Due to angular resolution and sensitivity constraints the most suitable target stars around which TPF-I can search for exoplanets are the stars nearest to us. However, the nearest stars encompass a large variety of stellar types, classes, ages, and multiplicities, and so the set of all nearby stars needs to be trimmed or culled, for both scientific reasons and engineering/observational constraints. The current scientific source selection criteria, defined at a workshop held at the Naval Research Observatory in Nov. 2004, are discussed in the following paragraphs.

Science Criteria

The identification of candidate scientific targets for TPF should begin with a complete list of nearby stars. The master list selected for this purpose is the Hipparcos list of nearby stars, out to a distance cutoff of 30 pc. This list does not include every star within 30 pc, because dim stars at the larger distances do not fall within the Hipparcos sensitivity limit, implying that the list grows less and less complete as stellar temperatures decrease. However, for the primary targets of interest, "solar-like" F, G, and K stars, the list should be fairly complete.

Figure 1. Candidate targets for TPF-I.

It is desirable to keep the number of culling criteria as small as possible. The criterion of interest can be summarized simply: bright, nearby, solar-like, main-sequence stars for which a binary companion is not too close. This statement translates in a straightforward fashion to the detailed technical requirements listed in Table 1.

The final column in this table gives the number of potential target stars remaining after the set of culls down to a given line has been applied. Even after all applicable science culls have been applied, over 1000 suitable targets remain. These 1014 stars are plotted in Fig. 1 on a space of stellar distance vs. projected inner habitable zones size, from which it can be seen, e.g., that for a sizable candidate population of stars to be observable, an inner working angle of 50 mas or smaller is needed. The final candidate list of 1014 stars was used to predict the performance of different TPF-I architectures.

Engineering Criteria

Next one also needs to include instrumental or engineering constraints which can eliminate certain classes of stars. With the current baseline design of TPF-I, these fall into two categories: first, sunshade constraints limit observations to stars within 45 of the ecliptic, and second, control of stray light implies that any bright companion stars must be more than 10 arc sec away from the target star of interest. As Table 1 shows, this eliminates about 40% of the scientifically interesting stars.