Report of the 75th meeting
of the Canada-France-Hawaii Telescope
Scientific Advisory Council

18, 19 May 2009 - Paris, France

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

After deliberation, the CFHT Board of Directors has approved SAC’s recommendations No. 7 – Impact by new LPs on current ones, and No. 8 – Call for new constrained LPs.

The remaining recommendations will be further discussed at the Board's next regular meeting, December 2009.


Recommendation 1: Solving the crosstalks issue of ESPaDOnS
Recommendation 2: Informing the users on the ESPaDOnS Crosstalks
Recommendation 3: Change of detector of ESPaDOnS
Recommendation 4: Data reduction software of ESPaDOnS
Recommendation 5: Update on science with SPIRou
Recommendation 6: Field of view of `IMAKA
Recommendation 7: Impact by new LPs on current ones
Recommendation 8: Call for new constrained LPs


SAC members Mark Chun, Pierre-Alain Duc (chair), Laura Ferrarese, Brett Gladman (vice-chair), Cecile Gry, Robert Jedicke, Denis Mourard, Coralie Neiner, Gregg Wade and Jon Willis attended the meeting. Shiang-Yu Wang (from the Academia Sinica Institute of Astronomy and Astrophysics in Taiwan) and Bruno Castilho (from Laboratorio National de Astrofisica in Brazil) attended the meeting as invitees. The CFHT director Christian Veillet, Derrick Salmon, Daniel Devost and Jean-Charles Cuillandre (for CFHTLS) participated in presentations and discussions. Jean-Luc Beuzit (CNRS/INSU representative) attended part of the meeting.

1. Technical activities

Derrick Salmon reported on the technical activities for the period December 2008 - April 2009.

SAC agrees with the following development priorities proposed by CFHT:

  1. Operational Priorities
  2. New Instrument Development
  3. Observatory Automation
  4. DIMM
  5. Mauna Kea Atmospheric Monitor

MegaCam operation

No major problem and only one minor problem with the readout and power supply boards were reported to SAC. Maintenance activities are scheduled for July 2009. As usual, engineers from CEA will be invited to attend them. A new draft of the Memorandum of Understanding between CEA and CFHT is currently being discussed at CEA/IRFU. No major changes, with respect to the previous one which ended at the completion of the CFHTLS, are foreseen.

The SNDICE device installed on MegaCam proved more difficult to use than expected. A paper relating the tests made late 2008 is in preparation.

WIRCAM operation

Operations with WIRCAM went smoothly. On-going efforts to improve the data reduction operations and calibrations require software developments.

ESPaDOnS operation

The long standing issue of polarization crosstalks, which was believed to have been solved by the use of new triplet lenses, has returned. The level of crosstalk was 4.2% in May, i.e. well above the goal of 1% recommended by SAC. The crosstalks are probably caused by stresses on the lens due to the aging adhesive used to hold the lens in their cells. The level of crosstalk may vary with position on the sky.

SAC extensively discussed the crosstalk issues as their return might have important impacts on some on-going and future ESPaDOnS programs. SAC considers the ESPaDOnS crosstalks to be the most serious instrumental problem currently facing CFHT.

Recommendation 1: Solving the crosstalks issue of ESPaDOnS

SAC recommends that CFHT implement a systematic program to investigate the ESPaDOnS crosstalks to confirm the source and to characterize their intensity and variability on short and long timescales. The ultimate aim is to reduce them to the 1% level.

Back to top


If no hardware solution is possible, the implementation of a calibration procedure able to model and correct for the crosstalks at the 1% level should be investigated.

Recommendation 2: Informing the users on the ESPaDOnS Crosstalks

SAC recommends that the director of CFHT inform the PIs who were allocated ESPaDOnS time in 2009B about the high and variable crosstalk level and get confirmation that their science program is not affected by this. In addition, the crosstalk issue should be clearly stated in future calls for proposals until it is solved.

Back to top

SAC was presented with on-going investigations into the value of replacing the liquid nitrogen cooling of the ESPaDOnS CCD with a CryoTiger closed-cycle cooler. The aim is to reduce the delivery and storage problems, and cost, of LN2. Promising results were obtained with a borrowed CryoTiger system.

SAC encourages CFHT to pursue the tests and discussions with the PI of ESPaDOnS on that opportunity. If there is consensus, CFHT may go ahead and proceed to the implementation of the CryoTiger system.

Derrick Salmon presented a report on CCD readout tests made with the sister of ESPaDOnS, NARVAL. Its CCD was read in split mode using two amplifiers - a technique also foreseen for ESPaDOnS in the framework of a change of its EEV1 detector. The data reduction proved not to cause any problem. The implementation of the new E2V detector would require updates of the data reduction sofware. The four options that were proposed by SAC during its November 2008 meeting were investigated by CFHT. After initial discusion with the ESPaDOnS P.I., CFHT determined that the options involving modification of the Libre-Esprit software will be difficult to implement, due to the currently proprietary nature of the software. Developing completely new software to replace Libre-Esprit would require too many resources from CFHT. The remaining option is to update an already existing alternative open source code.

CFHT staff has indicated that, in fact, a new detector would provide a time savings of at least 30 minutes per night (reported during the May 2008 meeting).

In addition, SAC points out that while many current programs would have only small improvement from the new detector, there are high time-cadence programs that effectively cannot be done, that would be enabled with this change. Therefore SAC estimates that the data reduction software issue should not be a barrier for the implementation of the new detector.

Recommendation 3: Change of detector of ESPaDOnS

SAC recommends CFHT move ahead with the implementation of the E2V detector in ESPaDOnS.

Back to top


In addition, SAC expressed the serious concern of the user’s community of ESPaDOnS of not having access to the sources of Libre-Esprit and having to use it as a black box. This further urges the development of alternative data-reduction software.

Recommendation 4: Data reduction software of ESPaDOnS

SAC recommends that CFHT identify and adopt existing open-source échelle spectropolarimetric data reduction software as the core of the ESPaDOnS pipeline. The open-source reduction pipeline should be available both for use at CFHT by the QSO team and for local execution and optimization by individual users.

Back to top


Telescope operation

Derrick Salmon reported to SAC a potentially-serious telescope pointing problem. Currently, in practice only ESPaDOnS observations are hampered by these larger-than-expected pointing errors in declination. The origin of the problem as well as its starting date are still unknown. It may be related to the 2006 earthquake. Its analysis may reveal other more serious mechanical problems although short-term alarm is not warranted.

SAC shares CFHT's worries that the Dec pointing problem may point to some deeper telescope problem (rather than simply impacting the telescope operations with small losses during field acquisition for ESPaDOnS). Beyond monitoring the problem, engineering time should be used to more actively explore the issue.

Environmental and dome seeing studies

The bad weather conditions since the May 2008 SAC meeting has not allowed collection of enough data with the newly-installed DIMM to measure and interpret trends with regard to the dome seeing. However a precise investigation of the MegaCam image quality (to be published in PASP) has concluded that the median observatory-induced seeing may be as large as 0.43” fwhm. Dome ventilation may be critical to reducing this induced seeing. The study also uncovered evidence that the image quality at some azimuths is clearly affected by the hot-air exhaust vent.

SAC encourages CFHT to pursue the studies of the dome seeing as image quality improvements are always desirable, as well as being a fundamental element in the study of an instrument like `Imaka.

The external seeing will be monitored by the Mauna Kea Atmospheric Monitor. This equipment procured by Keck will be installed on the CFHT site before the summer. According to a MoU signed between Keck, IfA and CFHT, CFHT will maintain the monitor and take care of minor failures. CFHT will be financially compensated for these efforts.

Observatory Automation Project (OAP)

The OAP has progressed according to schedule. The goal is still to have remote observations by the end of 2010. Six CFHT observers are already training for that purpose.


PUEO was put back on the telescope during period 2008B for engineering and science observations made in service mode. The setup turned out to be particularly painful. The finalization of the Fly Eyes experiment has been postponed for lack of human resources and some engineering nights were given back to the QSO.


Test of the `Ohana concept will be carried out with the `Ohana Iki experiment - two moveable 30 cm diameter telescopes feeding each one arm of the CFHT/Gemini interferometer. If the in-house design review is passed, first interferometric observations might occur in the fall. The CFHT-Gemini coupling should occur by the end of 2010B. This should enable the science characterization of the concept - proving that science can be done combining light with two different telescopes not designed as interferometers - and the end of the project. Science observations with the Keck telescopes have so far been unsuccessful mainly due to bad weather conditions.

Laser activities

Christian Veillet reported SAC on the laser studies for Laser Guide Stars generation and the past and current CFHT involvement. CFHT started a collaboration with the Laboratoire de Spectoscopie Physique (LSP, Grenoble) during the VASAO feasibility study. The VASAO project proved to be unfeasible and was terminated. The studies however lead to the development of a modeless laser able in principle to generate a polychromatic laser and thus to carry out tip-tilt corrections. The laser was purchased during summer 2008 with some CFHT funds and is planned to be tested on the Mauna Kea. These efforts have however been hampered by the rejection of a funding proposal by the Agence National de la Recherche.

SAC was informed that LSP will decide on future actions during a meeting scheduled on June 10, 2009.

SAC took note of this LSP meeting and expects CFHT to provide a report on the outcome of the project during the November 2009 SAC meeting.


2. New instrumentation

Derrick Salmon, Christian Veillet and Jean-Luc Beuzit reported on the latest developments of the new CFHT instrumentation projects.


Following the decision of the Board to ignite the Phase A study of SPIRou, CFHT sent the PI of the instrument a Phase A requirement documents together with the CFHT instrument design specifications. Misunderstandings regarding the objectives of the phase A - an assessment of the most critical aspects of the project or a more formal process leading to a Preliminary Design Review -, lead to numerous discussions between the CFHT management and the PI and have prevented the real start of the phase A. Jean-Luc Beuzit was assigned by CNRS/INSU to follow the project as a mediator.

The situation improved thanks to a productive face-to-face meeting which was organized in the Laboratoire Astrophysique de Toulouse-Tarbes (LATT) on May 15/2009. The LATT expressed a strong interest in developing SPIRou. The observatoire de Grenoble should also take part in the optical design and give support for the integration of the instrument. It was agreed that the so-called phase-A process should lead to a “Conceptual Design Review” which will be examined by external referees and SAC during its May 2010 meeting. The need for a project manager has been recognized by all parties and two local motivated candidates at LATT were identified: one for the phase A study and one for the construction phase. The absence of a system engineer was considered a concern by Jean-Luc Beuzit. No qualified candidate has yet been found. Christian Veillet expressed worries on the lack of human resources in the coming months. Indeed the MUSE project (for VLT) currently keeps the LATT technical staff, including the putative final project manager, very busy. Full manpower could be given to SPIRou only once MUSE is delivered. If MUSE is not delayed, the final design review of SPIRou may occur in spring 2012, with the instrument not reaching the telescope before 2015 (representing at least a year delay relative to roadmap in the SPIRou proposal).

SAC is worried about the current delay of the SPIRou timeline. Initially, one attractive aspect of SPIRou was that it could fill the gap between the current instrumentation and that considered as a long term option: an instrument as ambitious as `Imaka may indeed not be ready before 2017. Should the latter become a reality and become the unique instrument of CFHT - which might be imposed by its optical design - the time slot for SPIRou may become narrow. SAC wonders whether the scientific cases listed in the feasibility study will be competitive and timely enough if the instrument reaches the telescope on or after 2015, especially if it is available only for a small number of years. SAC was also told that instruments with similar scientific goals (the detection of exoplanets around cool stars) that do not require the polarimetric mode are currently being discussed at other observatories. The polarimetric mode may give a strong advantage to SPIRou, but this should probably be better justified given the competition.

Recommendation 5: Update on science with SPIRou

For the conceptual design review to be examined by SAC during its May 2010 meeting, SAC recommends that a science requirements document be generated which clearly delineates which science cases require which SPIRou capabilities. This will be used by SAC to assess the science impact of the now-delayed delivery of SPIRou.

Back to top


SAC was informed about the numerous activities since the beginning of the year and the decision by Board to position this instrument as a long term option. The CFHT staff played an active role in the instrumental developments, organizing regular telecons. Derrick Salmon currently acts as the project manager. To reinforce the involvement of France and Hawaii in the project, two French and a Hawaiian co-PIs have been appointed to complement the Canadian PIs. A half-day meeting was organized May 15/2009 in Paris to present the project.

So far most of the work has concentrated on three tasks:

  1. The optical design. Designers were contacted and asked to provide solutions for an ambitious requirement of 0.3 arcsec resolution over a 1 degree diameter; answers are expected by the end of the summer from 6 teams (who will receive some funds for this work). Preliminary studies indicate that it may be difficult/impossible to meet the 0.3" goal over such a large diameter.
  2. The optical turbulence profiling. The work is led by SAC member Mark Chun who presented the on-going experiments to characterize the ground layer turbulence at CFHT.
  3. Numerical simulations of the Ground Layer Adaptative Optics. CFHT staff member Olivier Lai is responsible for this task which should provide indications on the GLAO system performance.

SAC is pleased to see the obvious proactive energy with which the `IMAKA project is advancing and congratulates CFHT staff for its strong commitment. SAC eagerly awaits to see the progress by this coming November. SAC already noted however that the initial large field of view may not be achieved and would like to know the impact of a reduced field on the scientific goals. SAC repeats the position that the image quality should not be compromised.

Recommendation 6: Field of view of `IMAKA

SAC recommends the `IMAKA project poll the science team for a clear statement of what minimum field of view is needed to make impact in the various science programs and deliver this to SAC by the end of October 2009.

Back to top


FIRST was proposed as a possible visitor instrument. SAC was informed that the project received only a partial financial support from CNRS/INSU. A critical issue raised by CNRS is the lack of manpower as the team of the PI is involved in higher priority projects.


SITELLE was proposed as a possible guest instrument. A proposal to get funds from the Canada Foundation for innovation has been submitted. Results were not known at the time of the SAC meeting.


Christian Veillet informed SAC about the latest developments of the MegaMOS project and that of similar instruments. The PI of the MegaMOS concept has submitted a proposal to Calar Alto for an instrument similar to that proposed for CFHT. Its scientific case has been extended compared to the restrictive one already examined by SAC (which was not considered competitive enough in the November 2008 review). At CFHT, MegaMOS would benefit from a larger field of view. In an international context, the need for wide field-of-view spectroscopic capabilities has been identified at the European level; a working group under Astronet has been settled and will provide conclusions within one year. Further developments are expected following the decision of the Gemini Board to abandon WFMOS.

SAC will continue to monitor worldwide MOS instrumental development activities and would be open to seeing a revised science-case document for MegaMOS.

4. Telescope use and scheduling

QSO report

SAC was given the usual report on Queued Service Observing (QSO) for the period 2008B. The only major change is the integration of non-sidereal tracking in the QSO mode for all three instruments.

Like in 2008A, MegaPrime suffered from very adverse weather conditions, leading to an unusually low validation rate. While the completion rate remained acceptable, the average number of hours per night was 5 hrs instead of the expected 5.5. Due to the poor weather conditions, WIRCam only got 5.5 hrs per night instead of the expected 6 hrs per night. For ESPaDOns, the average is 6.6 hrs per night instead of the 7.5 hrs used in the telescope scheduling.

Only 40% of the ESPaDOnS programs could be completed because of the bad weather and the high number of time-constrained programs.

SAC expects to receive further details on the scientific impact of the low completion level of ESPaDOns programs. A recommendation on that matter had already been made during the November 2008 meeting. Further discussions on the need to update the hours/night conversion factors and/or change rules for the ESPaDOnS time-constrained programs are postponed to the next SAC meeting.

CFHTLS report

Jean-Charles Cuillandre presented to SAC the CFHTLS report. All observations, including the new SNLS i’ references, are now completed. The final time budget for each survey is: SNLS: 1186 hours (50%), Wide: 934 hours (39%), Very Wide: 253 hours (11%).

The CFHTLS T0006 release will occur in three steps: the Wide data in July-August, the Deep in August-September, and the Very Wide in September-October.

SAC was told that the incorporation of the Very Wide data (considered not a priority in its November meeting) should not cause any further delay.

The report on the CADC activities related to the CFHTLS did not include any statement on the development of tools to explore the CFHTLS database in a user-friendly way. The November 2008 recommendation on that matter has not driven any action on this issue. The SG is investigating other options at the Terapix level.

SAC congratulates the CFHT QSO team and the steering group for the successful completion of the CFHTLS.

Report on the on-going Large Programmes

SAC received a detailed report from the CFHT QSO team on the progress of the on-going Large Programmes. PIs were contacted by SAC to have further information and feedback.


This Large Program benefited from particularly-good weather conditions in 2008B and was completed at the 100% level. Issues with the photometric calibration were raised by the PI and addressed by the CFHT MegaCam team. The scientific results obtained by this program, which had previously already collected data in normal PI mode, have already been presented in several conferences.


The first semester of observations of NGVS was negatively affected by the adverse 2009A weather conditions. CFHT queue operations report that NGVS lost no time in 2009A due to collision with PI programs. It is expected that less than half of the allocated hours in 2009A (194 hours) will be validated. To allow more flexibility, NGVS will get 8 nights in 2009B to be deducted from the 2010A allowance. It may later be formally decided to split the NGVS observations between A and B semesters (as Virgo is already visible at the end of the B periods), instead of confining it to the A semester as originally proposed.

The observing strategy required some developments and updates in the Queue phase 2 and Elixir data reduction software. Interactions between the PI and the CFHT staff on that matter were reported to be highly effective.

A dedicated archival website and wiki site have been implemented. The first meeting of the NGVS consortium (open to the whole community for part of it) is scheduled in October 2009.


In 2008B, the completion rate of the MaPP program was fair: 80% despite the bad weather late in the semester.

A wiki site for the MaPP co-Is is available. The general Narval/ESPaDOnS database advertised in the LP proposal is not yet operational.

Taiwan has withdrawn from the LP as the principle Taiwanese co-I in the project changed institute. The T-share was then split between the F and C agencies.


MiMeS has progressed correctly in 2008B, with a completion rate of 73%. The data quality is nominal. Several new detections of magnetic field were already obtained. A workshop attended by about 50 persons was organized in May 2009 to discuss the first results.

The raw and pre-processed MegaCam data of the two LPs can be downloaded from CADC by F and C users who have registered for CFHTLS. Non-standard Elixir images produced for the NGVS are not yet available to the general communities. While the raw ESPaDOnS data of the two LPs are in principle available at CADC, they are in practice unusable; the reduced data produced by Libre-Esprit cannot yet be downloaded. CADC is currently working on an interface to deal with the ESPaDOnS products. The issue on the slow rate of data transfer between CFHT and CADC raised during the November 2008 SAC meeting has now been solved.

Discussion on the availability of the LP data to the C and F communities will be further discussed during the November 2009 SAC meeting.

In summary, SAC is reasonably satisfied with the Large-Program progress except for some worries about the unfortunate bad weather hindering NGVS. SAC estimates it to be too premature to take action (beyond being comfortable with the shift of some allocated time from the B to A periods).

Future LPs

SAC extensively discussed another call for Large Programs. The timeline for a program to begin in 2010B would necessitate an award in spring 2010, implying proposals due in early 2010 with a call in fall 2009. If a letter of intent is to be used, these would need to be due before the 2009 SAC meeting, and thus a call must go out during summer 2009.

In reaching its decision about another Large Program call, SAC considered the following inputs:

  1. In 2008B-2009A PandAS acquired 100% of its data, while the two ESPaDOnS programs were in the 80% range. NGVS will be significantly below this.
  2. CFHT operations staff indicate that with the current mix of large and PI programs, the queue is operating fairly effectively, although it would be easy to arrive at a state of empty queue if the program variety was significantly reduced. In that respect, another program at a low-pressure RA or using WIRCam may be welcome.
  3. The addition of another Large Program could however make the queue very fragile since LPs usually have a limited range of RAs or observing conditions, and thus more LPs could decrease the overall telescope productivity.
  4. Another time-constrained ESPaDOnS proposal would be extremely difficult to incorporate.
  5. A large MegaCam program anywhere near Virgo would result in both NGVS and the new LP to have minimal chances to obtain all their data.
  6. Having a new LP that acquired data beyond semester 2012B (when all the current LPs will have finished) was considered by SAC to be undesirable; a call in late 2011 for LPs beginning in 2012B would thus keep all LPs on a level playing field.
  7. Both C-TAC and F-TAC repeated that the quality of the science represented in the PI proposals was high, and thus poor science is not getting scheduled in the PI agencies. It was pointed out that for 2009B the number of proposals, the hours requested, and the pressure on the now slightly increased PI time all went up. The pressure is currently around 2 in both agencies.
  8. SAC felt keeping to the original limit of 40% LPs was reasonable. The current LP pressure varies from a maximum of 32% in 2011A to a minimum of 18% in 2012B when only the two ESPaDOnS proposals are still running. If the 40% cap is kept, then for 2010B-2012B (inclusive) there are about 95 total nights available (roughly 500 hours), producing a non-uniform availability ranging between 10 to 28 nights per semester.
  9. Finally, the overall quality of the proposals received after the first call for Large Programs was considered to be very good by the LPTAC and SAC. There is thus surely a pool for scientifically excellent projects.

In summary, SAC considers that CFHT might benefit from additional Large Program(s) provided that it/they do not impact operations of the current LPs or make the entire queue fragile. The probability of not receiving any proposal matching these criteria was considered high by SAC. For this reason, SAC proposes a process in which Letters of Intents would be required before the formal call for proposals.

Recommendation 7: Impact by new LPs on current ones

SAC recommends that CFHT be rapidly requested to provide a list of constraints any new LP would have to satisfy in order to have negligible impact on (a) the current LPs, and (b) queue operations in general.

Back to top

Recommendation 8: Call for new constrained LPs

SAC recommends Letters of Intent for another constrained Large Program(s) be solicited in late summer 2009, with Letters of Intents due October 2009.

Back to top

The announcement of the Letters of Intent (LoIs) will provide the CFHT constraints and indicate the expectation that only roughly 1 or 2 additional LP will be allocated, which should request a minimum of 100 hours, with up to about 500 hours if appropriately spread over 2010B-2012B inclusive. CFHT should provide a table of the current LP pressure of the existing LPs during that period. The solicitation should also clearly indicate that there is no commitment to award additional LPs if the scientific quality is not considered excellent. These LoIs will then be evaluated by CFHT solely to confirm that they satisfy the operational constraint (and thus several pages of text should be available for this). If one or more of the LoIs qualify, those proposals will be requested to submit a full LP proposal in February 2010.

SAC discussed the evaluation issue. Given the difficulty in assembling an LP-TAC last time, SAC thought it doubtful that it is necessary to assemble an external TAC to allocate at most 500 hours to between 0 and 2 programs.

Worries were expressed that using the full C+F TAC concurrently with the 2010B allocations might result in a tendency to 'preserve time for the PIs' and weight against an LP award. If Board shares this worry, then SAC proposes the idea that the LP-TAC be made of 2 C and 2 F TAC members selected by the Board for expertise and lack of conflict of interest. Final decision will then be taken by the full SAC.


5. Future of CFHT

Christian Veillet presented SAC the usual metrics of the Golden Age plan and the status of the current collaborations with Brazil and Taiwan. An overview of the instrumental developments on other 4-m class telescopes in Europe and elsewhere was also presented, giving an opportunity to discuss potential collaborations as well as potential direct competition with similar projects.


6. Next SAC meetings

The next SAC meeting will be held at CFHT November 12-14, 2009. Depending on the dates and location of the next User’s meeting, the following SAC meeting will either be organized in Canada or in Taiwan.

Back to top