Chandra Users Committee Report Minutes -- 25-26 January, 2005

Attendees: Vassiliki Kalogera, Julia Lee, Knox Long (chair), Smita Mathur, Chris Mauche, Chris Reynolds, Maria Santos-Lleo, Steve Snowden, Greg Taylor, Allyn Tennant (ex-officio) & Leisa Townsley

The Chandra users committee met in Cambridge on 25-26 January, shortly after one of the largest snowstorms of the last twenty years.  Nevertheless, most members of the committee were able to attend, including new members Smita Mathur, Chris Mauche, Maria Santos-Lleo, and Leisa Townsley. 

The first day of the meeting was devoted to presentations by CXC staff.  Briefly:

• Roger Brissenden described the health of the observatory, which on whole is quite good.  Indeed a recent study indicated that there was nothing obvious that would prevent Chandra to continue to operate for as long as 15 years.  The budget for increases in CTI (charge transfer inefficiency) of ACIS is 5% a year, and experience is that the actual degradation is 3-3.5% per year.  So one could imagine that the CCDs on ACIS will not degrade to unusable levels for a long time.  The guide star CCD is accumulating bad pixels from radiation damage as expected, and there will be some regions of the CCD that will have to be avoided in about 5 years, but this is not considered very serious.  The main spacecraft concerns are the long-term rise in temperature of the EPHIN, used to monitor and react to times of high charged particle background, and a concern that some of the propellant lines could cool below the hydrazine freezing point in certain situations.   Both of these effects reduce the portion of the sky in which, on a specific day, observations can be made, especially when a long exposure is required.
• CXC director Harvey Tananbaum described the use he had made of DD time in the past year.  Departing somewhat from past tradition, he had allocated a large amount of time 400 ksec for a program to observe 12 clusters of galaxies in an attempt to improve determination of the cosmological W parameter to 10%.  The  voiced continued support for the use of DD time for large programs that can significantly increase the overall Chandra science return.  The proprietary period for DD was discussed.  Most have no proprietary period, but the policy is that up to three months can be requested.
• Belinda Wilkes described when in the presentations responses to our last report would be provided during the day.  The committee had in June been concerned that despite the existence of joint proposals, for example for Chandra and HST time, that the TAC had occasionally rejected joint proposals, and needlessly subjected proposers to the kind of double jeopardy that joint proposals are intended to eliminate.  This had resulted in some of the available joint time not being given out. Belinda said that after review, the CXC felt the instructions they were giving to the TAC were appropriate.  She pointed out that it was important that the joint time be of high quality, to continue to justify the existence of joint proposals.
• Patrick Slane described in more detail how the combination of the EPHIN heating problem and the hydrazine freezing problem complicated mission planning.   Monitoring observations and long observations near the ecliptic poles have become especially problematic.  In response, the CXC is reducing the fraction of time-constrained observations from 20 to 15%.  This does not sound too serious, but in fact may be more significant than first appears because the CXC also intends to tighten up the definition of what is time-constrained.  By the new definition, as much as 40% of the Cycle 6 program would have been considered time-constrained.  The committee was concerned that the proposer community has clear understanding of the magnitude of the change.
• Belinda Wilkes described summarized the results of the Cycle 6 TAC and changes that are planned for Cycle 7.  The percentages of proposals being accepted are roughly the same in all categories, 15-25%, similar to that in Cycle 5.  Cycle 7 will be the first time it is possible to propose for coordinated Spitzer observations. Most of the other changes are in response to the increased difficulty of scheduling observations resulting from the spacecraft concerns mentioned earlier.
• Kathy Lestition gave an update on the E&PO activities.  In large measure she described, the CXC's response to the CUC's suggestion that it should make the activities of E&PO more obviously accessible to professional astronomers.   The CUC felt that the new web pages that highlight what is new and what is available will indeed made EPO's activities more accessible, and looks forward to the reaction by the community.  Ms Lestition also described the E&PO grant program; the CUC noted that there were very few individual grants awarded, and indeed that there were very few who proposed for individual grants.
• Jonathan McDowell, who has taken over for Martin Elvis as head of Science Data Systems, described intermediate term plans for CIAO, and responded positively to the committee's request from June that more attention be paid to informing users when they should upgrade, to pileup issues and to telling users how they might use DS9 in a more sophisticated fashion.  It's quite clear that the CXC has taken the pileup issue seriously, and the committee looks forward to the completion of the effort that is underway.  The committee continues to be interested in how one should best advertise GUI tools to the community; adding threads and collecting information on how to do this in more obvious places is a first step.  The committee appreciates that it is not obvious how best to suggest how users might maximally benefit from GUI-based tools, given the widely diverse needs (and desires) of the Chandra User committee.   The CXC is also beginning to embark on a uniform reprocessing of all of the data; the timescale for completion is not entirely clear.  For the most part users can get all the benefits of reprocessing by using the latest version of CIAO; aspect solution may be an exception as there are some changes in this area that are not part of the distribution.
• At the request of the committee,  Janet dePonte Evans described the plans of the CXC to create Level III products from the Chandra data.  The only product that is currently being worked on is a plan to develop a uniform source catalog.  Prototyping is now seriously beginning, but the schedule is uncertain because the priority for this activity remains lower than that given to improving the calibration and calibration tools for Chandra users.
• Paul Green described the large number of symposia that the CXC supports for Chandra.  The 6-year Chandra symposium and the next Chandra Calibration workshop will occur in the first week of November.
• Paul Plucinsky described developments in the Project to understand whether a bake-out of the CCD is advisable.  There have been some changes since the last CUC meeting.  In particular, it is now understood how to carry out the bake-out without use of TSC1 Shuttle Abort Heater on the Science Instrument Module  (which were deemed unsafe to use). In addition, MSFC has revised its model of what the bake-out would accomplish, given various assumptions regarding the properties of the material that one would like to remove..  The main continuing activity is a study to understand the effects of the bake-out on CTI.  When that is complete, a report will be presented to an independent review committee for their recommendation.
• Larry David described a large number of projects that the calibration group has undertaken.    The Committee was pleased to see that projects had been put on a schedule, and were being completed on schedule.  However, we found it difficult to assess whether these were actually the highest priority projects, and how large the steps were to reaching the goals for calibration.  It presumes that this is mainly issue of presentation, but hopes that in future meetings the significance of various issuers can be addressed more comprehensively, and the prioritization of future tasks explained in a way that are easier to understand.  The committee understands that there are two basic types/extremes of activities, one type that maintains the existing calibration, and one type that is actually intended to improve the calibration.  But for the committee to offer reasonable advice on these issues, it needs a presentation that includes this kind of information.
• Herman Marshall and Diab Jarius described various issues that are now coming to conclusion that will improve the ability to properly flux Chandra data in the wavelength range around the Iridium edge.  The committee is looking forward to the incorporation of the results of this calibration into the calibration DB, and if necessary the CIAO S/W

General Conclusions

Based on this meeting, our own experience, and from conversations with colleagues, the CUC applauds the high quality of support provided by the CXC to the community.  Although somewhat concerned about the impacts on observing implied by the rise in temperature of the EPHIN and of the problems associated with possibly freezing of the hydrazine lines, Chandra itself seems in good health and poised to continue to produce breakthrough science for the foreseeable future. 

The CUC appreciates the positive responses given to nearly all of the recommendations made by the CUC following its June 2005 meeting.   We also appreciate the time and effort of all the presenters, and of those who helped to assure that we were able to get to and from Boston without too much hardship.  We have detailed a number of specific suggestions for the CXC to consider for the future.  The fact that these recommendations appear longer in length than our General Conclusions should not, however, obscure the fact that the CUC feels that the users are quite generally positive about the level and quality of user support and enthusiastic about Chandra as a mission.

We have detailed a number of specific suggestions for the CXC to consider for the future.  The fact that these recommendations appear longer in length than our General Conclusions should not, however, obscure the fact that the CUC feels that the users are quite generally positive about the level and quality of user support and enthusiastic about Chandra as a mission

Specific Recommendations:

• The most serious threats identified by the CXC to the overall effectiveness of Chandra involve decreases in the effective field of regard of the observatory (the region of the sky where Chandra can observe on a given day)  as a result of degradation of MLI near the EPHIN and of temperature control of the hydrazine lines of the spacecraft.  While at present this degradation has resulted in only a moderate effect on the flexibility of scheduling the observatory, the CUC is concerned that further degradations will create serious problems in scheduling and consequent loss of science capability.  We urge the CXC to fully explore alternatives such as use of the HRC anti-veto channels, additional reliance on other satellites, and any other alternatives to assure that the observatory remains safe while maintaining the fewest constraints on observing possible.
• As part of the response to the problems associated with the EPHIN and the hydrazine lines, the CXC has reduced the fraction of time that can be orient constrained from 20% to 15%.  In the process of reviewing the existing policy, the CXC has found the definition of time-constrained was murky, and that mission planners regarded more observations as constrained than was assumed during the TAC process.  The CXC intends to tighten the definitions of time-constrained as a result.  While this may be necessary, it implies that the decrease in formal requirements for 20% to 15% may considerably understate the change in policy by the CXC.  Therefore the CUC recommends that the CXC aggressively advertise the change in policy, pointing out not only the quantitative decrease by the change in definition of what is time-constrained.  Moreover, the CXC should carefully consider instructions to the TAC in this regard, and how to deal with the problem of a VLP program that require orients, the acceptance of which could dramatically change the fraction of orient time, but which is also likely to be among the highest priorities for Chandra science.   There are various questions to be considered. These include whether to allow the TAC to accept proposals without the requested  constraints.
• The CXC recounted several examples of proposals that did not appear to have special requirements, but had such requirements buried in the text of the proposal.  The CUC feels that the CXC should be fairly strict in enforcing the rules that all special requirements be properly represented in the section of the proposal provided for this purpose. Partly this is an issue of fairness to other proposers and partly it is an issue of fairness to the CXC. The CUC also suggests that the CXC solicit the TAC panels to be more vigilant to inspect proposals to see if they have buried requirements, and to ask the CXC on how to handle such proposals.  Large and very large proposals constitute a special case.  These are presumably the highest priority programs for Chandra to conduct.  If they are approved there will be strong pressure to carry them out in a way to maximize their science return.  The CUC understands why it is impossible for CXC staff to conduct technical reviews of all proposals.  However, we believe the CXC should investigate a process for carrying out technical reviews of Large and Very Large proposals that emerge from the panels with marks that are sufficiently high that they are likely to be seriously considered by the merging panel. 
• The CUC appreciated the progress that has been made in addressing some of its previous concerns regarding instructions to users on the effects of new builds and tools to handle pulse pile-up issues within CIAO, and agrees that better tools for addressing background subtraction issues will be useful to the community.   And it understands that with Jonathan McDowell's just coming on board, it was likely difficult for the CUC to have been presented with longer-term plans for development of CIAO and tools for observers.  But at the next meeting, we would appreciate a more general discussion of what the CXC regards as the main deficiencies of CIAO and CALDB, and of its evaluations of priorities of various options to address the perceived deficiencies.  Areas that might merit investigation in this regard include spectroscopy of faint sources and PSFs in the PSF library.  (As an aside, the CUC does not regard PSF solutions that avoid use of the CALDB as a reasonable option for most observers.) Furthermore, the CUC understands that formal meetings of the CUC may not be well synchronized with the need to make decisions about development priorities; if that turns out to be case in practice we urge the CXC to ask for its advice more informally, via email or by telecon.
• With regard to overall software priorities, the CUC continues to agree that CIAO and activities that affect today's reduction of science data should have priority over longer-term issues, e.g. Level III and VO readiness issues, even as we understand progress needs to be made in those areas.   Given this, we understand that it is difficult to predict precise schedules for the development of source catalogs.  However, once again the CUC would like to be consulted regarding the characteristics of the catalogue and on other Level III issues as they arise.  With regard to the source catalog, we believe there ought to be a set of requirements, and these should be discussed at the appropriate time, where we would define the appropriate time to be following initial prototyping but well-before a commitment to production processing has been made.  Furthermore, we note that there is quite a bit of experience in the community with regard to creating catalogs of objects from Chandra data.  We urge the CXC to consult the developers of that S/W, such as Andy Ptak from JHU who has developed "xassist" and Patrick Boos at Penn State who has developed "ACIS Extract", to learn about their experience before embarking on an independent effort.  We believe this will result in the best catalog in the end.
• The CXC is embarked upon an impressive list of calibration efforts, and we were pleased, once again, to see that a calibration effort involves well-defined projects with well-defined end dates.  Some of these activities are intended to maintain the current state of calibration of the observatory and others are intended to improve the calibration, as outlined in the overall plan.  Unfortunately, the style of the presentation on calibration made it difficult to assess how rapidly the CXC is proceeding towards reaching its calibration goals and with what priority, and therefore it was difficult to offer advice on whether the priorities are correct.  The CXC needs to remember that CUC members are not party to many of the day-to-day meetings of the CXC, that most of us did not attend the Calibration Workshop, that many of the members were new to the CUC, and tailor the presentation to the expertise of the CUC. More specifically, in future presentations, the CXC needs to concentrate on showing more quantitatively how the calibration effort is (or is not) actually producing progress towards improving the calibration accuracy requirements and goals expressed on http://cxc.harvard.edu/cal/ and how the completion of future projects will aide in improving the calibration. 
  
• The X-ray community is fortunate to have two imaging X-ray missions returning excellent scientific data at the same time.  The two observatories have, in many respects complementary capabilities.  And indeed, proposals submitted to the Chandra TAC (and presumably vice versa) must often demonstrate that they are best suited for observations with Chandra.  Therefore, the CUC is disappointed with the progress made to cross-calibrate the two observatories, or more specifically that there do not appear to be concrete results associated with the cross-calibration effort.  We are aware of several meetings that CXC plans to attend to discuss calibration issues.  One possibility is that the CXC establish in coordination with the XMM data center a web site that compare observations of a selected group of cross-calibration targets with XMM.  The web site would be updated after every major update of the calibration files.  Alternative plans are also acceptable, but the CUC would like to see a real improvement in this area in the near term.  In addition, the CUC urges the CXC to consider a strategy for spectral cross-calibration with the Astro-E2 XRS upon a successful launch of this observatory currently scheduled for July 2005.
• The CUC has followed the various developments associated with the bake-out. It is evident that the CXC and the entire Project has approached the decision of whether to carry out a bake-out or not with the professionalism that a decision of this moment deserves.  Like many others, we see significant benefits in terms of low energy sensitivity, especially for extended sources.  We also see the risks.  Like others, we do not think that at the present time, it is obvious the correct course to take, and therefore all we can do is offer our continued interest..  The CTI tests appear to be the only major activity required to inform a recommendation by the CXC director on how to proceed.  We look forward to the completion of those tests, and hope these will lead to a definitive plan to proceed, or to defer consideration of a bakeout for several years. We are happy to participate in the process of formulating a recommendation whenever the Project and CXC Director feel the time is ripe. 
  
• The CUC recognizes that the CXC is bound by NASA HQ guidelines regarding policies for individual grant awards for EP&O activities.  However, it still wishes to go on record that in the absence of a simplification of the requirements and/or an increase in the maximum grant award to a number greater than $15K, it is very unlikely that this portion of the EP&O program will produce significant response by investigators. Without such response, it is hard to see how such grants can be expected to produce much impact in the communities these types of grants are intended to serve.
• Several years ago when VLPs were added to the Chandra program, there was some concern that after a few years, the quality of the VLP programs might decline. Therefore the possibility of reviewing the process after a few years was suggested.  The CUC believes that it would be appropriate to conduct an informal review of VLP and large programs following the selection of proposals this summer, and requests a presentation on this topic at the next CUC meeting in the fall of 2005.  The TAC panel chairs are probably one the best groups to assess whether the quality of VLPs remain high, and therefore we request that the CXC, as part of its overall assessment of the TAC, ask the panel chairs to comment on the relative quality of large, very large, and regular GO program.
• The CUC has concerns concerning the length of Chandra grants, and some concerns what may be a misperception in the community concerning grant extensions.  The basic Chandra grant is one year, and can easily be extended for one year.  The basic HST grant is for two years, and can be extended easily for one year.  The basic Spitzer grant is 3 years.  The CUC does not believe that Chandra data is intrinsically easier to analyze than HST or Spitzer data, and therefore suggests that the CXC review its policies and attempt to bring them more into line with NASA's other great observatories in time for awarding of Cycle 7 grants. The CXC indicated that they were fairly flexible regarding grant extensions in the course of the CUC meeting, but the user community's impression, at least as reflected in the minds of most CUC members, is that the CXC is more strict than other similar missions.   Consequently, we request that the CXC take prompt action to publicize what is required to justify an extension under the existing policies, and in a way that indicates that the CXC will be receptive to extension requests when justified.
• Finally, and especially in light of the fact that we have elected to move the next CUC meeting to fall 2004, some 9 months from now, the CUC requests that the CXC develop a short written response to the various suggestions contained in this report. We suggest that this response be circulated to the committee and ultimately placed on the web site containing the CUC report.  We understand even if the CXC fully agrees with a recomendation in the report, it may need time to develop a specific implementation plan, and hence the CXC's response may in some cases be a statement of intent or a description of a few first steps without a guarantee that the recommendation can be accommodated.  Nevertheless, we believe that a report would help both the CUC and  the community to understand the direction the CXC is attempting to take with regard to CUC recommendations.