IMPROVING GALACTIC CENTER ASTROMETRY BY REDUCING THE EFFECTS OF GEOMETRIC DISTORTION

We present significantly improved proper motion measurements of the Milky Way's central stellar cluster. These improvements are made possible by refining our astrometric reference frame with a new geometric optical distortion model for the W. M. Keck II 10 m telescope's adaptive optics camera (NIRC2) in its narrow field mode. For the first time, this distortion model is constructed from on-sky measurements and is made available to the public in the form of FITS files. When applied to widely dithered images, it produces residuals in the separations of stars that are a factor of similar to 3 smaller compared with the outcome using previous models. By applying this new model, along with corrections for differential atmospheric refraction, to widely dithered images of SiO masers at the Galactic center (GC), we improve our ability to tie into the precisely measured radio Sgr A*-rest frame. The resulting infrared reference frame is similar to 2-3 times more accurate and stable than earlier published efforts. In this reference frame, Sgr A* is localized to within a position of 0.6 mas and a velocity of 0.09 mas yr(-1), or similar to 3.4 km s(-1) at 8 kpc (1 sigma). Also, proper motions for members of the central stellar cluster are more accurate, although less precise, due to the limited number of these wide field measurements. These proper motion measurements show that, with respect to Sgr A*, the central stellar cluster has no rotation in the plane of the sky to within 0.3 mas yr(-1) arcsec(-1), has no net translational motion with respect to Sgr A* to within 0.1 mas yr(-1), and has net rotation perpendicular to the plane of the sky along the Galactic plane, as has previously been observed. While earlier proper motion studies defined a reference frame by assuming no net motion of the stellar cluster, this approach is fundamentally limited by the cluster's intrinsic dispersion and therefore will not improve with time. We define a reference frame with SiO masers and this reference frame's stability should improve steadily with future measurements of the SiO masers in this region (alpha t(-3/2)). This is essential for achieving the necessary reference frame stability required to detect the effects of general relativity and extended mass on short-period stars at the GC.