[asac] summary on demonstration science

Wed Aug 3 23:04:43 EDT 2005

Hi, everyone,

Here is my summary of information on demonstration science that I had so 
far only circulated to the sub-committee. If you are interested in looking 
at the actual documents, most of them are at

http://physwww.physics.mcmaster.ca/~wilson/www_xfer/charge3.tar.gz

Chris

---------- Forwarded message ----------
Date: Mon, 1 Aug 2005 22:01:30 -0400 (EDT)
From: Christine Wilson <wilson at physics.mcmaster.ca>
To: Lee G. Mundy <lgm at astro.umd.edu>,
     Ewine van Dishoeck <ewine at strw.leidenuniv.nl>,
     Peter Schilke <schilke at mpifr-bonn.mpg.de>,
     Satoshi Yamamoto <yamamoto at taurus.phys.s.u-tokyo.ac.jp>,
     Munetake MOMOSE <momose at mx.ibaraki.ac.jp>, rlaing at eso.org,
     David Silva <dsilva at eso.org>, Alwyn Wootten <awootten at nrao.edu>
Cc: Christine Wilson <wilson at physics.mcmaster.ca>
Subject: summary on demonstration science

Hi, everyone,

Here is my promised summary on some issues to do with demonstration 
science. I plan to circulate this to the whole ASAC tomorrow (Tuesday) so 
that people have time to read it before the telecon on Wednesday. 

So, please read it and respond; if you can respond before 2100 UT on
Tuesday, I will be able to incorporate your comments before I send it
around to the whole ASAC.

Thanks,
Chris

----------------------------

Some initial thoughts on Charge 3
Chris Wilson, 1 August 2005

The text of Charge 3 for the October 2005 ASAC meeting is as follows:

"Following from your Sept. 2004 discussions, the ASAC is invited to 
continue
developing proposed "demonstration science" guidelines or policies. The
ultimate goals include:
      a. providing a proposed framework (rationale, principles) that
         establish the value to the astronomical community of
         accomplishing demonstration science as you envision it during
         the early operations era; and
      b. facilitating evaluation of the concept, its proposed timeline
         and the planning for its implementation during a period when
         demand for ALMA construction and operational resources will
         likely be high"

In this note, I will try to synthesize the issues as I see them related to
demonstration science. This discussion is based on a summary of relevant
information from several documents relating to demonstration science. I
have already circulated that summary to the Charge 3 subcommittee and I
can make it available to the whole ASAC if desired.

The concept of "demonstration science" first appears in the May 2004 ASAC
report, where we recommended that the ALMA project focus on exploiting its
high resolution and high frequency capability to maximize its early
science impact. We recommended that these projects have community
involvement and recognized that multi-configuration observations would be
required. We also identified two categories of science demonstration
projects: public demonstration images for publicity and the general
public, and science demonstration projects to demonstrate ALMA's
capabilities to the general community.

In the September 2004 ASAC report, we elaborated on demonstration science
by recommending that they be carried out from end-to-end by a team of
community astronomers and ALMA experts. We also cautioned that it would be
unwise to attempt this too soon, as it would be hard to demonstrate ALMA's
superior capabilities with only a few antennas. We suggested that
demonstration science should take place before the first open call for
proposals and that demonstration science should be delayed until ALMA has
about 16 antennas.

The draft ALMA Commissioning and Science Verification Plan (2004-09-03)
covers the period up to but not including Early Science. It describes
demonstration science as being part of the science verification process.
Science verification, in general, tests the system end-to-end (from
proposal submission to final science) and involves outside observers. The
document also says that very limited reconfiguration will occur during
early commissioning (that is, the commissioning before early science) and
notes that scheduling the multiple array configurations required to
produce pretty pictures for demonstration science is a concern.

So, the fundamental difficulty we are facing is this: we would like
demonstration science to produce pretty images and to demonstrate the
power of ALMA to the wider astronomical community. Ideally, this would
take place before proposals are due for the first open call for proposals
to encourage the community to use ALMA. However, if Early Science starts
with only 6-8 antennas in the array, it will be difficult to make pretty
images with good publicity value. So the question becomes, how important
is it to spend a lot of time moving the telescopes to different
configurations so that we can make nice, sensitive images with lots of
spatial scales? And is this important enough to delay the start of Early
Science, or to shift project effort from commissioning and science
verification to demonstration science?

I find the wording of the Charge from the Board difficult to interpret,
but here's my best guess. In general, they want us to continue developing
guidelines or policies for the proposed demonstration science. One thing
we might consider is whether the "pretty picture" side of demonstration
science should be deferred until ALMA has more antennas and making images
is easier. In this case, we would be putting off the publicity side of
ALMA for a little while. But I think we could still demonstrate what ALMA
can do to the astronomical community by what will come out of science
verification. Each mode of ALMA that is used for early science will be
tested by one or more projects in an end-to-end fashion. Some of these
projects (particularly the high frequency ones and high resolution ones)
may be sufficient to get the wider astronomical community interested in
ALMA. It seems pretty clear to me that the science verification that is
proposed is absolutely necessary (no one wants to advertize a mode of
observing without having tested it first), and so we shouldn't have to
defend the need for this.

If we go this route, more thinking is required about how the community
involvement in the science verification process is managed. The CSV
document says that the science verification team is responsible for the
selection of a small number of proposals of scientific interest. Perhaps
it would be good to supplement them with a proto-TAC or at least an
advisory group with a broad range of scientific interests to make sure
that the team doesn't observe only outflows and high redshift galaxies
(just to pick two topics at random). We could look at how other telescopes
have managed calls for commissioning ideas and see how they did it.

I think the main thing we lose in the short term if we subsume
demonstration science into science verification is the publicity value of
pretty images. We should not underestimate the importance of producing one
or more splashy images suitable for the general public, funding agencies,
etc. However, we might be able to retain this capability, if only for one
particular image, by careful choice of science verification projects. For
example, I imagine the project will need to commission all the
configurations that are offered in early science. If the ALMA project was
careful to always observe some unique southern source (eta Carina, the
Galactic Center, or Cen A spring to mind) in small mosaic mode with the
appropriate correlator setup in each of the 6 early science configurations
at one of the higher frequencies (Band 7 or 9), that ought to produce a
pretty spectacular image.

The Board says that one of our ultimate goals should be to provide the
rationale and principles that establish the value to the astronomical
community of demonstration science in the early operations era. If we
adopt my scheme above (very limited "pretty image" demonstration science
coupled with thorough science verification of all early science modes), I
think it is very easy to establish the value to the community (although I
think we still need to think about justifying the pretty image). If we
want to stick with the more elaborate version of demonstration science
outlined in the two earlier ASAC reports, we need to think hard how to
justify them.

The Board also says that another goal should be to facilitate the
evaluation of the concept of demonstration science, its proposed timeline,
and planning for its implementation. Again, timeline is easy in the
scaled-back version (the CSV team needs to do science verification in any
case). Ditto for evaluating the concept, I think. Planning for its
implementation needs some work i.e. how to get the word out to the
community, how to pick which project to do, etc.