SPACE TELESCOPE USERS COMMITTEE REPORT
4-5 APRIL 2005
The Space Telescope Users Committee (STUC) met on April 4th and 5th
2005 at the Space Telescope Science Institute (STScI) and the Goddard
Space Flight Center (GSFC)
Attended: David Axon, Martin Barstow, Martin Elvis (Chair), Eric Emsellem
Laura Ferrarese, Pat McCarthy, C. Robert O'Dell, Regina Schulte-Ladbeck.
Unable to attend: Monica Tosi, Donald G. York
The STUC welcomed reports on the state of the Hubble Space Telescope
(HST) and the Institute from the STScI Director, Steven Beckwith;
NASA GSFC Center Director, Ed Weiler; NASA HST Project Scientist,
Dave Leckrone, Assistant Project Scientist, Malcolm Niedner;
NASA HAS Program Manager, Preston Burch; STScI HST Mission Office
Head, Rodger Doxsey; STScI Science Policies HEAD, Duccio Macchetto;
STScI staff scientists Gerry Kriss, John Mackenty, and Keith Noll.
1. Institute Outlook [Steve Beckwith]
The STUC is pleased at the continued pace of HST discoveries as
reflected in the Science News metric and the output of refereed papers.
The STUC welcomes Jim Manning as the new head of the Office of Public Outreach.
HST future is uncertain at this time. While the NAS/Lanzerotti report
recommended that Hubble be serviced by the shuttle, concerns were
expressed about the viability of a robotic mission. Perhaps as a
consequence, NASA has officially canceled plans for robotic serving.
The new proposed NASA's budget includes no servicing on any kind for
HST, in fact no funding provision is currently made for a de-orbiting
mission, which must be carried out to allow HST to re-enter the
atmosphere in a secure and controlled fashion. Congressional action on
the budget and confirmation of the new NASA administrator will extend
over several months, and we are unlikely to know the outcome
immediately.
It was noted that a robotic servicing mission which will install both
COS and WFPC3, recharge the batteries, and attach a de-orbit module to
HST, will cost NASA about 1.3 billion dollars. This is only twice the
cost of a de-orbiting only mission, which must be performed.
2. HST Status and Outlook [Rodger Doxsey]
The STUC is pleased to see that operations continue at normal efficiency,
and that the rework of the Archive has reduced access time to a mean
of a 1/2hour for direct retrievals not requiring processing.
A TGM test was run on 21-23 February 2005. Gyro 1 was left on but not used
during the testing. The main test was done using the ACS/HRC/F555W and an
exposure time of 100s, showing virtually no degradation in
performance. The results do not seem to be impacted by the magnitude
of the guide stars, i.e. the increase in the PSF FWHM as the
magnitude of the guide star is decreased is significantly smaller
than natural PSF variations which occur within an orbit due to focus
breathing and jitter (such changes can be of order 0.2
pixels). Further testing was done using exposure times between 50 and
500 seconds. The mean FWHM was 2.009 pixels (compared with 1.98 for
three gyro mode). Tests were also done with NIC1/F110W, and no
degradation was noticed. More tests were done with the coronographs on
both ACS and NICMOS.
The STUC heard a report of the impressive tests of Two Gyro Mode,
which have led to no degradation in the PSF, even when using
V=13 guide stars. The implications are that while sky availability
will be reduced by Two Gyro Mode, orient restrictions in the available
sky will be unchanged. Our congratulations to the Institute and the Project.
Given that the risk of turning off gyro 1 does not appear to be high,
the STUC agrees with the TAC that going to Two Gyro Mode early is a good plan.
We looking forward to the GSFC and HQ decisions expected in Mid-April.
[UPDATE: NASA has asked STUC to convene a special meeting, which will
be held on Weds May 4th, to gain assurance that the user community is
aware of and willing to take the risks associated with turning off gyro 1.]
3. Cycle 14 TAC
The Cycle 14 TAC process was completed successfully, with oversubscription
still high, though slightly lowered, probably due to loss of STIS. Cosmology
and galaxies still dominate the science areas, with most observing time being
with the ACS.
Very few of the Cycle 14 programs are impacted by TGM. If TGM is
activated, the switch-over date will be chosen to minimize impact on
Cycle 13 programs, and will likely be around August 1st, i.e. after
the Deep Impact Observations (scheduled for July 4). Most remaining
cycle 13 programs will continue without GO intervention. STScI has
already contacted the PI of the cycle 13 programs which might be
affected.
3a. There was concern about the factor ~4 underuse of Chandra and Spitzer time
awarded by the HST TAC. HST allows Chandra and Spitzer to award 100 orbits/
observatory. For comparison some 150 orbits are awarded by each HST panel. Most
HST-awarded Chandra time is for coordinated observations, while much
CXC-awarded HST time has no time constraints. STUC is concerned that the time
awarded for Observatories other than the main one a TAC is convened for is not
vetted at same high level as HST-only time, for example. The TAC finds that the
'double jeopardy' of having to convince two TACs is in general reasonable,
given the scarce resources being allocated. An exception is time critical
coordinated observations, including ToOs. For this purpose we support the
retention of the joint awarding of time. We suggest that the Institute look at
the productivity of jointly awarded time by means of the standard metrics. With
these results a review of the use of this time should be undertaken with the
other Great Observatories. The STUC would like to hear the results of this
review at its next meeting.
3b. STUC members who were on the TAC noted some confusion about TAC
instructions. (In particular whether panel members were expected to read all
proposals or not, and to send comments prior to the meeting). We are pleased
that new summary instructions will be produced for Cycle 15 that briefly
highlight the most important tasks and responsibilities of panel members.
3c. The purpose of the 'Previous Observations' section of HST proposals has
become unclear: Is it to demonstrate effective use of previous awarded time, or
to show that the proposers are knowledgeable about the use of HST? STScI
report that the vast majority of proposals lead to published papers in refereed
journals, mostly by the original team. (STUC would like to see these numbers.)
Hence the first possible use may be unnecessary. We suggest that STScI look at
options for this problematic section. Possibilities include: (1) Clarify the
purpose and contents [e.g. "include only PI papers published in refereed
journals for time awarded in the last 3 cycles"] of the section, both in the
call for proposals and with the TAC; (2) Change the section to be 'Team Roles',
as for Spitzer, or 'Previous Team Experience' in similar science; (3) Drop the
section entirely. (4) Substitute an STScI print out of PI/co-I accepted
proposals and papers published, using Archive/ADS links. The STUC hopes to
hear the Institute's considered opinion at its next meeting.
3d. ARCHIVAL PROGRAM
The Archival program proposals grew in Cycle 14, and this may be a
long term trend. A difficulty is the support on postdocs on 1-2 year
grants. The approximate distribution of GO funding is currently:
Observations ~$20M, ARCH+THEORY ~$3M, FELLOWS ~$2.6M. The STUC
discussed at length the balance between these areas, and within the
Archive area as a function of proposal size. No recommendation to
change balance. Can't solve scarcity of long-term funding in
astrophysics, highlighted by the cancellation of the FY05 LTSA/ADP
programs. This is a larger question beyond STUC's scope.
4. STIS Closeout
The closeout plan for STIS, to create a legacy quality archive,
appears to be well thought out, having a good schedule, FTE loading
and deliverables. It would be useful to have the new presentation
available on the web, and STScI are happy to do this. It is important
to gather the STIS expertise into products, software and documentation
before the teams disperse. A servicing mission would delay this
schedule. In view of the recently announced FY05, FY06 cuts we
encourage review of low priority items in view of the resources likely
to be available. The STUC members involved with STIS (Axon, York)
volunteer to give input to this process.
A concerted effort must be made to amalgamate the best practice of the
IDT and the institute in legacy pipeline and tools for UV spectroscopy
data. This will require minor additional funding from NASA for the IDT.
The STUC is pleased that, following a user request to the STUC,
non-standard STIS observations have been classified and that a policy
is being developed for their incorporation in the archive. We urge
STScI to work with original observers in the more difficult cases to
deposit their data products in the archive in a useable form.
For future instrument closeouts the STUC asks STScI to call for expert
user input at early stage of closeout planning (e.g. by email, web
clearinghouse). (The original drizzle web page is a good example.)
5. NICMOS Photometric non-linearity
Comparison of NICMOS/NIC3/grism and STIS fluxes, in the wavelength
region where they overlap, shows that the NICMOS/STIS flux ratio
depends strongly on the NICMOS (or STIS) flux counts. Comparison with
ACS data shows that the problem lies with NICMOS. The effect seems to
be wavelength dependent, occurring primarily at short wavelengths
(8000 A). No bias is noticed at 1.6 microns.
MULTIACCUM data show a deficiency in the count rate in the first few
steps of the ACCUM. Such count loss would result in underestimating the
NICMOS counts, and is likely the result of charge traps which exist in
crystal defects at the interface between the sapphire substrate and
the semiconductor layer. The charges are trapped at the beginning of
the MULTIACCUM sequence; when an equilibrium is reached depends on
the brightness of the source (brighter sources are less affected).
In practice, the problem is most severe for observations shorter than
the 160s, which represents the trap-release time. Few science
observations are shorter than this; therefore most data would not be
affected.
The discovery of this bias might prompt a change in the
recommendations for best observing strategies. At present, STScI
recommends to break up longer observations in several short dithered
integrations, although this might not be the best practice, at least
at short wavelengths.
Finally, there seems to be a NIC/F110 - J ~ 0.15-0.20 mag offset in
the photometric zero points (by comparing NICMOS and ground based J
photometry for targets in the HDF). The F160W magnitudes seem to agree
with the H band ground based magnitudes.
The STUC is pleased that this effect was uncovered and that a solution
is available.
6. Project Visit at GSFC: HRV facility, Project Status, Meeting with Director.
The STUC visited GSFC to hear from the Project and to view the lab where
robotic servicing work is progressing. The hardware shown was: (1) a full scale
volume mock up of the `ejection module' which would carry the new instruments,
the DEXTRE robotic arm and it tools, and in which the old instruments would be
placed after removal. This module would then be detached from HST and
de-orbited; (2) a working DEXTRE being operated 'remotely' (from the lab floor
but out of sight of the robot) to detach cables from their original connectors
on HST and reattach them to new connectors; (3) mock-ups of COSTAR, WFPC2, WF3
and COSTAR with attachments that DEXTRE can use or install itself. STUC watched
the successful detachment and reattachment of connectors.
The STUC thanks the HST Project for hosting a tour of Hubble Robotic
Vehicle (HRV) test facilities. We were impressed by the technology
demonstrations, and by the spirit, skill and can-do attitude of the
team. Our congratulations on highly successful PDR for HRV.
The success of the HRV PDR gives sense of reality to effort. We
wish them well for their CDR in October.
6a. Dr. Ed Weiler, GSFC Director, met with the STUC. GSFC has not received
orders to stop work on HRV, while Congress has required NASA to spend the $291M
allocated for this purpose. Hence work is continuing at GSFC on HRV. The cost
of Hubble servicing is widely talked of at ~$1.2B. This includes the Shuttle
launch for a manned servicing mission. However NASA has stated that the cost of
servicing was already budgeted, and a de-orbit mission has to be in the longer
term budget to comply with NASA's safety rules. Compared with these two
combined, and allowing for the cost of maintaining the core experience until a
later de-orbit mission, HRV requires only a relatively small extra investment
and would return much new science. Dr. Weiler feels HRV team has met all PDR
goals, and is on track for CDR in October.
STUC heard a report from Preston Burch, the HST Project Manager:
6b. HST status: HST is now healthy. Reports that 'Hubble is dying' are
misinfomed. Battery life predictions 6-months ago now appear too pessimistic;
the rapid rate of loss of capacity after the last servicing mission has not
continued, and the rate is now similar to before servicing. In addition some
tests of new procedures for fully dicharging and recharging cells, using the
`comparison sample' batteries at MSFC, indicates a better recovery. The net
result is that battery capacity is expected to remain above the safe level to
mid-2010, rather than the previous estimate of late 2009. This gives a 5-year
window for servicing HST while under attitude control. The success of the Two
Gyro Mode (TGM) tests suggests that a deliberate switch to TGM around August
2005, would extend the gyro operational life to mid-2008, a gain of 9 months.
Implementation of 2GM opens the possibility of a One Gyro Mode in future.
6c. Servicing: GSFC is studying two options: the 'Hubble Robotic Vehicle' (HRV)
which includes servicing the gyros, FGS and instruments (the hardware for which
was shown to the STUC, see above); the `De-Orbit Only Mission' (DOOM) which
attaches only a de-orbit engine to Hubble. DOOM uses many elements from HRV
(autonomous rendezvous & capture, grapple arm) but with many fewer than the 40
cameras on HRV as there is much less fine scale work to be done. DOOM needs
only a Delta II launch, while HRV needs a Delta IV or Atlas V. Both options
recently passed their PDRs and are proceding to CDR in October 2005.
The STUC strongly supports HST servicing. The STUC finds that WF3 and COS are
vital to continued progress in astrophysics with Hubble. We also recognise
that restoring STIS to operations, while a technical challenge, would have
major scientific benefits. The STUC is pleased that more than one servicing
option exists. STUC considers DOOM (De-Orbit Only Mission) to be a
short-sighted approach.
6d. The HST MO&DA has been directed to be reduced by $20M this fiscal year
(FY05). This is about 25% of the budget. An additional $12M cut has
been directed for FY06. FY05 impact is worsened as the fiscal year is
half way through. The STUC is alarmed to see such significant cuts
applied in the middle of a fiscal year. The STUC is of the opinion
that the GO program must be protected. A secondary concern is the
impact on user services and support.
6e. The WF3 IR background was very worrying. We are relieved that a
solution has been found within budget, and that result is a major QE
increase. The STUC notes that the independent and timely evaluation of
the detectors has been crucial to the successful development of WF3
instrument. The STUC is looking forward to an update on the WF3 IR
detectors status at the next meeting.
6f. Dave Leckrone reported on the process by which a decision on the
deliberate switch to TGM would be made. He also reported on the new
`HST Year Book' which shows science highlights from Hubble for the past
year. The STUC agrees that the having the selected authors of HST Year Book
science highlights be all white males was highly regrettable and does
not properly reflect the HST user community. The community reaction
was strong; many members of the STUC each heard from several members
of community about this, an unusual circumstance. We also note that
the great majority of astronomers came from the Baltimore-Washington area.
The proximity of the ACS team and the large ACS projects based at the
Institute makes such a concentration plausible, but again fails to
reflect the national and international nature of the Hubble user community.
We are pleased that steps will be taken to avoid similar situation in future,
and suggest that an outside panel be convened (using phone and email) to
generate the next list of highlighted science.
7. FUTURE DATA PROCESSING PLANS.
A level of effort move is underway, to produce enhanced archive
products. The STUC applauded this move at its previous meeting.
The STUC is worried about the impact of the just announced MO&DA cuts
on this vulnerable area. Servicing would not have a major impact on
these plans as the scheduling and pipelines for COS,WF3 are ready.
(The impact would be in calibration and documentation.) The STUC
offers to read the draft plan and provide input on priorities and
scope (by email).
8. CONTACT SCIENTIST AND PROGRAM COORDINATORS
STUC members report some problems with Phase 2 support. With Two Gyro
Mode there will be more scheduling issues, requiring Program
Coordinator and Contact Scientist coordination. STUC is pleased that
the Institute recognises that there are some issues in this area and
that it will be taking positive steps to improve the process.
STUC considers that a thorough review of proposals by Contact scientists
and Program Coordinators is instrumental in improving Hubble science.
9. CAPTURING COMMUNITY EXPERTISE
Several suggestions were presented for capturing expertise in the
community at a level less than that of fully robust pipelines: (1) A
contributed software page, as at the CXC; (2) A 'contacts' page for
users to advertise their expertise in specific areas; (3) A
'contributed memos' page for users detailing procedures they use to
solve data analysis problems. These all sound like good proposals to
the STUC. The STUC notes that capturing STScI expertise in software,
products and on-line documentation is also important for old
instruments in order to guard against inevitable loss of knowledge as
people move on. The same methods may be useful in this case too.
10. THANKS TO THE OUTGOING DIRECTOR
The STUC thanked the outgoing STScI Director, Steve Beckwith, for
his work, in particular his helpful dealings with the STUC.
11. SUMMARY OF REQUESTS FOR INPUT AT NEXT STUC MEETING
- Update on servicing options.
- Update on budget situation.
- Updates on TGM decision, operations; HST lifetime estimates.
- Cycle 15: Report on Joint HST/Chandra/Spitzer time productivity and
discussions with other Great Observatories; steps to clarify instructions
for TAC panels.
- STIS closeout prioritization (with input from STUC members D. York,
D. Axon), progress; IDT input to legacy pipeline tools.
- Update on WF3 IR detectors.
12. DATES OF NEXT STUC MEETING:
The next regular STUC meeting will be held on Tues-Weds 25-26 October
2005.