Prospects for future surveys in astronomy

Martha Haynes  5/13/2007

Surveys have played a critical role in astronomical discovery
providing unique, widely-used datasets to the community at
large. In some cases, the science resulting from a survey
is intimately tied to the design and execution of the
survey; in other cases, the survey dataset winds up
being used for entirely unanticipated science.
All national and to some extend private ones, facilities, both 
ground- and space-based, endorse the granting of significant 
blocks of telescope time for surveys. Extreme examples include
the 2MASS and SDSS telescopes which were largely dedicated to
acquiring the data for those surveys. National facility surveys
include the NVSS and FIRST VLA continuum surveys and HIPASS and
the surveys conducted under the Spitzer Legacy and Hubble 
Treasury programs.

To start, we need to define what we mean by a "survey", and in particular,
to distinguish "survey" from a "large proposal". The latter is any observing
program which requires at least N-hundred (differs at different telescopes)
hours of observing time. At Arecibo, there are "large programs" and 
"ALFA surveys" which have somewhat different requirements according
to posted documentation. A relevant reference is the talk that Tom Soifer
gave on "Spitzer: Lessons Learned" at the 2006 NRAO workshop on 
legacy surveys. According to Sofier, a "survey" is a large coherent 
science investigation which is not reproducible by an reasonable 
combination of smaller observing programs and which produces a
publicly accessible, unique dataset. He also notes that BOTH the
raw and pipeline processed data from the Spizer legacy surveys
were made available to the public with no proprietary time,
so that followup and archival studies were immediately enabled.
HST began with its "key projects" in the 1980's; these were projects
deemed to be so critical that "we'd better do them first, just in case"
(i.e. in case HST did not last long). The second generation HST surveys
fall under the Hubble Treasury program where again the undertaking
of surveys was viewed as the most strategic use of the facility and
the idea was to provide legacy datasets of value and relevance to the broad
community for years to come.

The optical astronomy community has significant plans for surveys,
beginning with the extension of SDSS into the area of the galactic
plan, sampling at selected longitudes (SEGUE) and a comparable high
latitude survey in the southern hemisphere (with the new VISTA telescope). 
An example of one that will use dedicated NOAO time is the Dark Energy 
Survey (DES) with its a 500 Mpixel camera, DES is slated to use
1/3 of the observing time on the CTIO Blanco 4m telescope for
5 years. Two very large optical synoptic facilities have been proposed.
One, Pan-STARRS, would use an array of 1.8m telescopes while
the other LSST, would use a single large telescope. For both,
the objective is to scan the entire sky on regular, short intervals
(e,g, 1 week) with the exact strategy of return dependent on the
scientific objectives of several different surveys. In contrast,
to SDSS, PanSTARRS and LSST are designed to explore the time domain,
although coaddition of frames of non-variable sources will also
produce deep images of the entire visible sky (sites: Hawaii and
Chile, respectively).

Several large sky surveys have been undertaken with radio
synthesis telescopes VLA, ATCA, WSRT and more are in the planning.
They will focus largely on source structure and, where resolved,
structure evolution (expansion, astrometry). The study of transients
is also a target for the ATA, LOFAR and the two arrays being constructed
at Mileura. 

Arecibo surveys should exploit its large collecting area, spectrometer 
capabilities, commensal observing, but we must recognize limitations in 
angular resolution, field of view and rfi environment. It is fairly
obvious that Arecibo can continue to excel in areas of time domain
science and, as discussed by Riccardo, in the exploration of evolution
of the HI content with redshift. 

In addition to thinking about the scientific applications, we should
also discuss how NAIC needs to organize itself and the community to
undertake such surveys. Programs to explore technical solutions to
rfi excision and the design of special purpose instrumentation need
to be established. Data management needs careful consideration and
planning. Personnel continuity and funding are real issues which
need to be addressed with the cooperation of the NSF. Some of these
issues are not unique to Arecibo, but we should take the lead in 
laying the groundwork for future surveys on behalf of the AO community.
This would be an excellent topic for discussion at the "Future AO Science"
workshop.