A very-very wide survey
Ray Carlberg

Abstract: The scientific case for a very-very wide survey has, if anything, grown stronger with time. The primary scientific goal is to improve the measurement of the Baryon Acoustic Oscillations in the galaxy power spectrum at low redshift. The resulting constraint on the dark energy parameter w is almost orthogonal in its errors with the CFHT supernova legacy survey measurements thereby significantly enhancing the power of the SNLS measurements. The first BAO measurement used the SDSS redshift survey of 48,000 luminous red galaxies (the statistics are primarily limited by the volume sampled, not the number of galaxies) to make a 2.8 standard deviation measurement of the BAO signal. Modeling has shown that an imaging survey alone can make the measurement with about ten photometric redshifts having the same statistical power as one spectroscopic redshift (the feature has a redshift width of about 0.03, so spectroscopic precision is not essential). The AAT is currently undertaking a spectroscopic measurement using quite blue galaxies, whereas CFHT would concentrate on redder galaxies and be a complementary measurement in the North. A survey of about 5000 square degrees would give about 100 million galaxies at a depth of about m I ~23 mag. It would be possible to do one pass of that area in a year. The huge area would put the survey on top of the existing SDSS data and other surveys such as UKIDSS, immediately enabling considerable secondary science. Furthermore, sensible time sampling of the survey would yield time sampling of interest to solar system and galactic structure studies. CFHT remains the most powerful telescope to undertake such work for the next few years, but with a window that is clearly closing. To fully justify and plan such a survey optimally requires careful consideration of many factors. An important outcome of the User Meeting is whether there is sufficient interest to merit further work.