The Explosion in Wide-Field Imaging: The last 5 years have witnessed a dramatic increase in wide-field imaging capabilities on moderate and large aperture ground-based telescopes (Table 2). The KPNO and CTIO 4-m telescopes, the CFHT, Subaru and the upgraded MMT all boast optical imaging cameras with fields of view ~30', and have prompted a host of wide-field survey projects that drive an immediate need for deep wide-field spectroscopy of extremely large samples. The wide-field surveys ongoing or completed at the time of writing are summarized in Table 3.
| Telescope | Aperture | Camera | FOV | Years of | # Sq. Deg |
| (m) | (') | Operation | by 20101 | ||
| Mayall | 3.8 | MOSAIC | 36'x36' | 1998-2010 | 1,240 |
| KPNO 0.9m | 0.9 | MOSAIC | 59'x59' | 1998-2010 | 220 |
| Blanco | 3.9 | BTC | 4x15'x15' | 1997-1999 | 250 |
| Blanco | 3.9 | MOSAIC | 35'x35' | 2000-2010 | 1,050 |
| CFHT | 3.6 | UH8K | 29'x29' | 1998-2000 | 200 |
| CFHT | 3.6 | CFHT12K | 42'x28' | 2001-2005 | 480 |
| CFHT | 3.6 | Megacam | 1 sq. deg. | 2001-2010 | 2,700 |
| Hale | 5.0 | LFC | 23'x23' | 2001-2010 | 720 |
| Subaru | 8.3 | SUPRIME-CAM | 34'x27' | 2000-2010 | 3,960 |
| MPG/ESO | 2.2 | WFI | 34'x33' | 1999-2010 | 1,000 |
| VST | 2.5 | CAM | 60'x60' | 2005-2010 | 2,300 |
| VISTA | 4.0 | Optical | 96'x84' | 2005-2010 | 12,600 |
| Magellan | 6.5 | IMACS | 27'x27' | 2005-2010 | 1050 |
| MMT | 6.5 | Megacam | 24'x24' | 2005-2010 | 700 |
| 3DT 8.4m | 6.9 | 180' dia. | 2007-2010 | 2x42,000 |
1 Equivalent area covered to R=26 AB (5, 2" aperture) in a single filter assuming 20% of time is utilized for imaging. In the case of the VLT Survey, VISTA and 3DT 8.4m imaging telescopes, it was assumed that only 60% of the time was used for imaging. The benchmark used was that the Mayall + MOSAIC reaches this depth per field in 2 hours (including overheads), and therefore covers roughly 1.4^ per clear night.
As an example, the NOAO Deep Wide Field Survey is obtaining images in BW, R, I, and K of two 9^ patches to a depth of 26 AB mag in the optical bands. The source catalogues generated by this survey are of the order of 105-106 per sq.deg. For subsets of objects, the source densities remain high: z>~3 galaxies, for instance, are ~4000/sq.deg., and emission line galaxies selected from complementary narrow-band imaging surveys in these fields have source densities to ~1500/sq.deg. (for sources with line fluxes brighter than ~3x10-17 erg/f/cm2).
The Need for Highly Multiplexed MOS Spectroscopy: Reaping the science benefits from these surveys requires a vast amount of spectroscopic follow-up. Although a few of the surveys are imaging-only programs (e.g., the weak lensing surveys), the majority of the surveys currently underway are designed to pre-select certain types of objects for detailed spectroscopic study. The surveys designed to investigate galaxy formation and evolution, the evolution of large-scale structure, cosmology (e.g., using SNe), Galactic structure, cluster searches, high-redshift ellipticals, the red-envelope galaxy population, radio sources, etc., all fall into this category. Imaging alone may indeed be adequate for a few projects. However, in order to understand the astrophysical processes, spectroscopy is essential. In general, the imaging surveys are only the first step: spectroscopic follow-up is critical.
| Survey | O/IR | MABlimit | Area | Completion |
| mag. | sq. deg. | Date | ||
| 2MASS | IR | 15 | 42,000 | |
| SDSS | O | 23 | 10,000 | |
| MACHO | O | |||
| LCO/UCO | O | 22 | 130 | |
| CFHT-RZ | O | 25 | 50 | |
| BTC40 | O | 25 | 40 | |
| NOAO-opt | O | 26 | 18 | 2001 |
| NOAO-IR | IR | 21 | 18 | 2001 |
| Hawaii | O | 25 | >15 | |
| EIS-Wide | O | 23 | 17 | |
| Deeprange | O | 23 | 16 | 1997 |
| CNOC2+ | O | 24 | 6 | 2001 |
| BFQS | O | 25 | 7 | 1998 |
| CNOC2 | O | 24 | 1.5 | |
| CFDF | O | 25 | 1 | |
| CFHT-VRZ | O | 26 | 1 |
The current suite of spectrographs available at large-aperture telescopes will barely begin to meet the need generated by the current imaging surveys. This is clearly evident from the fact that time requests for spectroscopic follow-up of imaging surveys from HST and various 4-m and 5-m telescopes alone exceed the time available on the Keck Telescopes. Moreover, a typical LRIS + Keck night results in low resolution spectra of about 15-30 objects with R~24.5. at a signal-to-noise ratio too poor to yield much other than basic redshift information. To go beyond this and study the internal dynamics, chemical composition, stellar content, age, and mass of the galaxies requires a significantly larger investment of observing time. Although the existing large-aperture telescopes will provide pathfinder science by spectroscopically observing manageably small, carefully selected samples from the imaging surveys, they will not be able to carry out spectroscopy on the scale that is required for a comprehensive understanding of the astrophysics.
Finally, the most interesting discoveries are sometimes serendipitous (e.g., the first galaxy at z>5); spectroscopy maximizes the discovery potential of a survey.
Synergy with Ground-based Radio/Sub-mm Observations:
A highly multiplexed spectroscopic capability provides a necessary complement to ground-based radio initiatives. For example, by virtue of the negative K-correction at millimeter wavelengths for distant dusty galaxies, future deep Millimeter Array (MMA) surveys have the potential to detect very high redshift sources (z~5-10). The existence of an abundant dusty star-forming galaxy population is supported by recent deep surveys at sub-millimeter wavelengths (e.g., Smail et al. 1997). For the faint sources detected in deep pointings with the MMA, optical and near-infrared spectroscopy will be essential in order to determine their redshifts and relationship to known star-forming galaxies at high redshift. The source densities of these objects is likely to be very high: a typical 1 hour observation with the MMA will identify sources to ~100 microJy at 850m (>10 sigma); the source densities to these flux densities are unknown, but current predictions (e.g., Franceschini et al. 1998) suggest >50,000/sq.deg.!