[Previous] | [Session 46] | [Next]
R.A. Wade (Penn State U.)
NASA's Space Interferometry Mission (SIM) promises breath-taking results in a wide variety of astronomical sub-disciplines. Establishing a suitable grid of ``effectively single'' astrometric reference stars for SIM, whose photocenters are stable to 4 micro-arcseconds is a challenging task in its own right, and one that is very demanding of telescope resources.
To reject stars which have unacceptable astrometric wobbles (owing to unresolved companions) from a list of candidate stars, it will be necessary to measure the radial velocity of each star repeatedly, to high precision. The requirement to visit large numbers of targets (N\approx 4000) many times each (M\approx 8) demands that the duration of each visit (t) be brief. Only if t \approx 5 minutes can the product N\times M\times t be kept below 3000 hours, or roughly 300 (clear!)\/ nights, spread over several years.
The requirement that the measurements also be of high precision (a combination of high signal-to-noise ratio, high spectral resolution, and high instrumental stability) indicates the use of telescopes with large light-gathering power, even though the grid stars must be relatively bright (V < 12). Since the grid must be all-sky, observations from more than one site will be necessary.
In this poster, we examine the parts of (m2, a) parameter space that can be explored by synoptic radial velocity studies, where m2 is the mass of the perturbing companion object and a is the orbital separation. We compare the accessible (m2, a) region with the region that is of concern to SIM. Within this framework, we examine whether G dwarfs or K giants would make better grid stars. We take into account the precision realistically achievable for wholesale measurement of radial velocities, the typical distance of potential grid stars of each type, and the statistics of binary incidence for stellar and planetary/brown dwarf companions. Finally, we ask whether other types of stars might have characteristics that make them suitable to contribute (at least in part) to solving the SIM grid problem.