The goal of correcting the Earth's atmospheric effects on astronomical images has seen considerable success through the use of laser guide star adaptive optics (AO) systems that provide a quantitative measure of the turbulence distributed over the aperture of a large telescope. However, laser systems by themselves cannot detect the ensemble shift in an object's image that is induced by the atmosphere, and the encircled energy in a diffraction-limited spot could be increased by more than 40% if this correction were made. Such a "tip/tilt" correction is only available through the use of a natural guide star - a rather bright natural star included in the telescope field of view - in tandem with the laser projection system. The near-IR tip/tilt sensor being constructed by Dr. Peter Wizinowich and collaborators of the California Association for Research in Astronomy for the Keck 10-m telescope on Mauna Kea, Hawaii will accomplish this goal. By working in the K (2.2 micron wavelength) atmospheric transmission band, the planned system will provide high bandwidth (up to 1 kHz) tip/tilt information using one to three stars in a 2 arcminute diameter area, and enable corrections over that entire field of view. Additionally, focus corrections are available at a much higher time resolution than for standard schemes.
The Keck telescopes are arguably the most advanced, reliable, and scientifically productive large-telescope AO systems available anywhere in the world. Scientists and technical staff at the managing institutions are among the most capable in the world, and the mountaintop is an outstanding observing site. Up to 30% of observing time is available to the general community through various means, so improvements in the imaging capabilities of the telescopes through mechanisms like near-IR tip/tilt correction systems benefit all of U.S. astronomy. Funding for this work is being provided by NSF's Division of Astronomical Sciences through its Advanced Technologies and Instrumentation program.
Turbulence in the earth's atmosphere blurs the images obtained by ground-based telescopes. Adaptive optics (AO) corrects for this blurring by sensing the wavefront distortions introduced by the atmosphere and correcting for them in real-time (~ 1000 times per second). Natural guide star (NGS) AO uses a star outside the earth's atmosphere to sense the atmospheric distortions; unfortunately bright enough stars are only available over ~ 1% of the sky. Laser guide star (LGS) AO uses laser illumination projected high up in the atmosphere to create an image for sensing the atmospheric distortions. A natural star is still needed to measure image motion (tip-tilt) and low order aberrations however it can be much fainter for this purpose opening up much more of the sky to AO correction. However, even with a LGS the sky coverage and performance of AO systems is ultimately limited by the NGS used for low order correction. This limitation can be dramatically reduced by measuring the tip and tilt of the NGS in the near-infrared (NIR) where the NGS is partially corrected by the LGS AO system and where stars are generally several magnitudes brighter than at visible wavelengths. The purpose of this NSF proposal was to demonstrate the scientific utility of NIR tip-tilt sensing for science observations. Our project has successfully implemented a NIR tip-tilt sensor for the Keck I telescope's LGS AO system using a 2048x2048 pixel Teledyne Hawaii-2RG detector within a cryogenic camera dewar. In order to achieve fast readout rates with low read noise a single small region (typically a 4x4 pixel region anywhere over the detector) is read out multiple times for each tip-tilt measurement. The multiple reads are inputs to our modified real-time control processor which coadds the readouts and then, since the camera is read out non-destructively, subtracts the previous coadd. The resultant image is used to determine the tip and tilt which is applied to a fast tip-tilt mirror. The NIR tip-tilt sensor camera is installed on the Keck I AO system optical bench just before the science instrument (OSIRIS) which is an integral field spectrograph and imager as shown in Figure 1. A portion of the light to OSIRIS is picked off with a choice of two dichroic beamsplitters. The performance of the NIR tip-tilt sensor and the existing visible tip-tilt sensor are quite comparable on bright NGS. The NIR tip-tilt sensor's performance gains have been demonstrated to become more and more obvious as fainter NGS are used. The NIR tip-tilt sensor is undergoing final on-sky commissioning and performance characterization with science observations scheduled to begin in early 2015. A gravitational lens has been observed as a science verification target and the results have been sufficiently impressive that they are being prepared for publication as an Astrophysical Journal Letter.
