2000/11/04 by Karl Glazebrook, Joss Bland-Hawthorn, Joss Bland‐Hawthorn · 4 citations
Chemistry · Physics and Astronomy · #Adaptive optics and wavefront sensing #Artificial intelligence #Astronomy #Astrophysics #Computer science #Noise (video) #Optics #Physics #Sky #Spectral line #Spectrograph #Spectroscopy #Spectroscopy and Laser Applications #Stellar, planetary, and galactic studies #Telescope #William Herschel Telescope #astro-ph
paper · pdf · doi:10.1086/318625
published as Publ.Astron.Soc.Pac.113:197,2001 · Accepted for publication in PASP; 25 pages, 12 figures. A higher-quality compressed Postscript file (2.2Mb) is available from http://www.pha.jhu.edu/~kgb/papers/nodshuffle2000hq.ps.gz
arxiv created 2000/11/04 · openalex publication_date 2001/02/01 · arxiv updated 2011/03/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
We describe a new approach to obtaining very high surface densities of optical spectra in astronomical observations with extremely accurate subtraction of night sky emission. The observing technique requires that the telescope is nodded rapidly between targets and adjacent sky positions; object and sky spectra are recorded on adjacent regions of a low‐noise CCD through charge shuffling. This permits the use of extremely high densities of small slit apertures ("microslits") since an extended slit is not required for sky interpolation. The overall multiobject advantage of this technique is as large as 2.9 times that of conventional multislit observing for an instrument configuration which has an underfilled CCD detector and is always greater than 1.5 for high target densities. The "nod‐shuffle" technique has been practically implemented at the Anglo‐Australian Telescope as the "LDSS++ project" and achieves sky subtraction accuracies as good as 0.04%, with even better performance possible. This is a factor of 10 better than is routinely achieved with long slits. LDSS++ has been used in various observational modes, which we describe, and for a wide variety of astronomical projects. The nod‐shuffle approach should be of great benefit to most spectroscopic (e.g., long slit, fiber, integral field) methods and would allow much deeper spectroscopy on very large telescopes (10 m or greater) than is currently possible. Finally, we discuss the prospects of using nod‐shuffle to pursue extremely long spectroscopic exposures (many days) and of mimicking nod‐shuffle observations with infrared arrays.