayes' theorem to the classical isochrone fitting problem. In an extension of the method, the age distribution of a sample of stars can be derived, which in turn depends on the rate of star formation as function of time (the star formation history).
Main references: Jørgensen & Lindegren (A&A 436, 127, 2005), Jørgensen & Lindegren (ESA SP-576, p.171, 2005), Nordströt al. (A&A 418, 989, 2004).

The local mass density (Holmberg, Flynn):

The Hipparcos data permit a detailed mapping of the spatial distribution and velocities for well-defined samples of stars in the solar neighbourhood, out to distances of a few hundred parsec for the more luminous stars.  These data can be used to trace the gravitational field, and hence the dynamical mass density in our vicinity, and to identify and study kinematic groups of stars with a common origin.  Using a complete volume-limited sample of A and F stars, an estimate of the local dynamical mass density of 0.102 ± 0.010 M_Sun pc^-3 was derived, to be compared with 0.095 M_Sun pc^-3 of visible disk matter.  Thus we find no significant amount of dark matter in the galactic disk.
Main reference: Holmberg & Flynn (MNRAS 313, 209, 2000).

Old moving groups (Feltzing, Holmberg):

Stars scattered over the whole sky, but identified as a group by their common space motion, provide an important evolutionary link between clusters and field stars.  The large number of accurate parallaxes provided by Hipparcos has made it possible to reliably identify moving groups and to study their ages, metallicities, and dynamical evolution.  The reality of the old (2 Gyr) metal rich ([Fe/H] = 0.2) moving group HR1614 was proved by combining Hipparcos data with metallicities derived from Strömgren photometry.  Supported by dynamical simulations, an extended sample of probable member stars was derived.
Main reference: Feltzing & Holmberg (A&A, 357, 153, 2000).

Cluster kinematics (Madsen):

The kinematics of classical moving clusters and associations were also studied as part of the programme to determine astrometric radial velocities. In particular, the internal structure and velocity dispersion of the Hyades cluster has been studied in detail, including fine structure of the HR diagram based on kinematically improved stellar distances.
Main references: Madsen (A&A 401, 565, 2003), Madsen, Dravins & Lindegren (A&A 381, 446, 2002), Lindegren, Madsen & Dravins (A&A 356, 1119, 2000).