Re: Science goals

[Ray Carlberg <carlberg@moonray.astro.utoronto.ca>]

There is huge potential interest in a multi-band survey! But, let's be
really ambitious about this to undertake something of really wide and
lasting interest. This is probably in the 1000 square degree
range, assumiing we stick to the 5 Sloan filters.

It seems to me that we have three choices:
o another 2-3 fields like XMM-LSS
	this is "more of the same" and lacks all the IR and X-ray
o many small fields to the XMM-LSS depth
	this is "more of the same but not nearly as useful"
o a shallower survey which covers much area. Note that the nature of
	number counts (stars or galaxies) is that we always get more
objects if we reduce
	the flux level.

The guiding science then becomes a "census of the galactic halo" (and
it is of course useful for KBOs, white dwarfs, and cosmology).

The one major constraint is the Sloan Survey. The main SDSS does not
cover the XMM-LSS field. It has a nominal exposure of about 50
seconds, equivalent to about 25 seconds on CFHT (assuming same image
quality, which is of course not true). Their nominal depth in R is
about 23-24 mag, which is enough to pick up main sequence stars out to
about 10kpc, and sub-giants etc over the entire galactic halo. They
do have a significant issue with star/galaxy separation at the fainter
magnitudes where we can make quite a large impact.

An "interesting" exposure time then is about 30 minutes to 1 hour. At
30minutes we can do 1280 square degrees in all 5 filters, plus repeat
the entire area again in 5 years in say r, to get the proper
motions. That would cost 3840 hours, or about 480 nights.  The
limiting magnitude would be roughly 25.5 in r (some secure numbers
needed!).  Of course 640 square degrees could be done to a 0.3 mag
deeper, but the payoff is pretty low. My one concern in this is the z filter
which looks pretty slow on megacam relative to CFH12K. For the purpose
of both very late type dwarfs (and cosmology) this is of huge importance, 
and it could cost as much as 2 hours to get to the appropriate depth.

A reasonable way to layout 1200 square degrees is a strip about 80
degrees by about 15 degrees. That surely covers quite a lot of
everything in the galaxy! It should cross the NGP (which, from Hawaii,
is a better place for a supernova and other faint variable/moveable
survey). Some of this strip could be placed in the South, or, it could
even be broken into a few somewhat more square chunks. 

A huge advantage of this is that it is "behind Sloan"  so can be calibrated
to the same magnitudes and takes advantage of all their spectra.