Abstract:
One of the most ambitious goals in early part of the 21 century is the detection of an Earth-twin planet in the habitable zone of a nearby star. Many concepts have been proposed, most of them costing many billions of dollars. One of these SIM (Space Interferometry Mission) was recommended by two US astronomy decadal reports, but SIM-Lite at a cost of $1.2~1.6 B was not recommended in the 2010 decadal report. The metrology technology developed for SIM however gave rise to the NEAT concept that can detect an Earth twin around ~100 of the nearest stars for potentially 1/4 of the cost of SIM-Lite.
NEAT is a long focus 1m telescope that does relative astrometry over ~0.6 deg field of view with an accuracy of ~0.8 uas (microarcsec) in 1 hr and a mission accuracy of 0.04 uas. This level of performance is needed to search and detect an Earth twin around the nearest ~100 stars. This talk describes the concept and the techniques used to advance the state of the art in a few areas by 2-3orders of magnitude. Conventional star trackers can centroid a star's image to 1/100 of a pixel. Astrometry with the Hubble telescope CCD cameras has been demonstrated to 1/300 pixel. NEAT will centroid a star's position to 1e-5 pixels. To do this we need to calibrate the detector and the optical PSF to much higher accuracy than previously done. Also 1 uas over a ~2000 arcsec field means the geometry of the focal plane needs to be accurate/stable to 5e-10. The geometry of the focal plane (the location of all the pixels) will be measured to ~0.1nm, with a SIM derived metrology system.