The Evolution of Galaxies to a Look-back Time of 4 Gyr  --  RIG
======================================================

HI can be currently detected in emission from extragalactic objects
out to a z~1/3, with integration times of order of 10 ksec at AO.
A redshift of one third corresponds to a lookback time of about 4 Gyr,
a significant evolutionary step. An important, unexplored science
niche untouchable by any telescope other than AO at this time
consists in the study of the evolution of the HI mass to optical
luminosity ratio, which is linked to the evolution of the cosmic
star formation and merger rates.

HI sources at distances > 100 Mpc are most unlikely to be starless, as
the masses required for their detection with AO correspond squarely in
the category of giant galaxies, for which no theoretical expectation
exists of any inhibition to profuse star forming activity. Thus the study
of such objects is best done via targeted, rather than blind searches.

The Sloan Digital Sky Survey (SDSS) is delivering optical photometry
and spectroscopy for millions of galaxies. The one important datum
SDSS is missing is gas content. A survey of SDSS/GALEX sources over
the  redshift range 0.02<z<0.04 is being currently planned (GASS),
which will target abotu 1000 sources. This study, which will fall in
the category of "large surveys" as it will require on order of 1000
hours of telescope time, will characterize very well the properties
of galaxies at z~0. A proposal is in preparation by an international
team led by David Schiminovich of Columbia and Guinevere Kaufman of
MPIfA/Garching. An even more ambitious effort, aimed at characterizing
the evolution of galaxies back to 4 Gyr, will likely follow a few
years hence. Such an effort will likely require several thousand hours
of AO time and be carried out as a legacy survey with broad community
participation. 

Technical challlenge: because it will be a targeted search, multifeed
capability is not necessary to multiplex sky coverage. However, special
attention will be required to dealing with rfi. Dual beam option, as
well as sophisticated rfi excision and managing techniques will need
to be applied. I think a committed, intelligent approach to dealing
with this problem is of paramount importance. Obviously, minimization
of system noise is also desired