Report of the 73rd Meeting of the Canada-France-Hawaii Telescope Scientific Advisory Council

Toronto, Canada, 15-16 May 2008

NOTE: The report presented below is the public version of the SAC report as amended by the CFHT Board of Directors.

The recommendations contained in this public version have been endorsed by the Board.

Recommendation 1 ESPaDOnS Users' Feedback
Recommendation 2 ESPaDOnS QSO-PI communication
Recommendation 3 CFHT-TAC interaction
Recommendation 4 CFHT-PI communication on time allocation
Recommendation 5 C Programs
Recommendation 6 Phase II no-tolerance policy
Recommendation 7 CFHTLS 2008B allocation
Recommendation 8 CFHTLS completion
Recommendation 9 CFHT-LP interaction on RA
Recommendation 10 LP review

SAC members Pierre-Alain Duc (chair), Laura Ferrarese (vice-chair), Chris Ftaclas, Brett Gladman, Jean-François Gonzalez, Robert Jedicke and Dae-Sik Moon, attended the meeting. Cecile Gry, Henk Hoekstra and Denis Mourard were unable to attend. Jeremy Lim from the Academia Sinica Institute of Astronomy and Astrophysics in Taiwan and the chairs of the Canadian and French Time allocation committees, Gregg Wade and Coralie Neiner, also attended the meeting as invitees. The CFHT director Christian Veillet, Derrick Salmon, Jean-Charles Cuillandre (for CFHTLS) and Pierre Martin participated in presentations and discussions.

1. Technical activities report

Derrick Salmon reported on the technical activities for the period November 2007 - May 2008.

SAC agrees with the following development priorities proposed by CFHT:

1 Operational Priorities

2 New Instrument Development

3 Observatory Automation


MegaCam operation

SAC was informed that no major new failures of MegaCam occurred during the last semester; however, minor problems had occurred and were investigated. The MegaCam team investigated intermittent communication problems experienced between the controller and the detector host computer via the S-LINK. Due to a marginal drive signal, clock resets occasionally fail after an exposure causing problems for the shutter timing.

SAC is pleased to see that the overall operation of MegaCam has been smooth. The MegaCam team should address the issues of the S-LINK communication board and shutter timing board as soon as possible.

SAC was informed that a major maintenance period for MegaCam would be carried out in June and July with the participation of CEA. Among the planned tasks are the modification of the shutter timing circuit to fix the above-mentioned problem, the replacement of the side rails of the jukebox, the cleaning of the pulse tube, the installation of the new rotating valve delivered from CEA (and successfully tested by CFHT), and possibly the installation of SNDICE.

The SNDICE (SuperNova Direct Illumination Calibration Experiment) project is aimed at improving the photometric accuracy of MegaPrime data using a LED source already installed on the dome and a CLAP (Cooled Large Area Photodiodes) detector to be installed in MegaCam during the major maintenance period. This is a joint effort between the SNDICE team based at the LPNHE laboratory (IN2P3, France, currently involved in the calibration of the SNLS) and the CFHT. Christian Veillet informed SAC that a review of the experiment is planned in the near future to make the final decision on the implementation of SNDICE.

SAC has not yet been given the opportunity to evaluate the cost and benefits of SNDICE in detail. So far, SAC was not given convincing arguments that future programs with MegaCam would scientifically benefit from the implementation of SNDICE. Furthermore, SAC is concerned about the cost of SNDICE and wishes to know whether its installation and maintenance may cause a risk to the instrument.

SAC wishes be informed about the outcome of the SNDICE review before any implementation of SNDICE on the MegaCam camera.

WirCam operation

SAC was informed that WIRCAM did not have any major problem over the last year. After taking care of the negative and positive crosstalk that affected the deep stacks, the WIRCAM team successfully suppressed the edge crosstalk. They were caused by a grounding problem in the amplifiers of the video channels. The first run with the improved control electronics took place mid April.

SAC is very pleased to see that the problem of the electronic crosstalk has been solved, and appreciates the efforts of the CFHT staff on this matter. CFHT should however still check deep stacks to ensure that all the crosstalk has indeed disappeared.

ESPaDOns operation

SAC was informed about several minor improvements to the instrument: reduction of oscillations, a better red sensitivity for the guider, implementation of a seeing estimator and of a new automatic procedure for the focus, setting up of a new device and automation procedures to help and secure the instrument setup. Finally, CFHT is about to purchase a new triplet lens in the polarimeter. The expectation is that it should help to reduce the polarization cross talk, a long-standing recommendation of SAC.

SAC was presented a report to replace the EEV1 detector that suffers from a few cosmetics problems and long readout times due to the failure of one amplifier. CFHT plans to purchase a new detector (E2V) which would offer a better quantum efficiency in the blue and red, better cosmetics (if a grade 1 device is ordered), less fringing in the red, no smearing and shorter readout times using two amplifiers. According to CFHT, the gain will provide about 30 more minutes per night on the sky. The cost of the new detector is $100K while no additional cost for the readout electronics is expected.

SAC recognizes the potential of replacing the EEV1 detector to improve the efficiency of the instrument. It noted however that there is a risk associated with the use of dual amplification.. SAC was informed that tests are being carried out on the sister instrument at Pic du Midi, Narval, and understands that the final decision to purchase the detector will depend on the results of these tests. The purchase order may occur before the next SAC meeting. SAC also discussed issues of instrument ownership and whether it will be necessary to have the consent of the instrument PI to make major changes. SAC wishes to be informed of the results of the tests as soon as they are available.

Observatory automation

SAC was informed of the high priority given to this activity that is now managed as a formal project with newly allocated technical resources. The expectation is to implement routine Waimea based remote observations by 2010.

SAC sees this as an important project for the observatory and supports its continuation, although the planned completion of automation in 2010 may be somewhat optimistic. The SAC wishes to see the cost estimation report in the next semester, together with the estimation of potential risks and a risk management plan.


SAC was informed that the DIMM was installed in the dome shutter well but that a few major tasks should still be completed before it is fully operational. The once unreliable motors of the DIMM telescope were replaced. Inexpensive CCDs were successfully tested and purchased. Dedicated software and web applications were written.

SAC sees the construction of DIMM and monitoring of the seeing as an important long-standing issue for CFHT and supports the continuation of the project. In the long term, the DIMM information should be implemented in the file headers of the science exposures.

Mauna Kea All Sky Mid-IR camera

SAC was presented a report written by Glenn Morrison (CFHT/IfA) on the plan for a Mauna Kea All Sky Mid-IR camera. It will allow a measurement the mid-IR emission from all clouds, including thin cirrus, and map their distribution on the sky even on moonless nights. A commercially available instrument is being considered and a demonstration unit will be installed this summer near the UH 88-inch. If the tests are successful and if the project is accepted by the Mauna Kea site monitoring working group, the instrument will be purchased. Various Mauna Kea corporations will share the cost. The contribution of CFHT should be about $10K.

SAC sees this as a worthwhile project and supports CFHT to lead the project for the Mauna Kea observatories.

2. AO activities at CFHT and their future

SAC was asked by the Board to evaluate the activities on adaptive optics at CFHT and to comment on their possible link with the future of AO with this telescope. Christian Veillet presented to SAC a report summarizing the past and on-going efforts on AO at CFHT and giving some reflections for the future. He emphasized that the AO system, PUEO, was particularly productive until 2001 and acted as a reference system for other telescopes. Since then PUEO is no longer much used. It has nevertheless still been scheduled on the telescope each semester to fulfill the need of a few scientific programs but mainly to carry out engineering tests (12 nights between October 2007 and April 2008) and thus maintain AO expertise at CFHT.


FlyEyes is an in-house development project to test the use of CCDs instead of the Avalanche Photodiodes (APDs) that are currently used with PUEO. Since last October the wavefront sensor for FlyEyes has been tested three times on the sky. The Strehl performance, limiting magnitude and loop-sampling rate were estimated.


The feasibility of the VASAO (Visible All-Sky Adaptive optics) system was studied at CFHT in parallel with the development in France of the ELP-OA project (polychromatic laser guide star for the adaptive optics). The concept using a two-color laser guide star was found to be unfeasible in visible light in the 2012/13 period when new instruments are being considered.

Nevertheless, in November 2007 CFHT carried out a two-color experiment with natural stars to test the long-term potential of a polychromatic laser guide star to make tip-tilt corrections. In principle, the atmospheric tip-tilt can be inferred from the centroid displacements of the light from a single source filtered in two bands but no correlation between the tip-tilt and the centroids was measured. The experiment might have failed due to poor signal-to-noise and an inadequate field size.

The long-term plan for creating this polychromatic laser guide star is to use a mode-less laser (MLL) and the radiative cascade at 589 and 330 nm following the excitation of the mesospheric sodium by a single photon. CFHT has started a collaboration with the Laboratoire de Spectroscopie Physique (LSP, France) to carry out an experiment with an MML purchased on CNRS and CFHT funds ($30k for the CFHT contribution). The evaluation of the experiment will require the use of the other Laser Guide Stars available on the mountain (Subaru, Keck and Gemini) and the use of a lidar telescope.

A key ingredient for the success of the VASAO/ELP-OA concept is the sensing of the vibrations of the telescope that induce an extra component in the tip-tilt correction that cannot be monitored with the laser guide star alone.. The installation of a seismometer on the telescope is thus required. Vibration measurements of the declination axis with a seismometer mounted on the central pier of the CFHT were reported at the 72nd meeting of the SAC.

Future CFHT instruments

In his report, Christian Veillet noted that two of the future instruments proposed for feasibility studies, `IMAKA and FIRST, could benefit from the Adaptive Optics developments at CFHT. The one deg imager `IMAKA does not initially plan to use laser guide stars (LGS) but instead a constellation of at least five natural stars associated with AO systems. In the longer term the use of LGS could be considered as an option. The high-resolution and high dynamic range imager FIRST would be even more efficient if coupled with an IR-optimized AO bonnet such as PUEO. The gigantic interferometer `OHANA will eventually require an AO-correction at each telescope.

[Note added]

SAC opened a discussion on the future of AO at CFHT and on the AO-related developments currently under way at the observatory. This discussion will continue at the next SAC meeting together with the review of new instrumentation proposals for 2013 and beyond.

3. Telescope use and scheduling

Queue ESPaDOnS

SAC heard the CFHT report on the new Queue mode for ESPaDOnS and received feedback from a few PIs. Despite some initial problems with the control software, data access to PIs and unfriendly weather, the QSO went well (technically) during the first run. The implementation of the QSO required the development of a new pipeline called Upena that uses several routines from the data reduction software Libre-ESpRIT.

SAC is grateful to the QSO team for the efforts made in this achievement. It is however too early to determine the overall efficiency of the queue mode for this instrument which, compared to MegaCam and WIRCam, faces many more observing constraints, in particular severe time-critical conditions.

Recommendation 1

SAC recommends that PIs of all ESPaDOnS proposals be contacted at the end of semester 2008A to get their feedback on the new QSO mode, including the phase 2 process and interaction with the QSO team.

Back to top

The report given to SAC lists several suggestions to improve the Phase 2 Tool (such as better finding charts or more flexibility to enter the time constraints), run planning and automatic data reduction.

SAC noted that the success of the QSO mode of ESPaDOnS would largely rely on the PIs being able to monitor the observations and thus be informed about them. Indeed, for monitoring programs, an incorrect instrument setting might imply the loss of a whole run even if the problem is identified early.

Recommendation 2

SAC recommends that CFHT alert the PI when their program is started and provide access to all required information to monitor data quality. When properly justified (e.g. by the need for coordination with observations at another telescope) the PI should be informed in advance when his/her observations are likely to be scheduled.

Back to top

MegaCam/WIRCam QSO report

Pierre Martin presented to SAC the usual QSO report for WIRCam and MegaCam. During the fall 2007B run the weather was worse than usual. More than 35% of the MegaCam time was lost due to bad weather and an additional 5% was lost due to technical problems. As a result, the completion level for B programs was only 60%. The improved focus model contributed to a global high validation rate. The fraction of WIRCam time lost due to bad weather was about 30%. Almost no time was lost due to technical problems.

As usual, the balance between the agencies has been strictly respected. The Deep part of the CFHTLS, which was affected by the break of the i' filter, did not get its full share of time.

SAC notes that the QSO statistics provided by the CFHT to the SAC could be of interest to the chairs of the national TACs. Even more useful would be the success rate of all proposals for which an allocation of time had been recommended. Such information is already available on the QSO web site but is not easy to retrieve. Compiling the data in a compact table would be more practical.

Recommendation 3

SAC recommends that CFHT inform the chairs of the national TACs about the status of all A, B and C programs after the end of each semester including an indication of their level of completeness.

Back to top

Status of C programs in QSO mode

Some misunderstanding has arisen in the community about the status of the so-called C ranked proposals, i.e. proposals ranked below the cut-off line that may still be considered for a phase 2 proposal. SAC received a letter from a PI asking for clarification. The director of science briefly explained to the SAC that C-ranked proposals are considered as "bonus" proposals. Depending on their probability of being executed some of them are allowed by CFHT to pass through a phase-2 step and others not. Contrary to common belief the ranking of the proposal below the cutoff line is not the main driver for the decision.

SAC discussed new procedures to properly and rapidly inform PIs about the status of the proposal given the additional fact that the cutoff line between "accepted" proposals (ranked A and B) and below (ranked C) is usually not known at the time the national TACs meet.

Recommendation 4

SAC recommends that after their meeting the TACs only provide the PIs the relative ranking of their proposal and the referees' comments. If however the TACs decide for any reason not to allocate any time to a proposal, the PI may be informed immediately. CFHT will then be responsible for informing all PIs in due time about the final status of their proposal: A, B (accepted), C (only accepted as a backup and going through a phase-2) or rejected (i.e. not allowed a phase-2).

Back to top

SAC heard from Pierre Martin that the success rate of the C-ranked proposals that went through a phase 2 had been quite high during the last semesters. It is actually sometimes as high as for the B-ranked proposals even though the number of C-hours is much less.

SAC understands the need for having backup C programs in the Queue mode context. However, it felt that their impact had not yet been fully addressed. Their importance has increased with CFHT only operating in service mode. The level of the cutoff line (counted in nights) between the B and C ranked proposals is sensitive to the arbitrary conversion factor between hours and nights. The conservative decision of CFHT to reduce the conversion factor to 5.5 hours per night for MegaCam and 7.5 hours per night for ESPaDOnS will artificially increase the number of potential C-programs in the coming semesters. Furthermore, the regulation of the balance of time between the agencies is made on a per instrument basis. With that rule, C-ranked proposals from a given agency requiring an instrument for which no time was granted by the same agency in the A+B categories cannot be executed irrespective of their ranking or their ability to fill RA holes! Adding more flexibility in the process will help ensure that the best possible science is carried out and that the time allocation for C-ranked proposals is less random.

Recommendation 5

SAC recommends CFHT to consider various solutions to control the balance of time between agencies that would address the issue with the C-ranked proposals.

Back to top

Suggestions will be discussed during the next SAC meeting in November.

Phase II templates and quality control

SAC discussed ways to improve the quality control of the data obtained in QSO model based on the experience acquired with the CFHTLS.

In his CFHTLS SG report, Jean-Charles Cuillandre explained the reason why some CFHTLS images validated by CFHT had been rejected by Terapix. To improve throughput efficiency CFHT applies a tolerance to the constraints on the data quality (e.g. on the seeing) off which the Terapix team was not aware.

Although this rule is explicitly mentioned in the phase-2 document, the SAC is concerned that some PIs might not be aware of the policy when filling their phase-2 proposals. Since some scientific programs cannot accept any tolerance, SAC recommends that the Phase-2 template be modified as follows:

Recommendation 6

SAC recommends that the Phase II template be modified with the addition of check boxes next to the image quality, sky brightness and cloud coverage that offer the option to refuse any tolerance limits.

Back to top

The box would be labeled by the heading "please click if you do not accept a 15% tolerance in these parameters. Note that selecting the field decreases the chances of the data being acquired." A link to the relevant web page could be provided.

4. Current and future Large Programs

CFHTS progress and time allocation

On behalf of the CFHTLS Steering Group (SG) Jean-Charles Cuillandre gave the SAC a detailed presentation including a progress report and justification for future time allocation.

Following SAC recommendations, the 2008A CFHTLS observations were carried out with priority over PI programs. During the first three 2008A MegaPrime runs, CFHTLS observations comprised approximately 50% of all validated MegaPrime hours (although the fraction was significantly higher in the first run which was badly impacted by bad weather). This is typical of previous semesters, i.e. PI programs were not severely affected by the decision to grant priority to the CFHTLS.

There are four main issues for which the steering group requests additional time in 2008B and 2009A: 1) completion of the Wide Survey (92.2 hours requested for 2008B); 2) patching of the Wide field (essentially re-observing fields validated by the CFHT QSO but rejected by Terapix,18 hours requested for 2008B); 3) photometric calibration of the Wide (11.2 hours requested for 2008B); and 4) photometric calibration of supernovae observed with the new i' filter (40 hours requested in 2008B and 10 hours requested in 2009A). The total time request for 2008B is 154 hours, and 10 hours for 2009A.

During its Nov 2007 meeting, SAC recommended allocation of 50 hours for completion of the Wide survey and 4 hours for the photometric calibration of the Wide, all to be carried out in the 2008B semester. In its May 2007 meeting, SAC recommended that these observations be carried out with priority over PI programs. Furthermore, SAC discussed whether additional time allocation in 2008B would be considered following an evaluation of a detailed justification from the steering group.

SAC carefully considered three separate issues in evaluating the additional time request made by the CFHTLS steering group: 1) the interplay between CFHTLS, PI and Large Programs in 2008B; 2) whether the 100 hours requested in addition to the 54 already allocated by SAC in 2008B are justified; and 3) the request to extend the program to the 2009A semester to complete the i'-calibration of supernovae discovered and followed-up during the 2008A semester.

SAC feels that the explanation provided by the CFHTLS steering group justifies granting the additional 100 hours requested beyond the 54 hours previously allocated. However, SAC feels that allowing these 100 hours to take priority would have too negative of an effect on the PI and/or Large Programs.

Recommendation 7:

SAC recommends that an additional 100 hours be allocated in 2008B for the CFHTLS and scheduled with the same priority as the PI and Large programs.

Back to top

Recommendation 8:

Any additional CFHTLS related time request beyond 2008B should be made through regular PI programs.

Back to top

This applies to the 10 hours requested by the SNLS in 2009A for the photometric calibration of the Wide and to any observations that are not completed in 2008B.

Time allocation and implementation of the Future Large Programs

A total of twelve proposals for Large Programs (LPs) were submitted. Following the ranking proposed by the Large Program Time Allocation Committee (LPTAC) and taking into consideration the feedback given by the national Time Allocation Committees (TACs) and information on the technical feasibility provided by CFHT, SAC recommended to the Executive Director of CFHT to accept the following programs with the full amount of requested time:

The Pan-Andromeda Archaeological Survey (PandAS)
PI: McConnachie – Instrument: MegaPrime

The Next Generation Virgo Cluster Survey (NGVS)
PI: Ferrarese - Instrument: MegaPrime

Magnetic Protostars and Planets (MaPP)
PI: Donati - Instrument: ESPaDOnS

Magnetism in Massive Stars (MiMeS)
PI: Wade - Instrument: ESPaDOnS

For all four programs, time is shared between the French and Canadian communities. Since no PI objected to the time sharing all the raw data from the new LPs will be available to the entire French and Canadian communities with no proprietary period. Taiwan provided additional time for one of the proposals. The Hawaiian TAC decided not to join the Large Program scheme but will consider participation on a per semester basis alongside regular PI programs.

SAC is disappointed with the Hawaiian TAC decision since it contradicts the official wish of the Hawaiian agency to take part in the Large Programs. Furthermore, it adds uncertainty to the total amount of time allocated to the LPs that include Hawaiian partners.

Following deliberations by the CFHT Board, LPs and PI programs are treated as separate agencies and must compete for observing time.

Because of the severe RA pressure produced by some of the LPs, SAC makes the following recommendation:

Recommendation 9:

Each semester, before the call for proposal for PI programs, the PIs of LPs should provide to the CFHT their expected RA distribution in exposure hours and the parameters requested for the image quality. This will be included in the call for proposals to inform the PIs where there is heavy pressure.

SAC discussed the procedures to monitor the progress of the LPs. It proposes the following steps:

Back to top

Recommendation 10:

SAC recommends that the following procedure for review of the LPs be implemented: Every semester, approx. two weeks before each national TAC meeting, LP teams shall submit to the chair of the national TACs and to SAC a short outline summarizing the status of data acquisition, processing and analysis. During the course of the LPs, with at least one semester's notice, SAC will ask for a detailed report describing the science results obtained up to that point, any problems encountered, and a proposed strategy for the remainder of the survey. As a guideline, the detailed report should be expected at the halfway mark.

Back to top

5. Future of CFHT

CFHT's Golden Age plan and the future of CFHT

Christian Veillet presented to SAC a quick update on the CFHT's Golden Age plan. Initially made for the period 2005-2010, it will have to be extended to 2012. This will be addressed during the next SAC meeting.

SAC was informed that the staff has decreased naturally (with departures and retirements) and the number is already at the level foreseen for 2010.

The Executive Director informed SAC that the Board supported his efforts in seeking new CFHT partners. Up to 30 CFHT nights per year could be sold or exchanged. SAC was informed about a plan for an exchange of nights between CFHT and the AAT starting in 2009B. Five nights per semester are being considered.

New instrument feasibility studies

Following the recommendation of SAC and endorsement of the Board, four teams received CFHT funding to finance their instrument feasibility study. The Executive Director postponed the deadline for submission of feasibility studies to October 1st, 2008. The new deadline has been advertised on the CFHT web site. SAC will discuss these studies during its November meeting and possibly recommend one or several instruments for a phase A study.

SAC was informed by the Executive Director that CFHT will itself evaluate the technical feasibility while SAC should mainly discuss the scientific interest.

Given the expertise of several SAC members on instrumentation, SAC felt that it should also be associated with the evaluation of the technical feasibility of the proposed instruments. Three members of the SAC, Dae Sik Moon, Denis Mourard and Christ Ftaclas, were nominated to discuss this matter with the CFHT, external experts and the teams proposing the new instruments and will report to the SAC the technical aspects of the proposed instruments. Given the short delay between the deadline for the submission of the feasibility studies and the SAC meeting, SAC asked the Executive Director to inform the communities about the proposed instruments well before the deadline. The feedback from the communities will be an important element in their evaluation.

Back to top