Subject: notes from workshop on deep imaging with LSST

From: Sidney Wolff

Submitted: Thu, 5 Dec 2002 14:00:58 -0700

Message number: 29 (previous: 28, next: 30 up: Index)

Following is a summary of the first LSST science workshop held at
NOAO, which focused on deep imaging.  The goal of the workshop was
to look at a range of science problems and to determine whether
they could in fact be done with the baseline telescope, instrument,
and operational mode worked out at Aspen and discussed with
many audiences by Tony Tyson.  This was not an effort to develop
a detailed scientific justification or to determined a detailed
set of specifications but rather to make sure that there was
likely to be sufficient top level agreement to make it possible
to proceed with optical design and concepts for the telescope
and instrument.

A second workshop dealing with the time domain and solar
system studies was held in April, and I will write that up
later.  A final workshop on galactic astronomy is being planned
for January or February and input from that workshop will be
provided to the SWG in order to help with the evaluation of
whether this kind of science is a driver and if so how it
affects the LSST requirements.

The meeting summarized below was held in March, 2002
and focused on problems that could be explored through deep imaging.
Participants included Chuck Claver, Andrew Connolly, Ian Dell'Antonio,
Dave DeYoung, Richard Green, Buell Jannuzi, Tod Lauer,
James Lowenthal, Jeremy Mould, Marc Postman, Abi Saha,
Michael Strauss, Tony Tyson, Sidney Wolff, and Dennis Zaritsky.

Since I am not an expert in this area of science, I welcome
corrections, but this summary should give the flavor of what
was discussed.

SCIENCE

Tony made a presentation on deep lensing, which is amplified in the
recent e-mail to lsst-general.  The science highlighted included:
1) tracing dark matter structure over 7 Gyr; 2) the determination of
the equation of state as a function of time; 3) observations of a
billion galaxies with known photometric redshifts; 4) observations
of 100,000 mass clusters; 5) the determination of mass structure over
wide scales (well beyond the few magparsecs of current studies) and the
sensitivity to the growth of structure.  He noted that the current
DLS (deep lens survey) resolves more than 50 percent of the galaxies
observed.

He favored short exposures to control systematics.  Short exposures
are consistent with the requirements for searching for NEAs, since
~20 sec exposures are thought to be about the maximum desirable in
order to avoid trailing.

Strauss talked about both the science and the lessons learned from
the SDSS.  He showed that it will be possible to do semi-quantitative
quasar science to z = 5-6 and to obtain 3-d maps and luminosity
functions using photometric redshifts, but with substantial scatter.
He argued that going fainter, as LSST will, makes it possible to extend
the survey to lower luminosities but not greater distances.  Variability
properties would provide a possible new method for selecting quasars.

He noted that the u-band is very important for picking out unusual
objects, including stars.  U is essential for low redshift quasars.  Strauss
also commented that SDSS has not found any truly strange
objects.

Strauss also stressed the need for spectroscopic follow up for many kinds of
science.

AGNs were described as not being a driver for the LSST but rather an
area of research that would benefit from the survey.  AGNs are useful
tracers of over density; studies of the power spectrum of flickering might
yield useful contraints.   Black hole evolution is an important subject,
but it is not clear how to obtain observational constraints.

Lowenthal talked about star-forming galaxies and the structures that
they trace at redshifts > 2.  Supernovae in Ly-break galaxies might
provide clues about star-forming histories; there is enough time for
SN Ia at z = 3.  There is also some possibility of using SN Type II as
standard candles, but only if spectra are available, say with GSMT.

Coupling LSST photo-z's with LMT/SIRTF/ALMA would get around the dust
extinction problems and enable tracing the galactic mass assembly history
of the universe further down the luminosity function of Ly break
and SCUBA galaxies.

Postman suggested that LSST would find interesting numbers of
Z>5 galaxies through lensing.  The large area coverage would suppress
cosmic variance and allow an attack on the problem of clustering as
a function of morphology, the evolution of higher order moments, biasing,
the relation between mass and light clustering.  Clusters and other
rare objects could be used to trace structure on gigaparsec scales;
mass could be estimated from the DM survey.

Strong lensing could be used to search for high redshift (z>4 galaxies)
and select a sample for follow up spectroscopy.  Time domain observations
of strongly lensed objects and SN in arcs also offer potential; they would 
pop off at different times in a quasi-Einstein ring.

REQUIREMENTS

There was convergence on short exposures, either 20 seconds
or two 10 second exposure of each field

General agreement that the initial phase of LSST operations
should provide a multi-color (griz) survey of the entire sky to
26th magnitude

Basic survey should be followed by repeated scans of sky that
emphasize sky coverage rather than color information.  Not
determined whether this repeated coverage requires any color
information or could be done in R only

Orginal baseline operations plan had a deep 1000 degree survey
in early phase of project; this group felt that this survey was
really good only for TNOs (trans-Neptunian objects) and should be
deferred until LSST was fully operational and all the bugs wrung
out.

Strong requirements (see Tony's e-mail) on image quality,
continuous maintenance of focus, control of aberrations,
alignment, control of PSF for weak lensing work

LSST should not be a "Noah's ark" taking on board
a little bit of everything

Need for photo z's; implies multi-color photometry; subsequent
modeling by Andrew Connolly indicates at least 4 filters needed
including B

Big debate about importance of u and whether it needs to be
contemporaneous.  Ratio of exposure times, e.g. u/g about 10.
Question about whether really necessary to reach u of 26 to do
science.  General consensus that, since current design does  not
preclude u, the right strategy is to optimize for the red and
take what we get in u.

Unclear how many filters should be available during the
night, but it seemed that 4 should be adequate with the
possibility of changing the specific 4 not more often than
during a lunation.

SN do require simultaneous color information; how is this
requirement to be met?

The use of red-enhanced detectors will require definition of new filter 
system; can't inherit an existing system

Calibration strategy recognized as problem that needs to be addressed,
but mechanism for doing so not identified; role of an auxiliary telescope
to measure extinction discussed

Many of the needed elements of lens pipeline currently under development

Unclear how much cirrus is acceptable

Questions about what to do about marginal data

LSST LSST LSST LSST LSST    Mailing List Server   LSST LSST LSST LSST LSST LSST
LSST
LSST  This is message 29 in the lsst-general archive, URL
LSST         http://www.astro.princeton.edu/~dss/LSST/lsst-general/msg.29.html
LSST         http://www.astro.princeton.edu/cgi-bin/LSSTmailinglists.pl/show_subscription?list=lsst-general
LSST  The index is at http://www.astro.princeton.edu/~dss/LSST/lsst-general/INDEX.html
LSST  To join/leave the list, send mail to lsst-request@astro.princeton.edu
LSST  To post a message, mail it to lsst-general@astro.princeton.edu
LSST
LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST LSST