Chandra Users' Committee Recommendations to Chandra Director Chandra Users' Committee Meeting, January 20-21, 2000 1. GTO TOOs during Cycle 2 The committee was very interested in the proposal by George Ricker to use a portion of his GTO time to do follow-up TOO observations of GRBs detected with HETE II. Although this violates the current observing policy, we think the CXC should consider an exception in this case because 1) the Chandra data in question would be made public immediately, 2) the HETE II data will be public as well, and 3) HETE II is a time-critical resource. In fairness, the CXC should consider other Cycle 2 GTO TOO proposals, but we strongly recommend that any accepted proposals should make the Chandra data public immediately. The total number of GTO TOO targets allowed should be limited to a fraction of total TOO targets which is equal to the proportion of Cycle 2 time which goes to GTOs. We believe that this means that, at most, 3 GTO TOOs would be allowed. 2. Funding for Guest Observers Because of the adverse effect of the Chandra launch delays and the early uncertainty in the observing schedule on the hiring of graduate students and post-docs by Chandra observers, we recommend that the CXC immediately notify PIs that grants will routinely be granted a 1-year no-cost extension following a request and a very brief interim report. 3. Science Conflicts during Proposal Review The committee recommended (by a 4 for, 2 against, 2 abstaining vote) that significant science conflicts be considered during the proposal review process. Such conflicts occur when a panelist is a PI or Co-I on a proposal with the same major science objective(s). Such conflicts should be identified prior to the discussion by the panelists and/or the panel chair. Any affected panelist should be asked to leave the room briefly near the end of the discussion of the affected proposal, while the panel discusses the possible effect of this conflict. The affected panelist will be allowed to vote. 4. Support of Chandra Observers We applaud the CXC decision to assign scientific staff to individual GOs to help introduce them to the analysis of Chandra data. However, we are concerned that this will be a significant drain on staff time. We hope that enhancing the on-line documentation will take care of many of the routine problems that are coming up now. As generic problems and solutions are encountered, they should be placed in the "Things to Watch Out For" web page and/or a FAQ web page, as well as incorporated in the appropriate software/instrument guides. We realize that some of our recommendations for improved documentation and information for observers will require a significant time investment by CXC staff in the short term. We also understand that people at the CXC have been extremely busy with the problems associated with getting a major observatory working. However, it is our belief and hope that improvements in the public information available to users will reduce the number of queries which require individual attention from the CXC staff. Thus, we hope the effect will be to reduce the burden on the staff in the long run. One concern which came up in the discussions of improvements to the webpages, to the CIAO documentation, and in the handling of questions by users was that several of the CXC people who are responsible for these functions have not had the time to actually try and analyze Chandra data or use the software themselves. Of course, a primary reason why any observatory like Chandra has scientists (as opposed to managers or software engineers) on its staff is for the practical scientific experience they bring to their service roles. This is also a prime motivation behind the contractual science research time for the scientific staff. Although we realize the time pressures on the scientific staff are very great at present, it is essential that the people who work on documentation and user services have experience working with Chandra data and software. We encourage the CXC Director to try and protect a portion of their time for scientific research with Chandra data, perhaps in collaboration with other scientists at CXC or elsewhere. The User Support scientists should use their experience with the data and software to improve the documentation and information available to all users, and this needs to happen very soon. 5. Webpage Improvements The committee is happy that the CXC is responding to committee suggestions for improvements to the CXC web pages, including the creation of a "Things to Watch Out For" page. The committee recommends that one person should be responsible for organizing the data analysis documentation. It is essential that this person actually analyze Chandra data, using existing public documentation and software to see where problems lie. The best way for this to occur is if people responsible for documentation have science time to analyze Chandra data. 6. CIAO Software The committee was very pleased with CIAO 1.1. We endorse the high priority placed on producing patches which will a) allow FEFs to be selected automatically, and b) facilitate the analysis of extended source spectra with ACIS. Observers need to be told immediately about the circumstances when many FEF/RMFs are required and when they are not, and when spectra should extracted in PI space vs. PHA space. Because of the limited staff time available and a number of more pressing tasks, the committee recommends that the CXC not pursue the porting of CIAO to IRAF or worry about PROS compatibility. 7. Software Documentation Issues Not surprisingly, the software documentation is less up to date than the software. There is an urgent need for "walk throughs" (example threads of software commands) for the most basic initial data analysis tasks. The CUC very strongly recommends that documentation efforts concentrate, in the short term, on producing such walk throughs for the extraction and initial analysis of basic scientific products for each common instrumental configuration. We estimate there may be roughly a dozen such walk throughs required. Specifically: a) currently, there is almost no public documentation and no walk throughs for analyzing grating data. b) the walk throughs for analyzing imaging data are confusing and incomplete c) the walk throughs for analyzing spectral data, particularly from extended sources, are confusing, incomplete, and probably incorrect. d) given the gain file and other inconsistencies which have occurred in the pipeline processing, a walk through is need for using acis_process_events to re-extract level 2 events and correct the PIs. The need for these walk throughs underscores the the importance of the CXC User Support Group (or others charged with generating these walk throughs) being involved in the analysis of Chandra data. Ideally, the walk throughs would be generated during such scientific research. In the short term, anyone at CXC who is or has analyzed Chandra data should be prevailed upon to generate walk throughs for the tasks he/she has done. On a somewhat longer term, better documentation on the Chandra data files, pipeline processing, and calibration are needed. 8. Observation Scheduling The CUC was very happy with the improvements in scheduling observation and providing information to observers about the status of their observations. One area which remains confusing is that nature of the "pool" targets at the end of the Long Term Schedule. The presentation at this CXC meeting was the first time that any of the members of the CUC had heard of this, and the first time we understood the meaning of the long pseudo-weeks at the end of the observing schedule. Everyone simply assumed that these targets would not be observed until Fall 2000, and thus most of us had not given any immediate thought to modifications to the observing plan for these targets. We assume that most other GOs are equally confused. Thus, we recommend that the CXC immediately send a short E-mail to all PIs pointing out the meaning of this "pool", and that the information about this (which is part of several paragraphs at the start of the Long Term Schedule) be highlighted in some way. 9. Cas-A HRC-I Observation The CUC was convinced that the CXC Director acted reasonably and in the interest of maximizing the scientific output of Chandra in allowing Steve Murray to use part of his Cycle 1 GTO time to do an HRC-I observation of Cas-A. In the future, it would probably be useful if the CUC were informed of such potentially controversial decisions, so that we can respond usefully to complaints from users. (Of course, it may be difficult to know what is likely to raise controversy.) 10. Prior Human Technical Reviews in AO-2 Delays in the release of the NRA for AO-2 mean that the time from the release of the NRA until the final selection of targets is uncomfortably short. In order to save time, the Committee recommends (by vote of 2 for reviews, 5 against, 4 abstaining) that prior human technical reviews NOT be done by the CXC for AO-2. The CUC recommends that technical reviews be available upon request by either the reviewers or the panel chairs prior to the panel meetings (during the time when reviewers are reading the proposals) and during the panel meetings themselves. We also recommend that the CXC provide the reviewers with information on typical technical problems and up to date information on the instruments at the same time that the proposals are sent to the reviewers. During the meetings of the review panels, the panels should be instructed that they should not reject a proposal on purely technical grounds without consulting with a technical expert from the CXC. Based on the experience in AO-1 and AO-2, the CUC would like to reconsider this issue after the AO-2 review and prior to the AO-3 review.