SIM EPICS/Marcy Tutorial

Background

Two Teams were selected by NASA to search for exoplanets using SIM. The EPIcS team is lead by PI Shao & is called Team A. The other team is lead by PI Marcy & is called Team B. While nominally competitive, the teams work closely together. Competition consists essentially of the selection of stars by the respective teams. This was completed in the SIM draft during which teams A & B selected stars (using different criteria).

The EPIcS team had a larger telescope time allocation. EPICS chose a 2 Tiered observing strategy. Tier 1 stars would be observed at high precision, Tier 2 stars at low precision. Since EPICS' Tier 1 stars will be observed the same way as all of Team B's ( Marcy's) stars, these two lists of stars are treated together and all called "Tier 1". (Marcy's team has no "Tier 2")

Reference Stars

Every Tier 1 star will have its position measured astrometrically by SIM with respect to ~4-5 Reference Stars which must themselves be astrometrically stable. These observations are referred to as "Narrow Angle", since the reference stars are close to the targets on the sky (~1 degree). (This is in contrast to observations which reference the SIM grid which are referred to as "Wide Angle".)

Selection

K Giants at a distance of ~200-2000 pc were identified as the ideal reference stars. A preliminary survey by Sabine Reffert (then known as Sabine Frink) confirmed this strategy. Reference stars for the Tier 1 Stars were selected using criteria similar to those developed by Dave Ciardi for the SIM Grid. Ciardi has included several files in the SIM library on his method.

Summary of Ciardi Method

Vetting

To insure stability, every reference star assosicated with every Tier 1 target star will be "Doppler Vetted". That is, the Radial Velocity (RV) of each REF star will be measured over a 5 year timescale. The longer this time baseline the better we can vet out long period binary stars. (However note that binaries with periods much longer than SIM's mission lifetime might not be a problem) Stars whose RV is found to vary are likely to have an unseen stellar companion, and will be dropped and replaced with a different star. Vetting was done mainly at Keck/HIRES, with some data taken at Lick/Hamilton. The RV vetting regimen goes something like this:

For each of ~170 Tier 1 Target stars (Teams A+B), roughly 10 candidate reference stars are observed at Epoch 0. (~1700 RV observations)
A few such stars can be rejected immediately, as binaries, etc.
Remaining are then observed twice more (Epochs 1,2) in RV, spanning a ~5 year time baseline.
Stars which show RV variations greater than the expected observational errors of ~10-20 m/s (or perhaps less) would be discarded and replaced with other candidate reference stars from a pre-chosen list.
The exact procedures for replacing ref stars were discussed, including: observing more REF stars than needed at epoch 0 so as to ensure a good time baseline, selecting replacement REF stars so as to ensure good geometric coverage around the target star, etc. In practice, such concerns often took a back seat to the demands of observing time. (Weather, season etc.)
After ~5+ years ideally each target star would have ~4-5 RV quiet REF stars. One of these might, on average be have astrometric noise undetected in RV, but that is acceptable.

Observations of REF stars began June 2004 (run rk86) and were suspended after Sep. 2006 (rj23), as SIM incurred launch delays. Roughly 400 observations were made by this point. A few REF stars were ruled out, but none had yet passed the vetting process because final third epoch data had not yet been obtained. See Ref star status report (LINK) section for current status of these observations. Roughly 70 REF stars observations were made at Lick.

Internal Page