Report of the 74th meeting of the Canada-France-Hawaii Telescope Science Advisory Council

12-15 November 2008, Waimea

Board Resolution on the New Instrumentation Proposals
Recommendation #1 ESPaDOns new detector tests
Recommendation #2 ESPaDOnS new detector software
Recommendation #3 CFHTLS; Data Quality, Very Wide
Recommendation #4 CFHTLS; “Google Sky” graphical interfaced
Recommendation #5 Large Programs, next call

SAC members Pierre-Alain Duc(chair), Laura Ferrarese (vice-chair), Chris Ftaclas, Brett Gladman, Cecile Gry, Jean-François Gonzalez, Robert Jedicke, Henk Hoekstra, Dae-Sik Moon and Denis Mourard attended the meeting. Jeremy Lim from the Academia Sinica Institute of Astronomy and Astrophysics in Taiwan and the chair of the French Time allocation committee, Coralie Neiner, attended the meeting as invitees. The CFHT director Christian Veillet, Derrick Salmon, Daniel Devost, Nadine Manset, Glenn Morrison, Loic Albert, Jean-Charles Cuillandre (for CFHTLS) participated in presentations and discussions.

1. Science talk

Following the decision of the SAC to include in its yearly meetings in Waimea a scientific talk to the CFHT staff and the SAC members, Laura Ferrarese presented a study on the scaling relations of galaxies, as investigated thanks to the ACS Virgo Cluster Survey. She also reported on the objectives of the Next Generation Virgo Survey, one of the new CFHT Large Programs.

2. Report on feasibility studies for CFHT next generation instruments

Subsequent to SAC’s examination of the five feasibility studies submitted after the call for the next generation instruments, SAC’s recommendations were considered by the Board of Directors at its December 2008 meeting; the Board’s deliberations concluded in the following resolution: [Board of Directors Resolution]

3. Technical activities report

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

SAC agrees with the following development priorities proposed by CFHT:

1  Operational Priorities

2  New Instrument Development

3  Observatory Automation


5  Mauna Kea Atmospheric Monitor

MegaCam operation

MegaCam has become much more stable over the past few semesters. There were no reports of instruments failures since May. The CFHT staff performed a major maintenance of the instrument in June in which three engineers from CEA participated. These operations contributed to: making future interventions on the instrument much easier (e.g. modification of the shutter switches), preventing minor intermittent failures (e.g. modification of the  shutter timing circuit, replacement of the SLINK and serial communication fibers) and avoiding sources of wear (e.g. replacement of the rotating valve). The pulse tube, connected hardware and helium lines were cleaned. They had shown prominent signs of contamination with dust generated by wear in the previous rotating valve. The SNDICE clap, a photodiode used for photometric calibration of MegaCam, was installed. SAC was told that it is harmless to the instrument. At the next SAC meeting a report on SNDICE will be presented.

SAC congratulates the CFHT staff for all the improvements that make the instrument more reliable and less subject to wear. The SAC chair received positive feedback from the CEA team that was invited to CFHT during the MegaCam maintenance.

SAC was informed that the MoU with the CEA regarding MegaCam will expire soon and needs to be extended. At CEA, ongoing discussions with IRFU are attempting to keep an official contact person for MegaCam in case of a major instrument failure.

WIRCam operation

Derrick Salmon confirmed to SAC that all crosstalk has been eliminated. The suppression of the edge cross-talk has generated increased ground noise which was solved using an improved controller ground configuration.  It was noted that the readout noise was a factor of up to 2 higher than measures obtained in the storage environment or at other telescopes but lowering it would require much work.

In view of the limited impact of a higher readout noise for infrared imaging observations (usually dominated by the sky background), SAC thinks that the reduction of the readout noise level should not be a priority.

Loic Albert presented to SAC a new observing mode now offered with WIRCam: the guide window may be used for science acquisition at high frame rates. Various science cases were presented that might benefit from this observation mode:  e.g studies of exoplanet secondary eclipses.

ESPaDOns operation

SAC was informed of the success of a major improvement to the instrument: the reduction of the polarization cross-talk, which had long been recommended by SAC. The use of one of the three new triplets lens purchased by CFHT results in a polarization cross-talk at the goal value of 1%. Unfortunately the two other triplets do not work properly for reasons which are still being investigated.

Among the other improvements achieved during the last semester are the implementation of an auto-focus model, the modification of the guider CCD shutter that had intermittent problems, and the use of extra-mechanical supports to secure the electronics crate (that had failed in June causing the loss of a calibration fiber). Among the planned improvements is the implementation of a new telescope pointing model. Due to the poor accuracy of the current one (between 7” and  24" rms) the time spent on target acquisition is long, amounting to 15 to 30 minutes per night.

During its May meeting, SAC was presented a plan to improve ESPaDOns efficiency by changing its EEV1 detector to a new E2V detector. The ESPaDOns PI has expressed concerns about data reduction with the E2V dual amplifier while other astronomers believe it is feasible. SAC was informed that the decision depends on tests carried out with ESPaDOns’s sister instrument at Pic du Midi - NARVAL. The data were obtained but not yet reduced at the time of the November SAC meeting due to lack of time and manpower.

Recommendation 1 – ESPaDOns new detector tests

Given the increased efficiency of the new detector and the fact that ESPaDOns will still be used at CFHT for at least four years, SAC recommends that CFHT analyze the data already obtained with NARVAL and provide a report before the next SAC meeting.

Back to top

Recommendation 2 – ESPaDOnS new detector software

SAC recommends that CFHT investigate options to allow the ESPaDOnS reduction software an appropriate handling of two amplifiers on the detector, should this configuration be selected.

Back to top


A new science run is scheduled at the Keck telescopes. Once carried out, material will be transferred to CFHT in preparation for the CFHT-Gemini link. The above-ground conduit (underground at the CFHT parking lot) between the two telescopes is in place. The ‘OHANA team requires at least one more year of work before the first fringes may be obtained. The interferometer will probably be tested first on two small telescopes, a project referred to as ‘OHANA IKI.

CFHT environmental and seeing studies

CFHT provided a detailed report on the DIMM to the SAC. In August it was re-located to the shutter well of the dome and uses a new mounting structure designed by CFHT. It operates autonomously and reports seeing estimates in real time on an open web site. CFHT has initiated correlation measurements between the DIMM seeing, the MegaCam/WIRCam image quality, and the dome temperature. Some interesting trends have already been discovered but require confirmation with more data.

SAC congratulates the CFHT team for the finalization of the DIMM unit. It will provide useful data for CFHT operations and the phase A study of an instrument like ‘IMAKA. In fact, all Mauna Kea observatories might benefit from the DIMM.

Estimates of the external seeing are required as reference studies for the dome seeing. For this purpose, CFHT envisions using the TMT seeing monitor recently acquired by the Keck observatory.

Glenn Morrison presented SAC a detailed report on the Mauna Kea All Sky mid-infrared camera, an initiative lead by CFHT and IfA to provide all Mauna Kea observatories a tool to map the nighttime cloud distribution. During the summer 2008, a demonstration unit - the Solmirus All Sky Infrared Visible Analyser (ASIVA) - was installed on the roof of the MKO summit lunchroom near the UH 88-inch. It consisted of a mid-IR camera measuring the black-body emission of clouds and a visible wavelength camera. The tests were very successful, showing that even thin cirrus, invisible at optical wavelength, is detectable in the mid-IR. The Mauna Kea directors have approved the purchase of the unit which should be installed by mid-2010. Data will be sent via an optical fiber to a server accessible to all MKO telescopes, located in the CFHT.

SAC is pleased to learn that this useful facility will soon become a reality. The mid-infrared all-sky camera will make queue scheduling even more efficient.

Observatory automation

A report was given to SAC on the significant progress made on the Observatory Automation development plan. Some of its sub-projects have already passed their final design review and are in the implementation phase. The first systems to be remotely operable will be the dome shutter, mirror covers and dry air systems.

A risk management plan was presented to SAC. A list of risks was compiled pertaining to the absence of summit night staff. They range from earthquakes to a coffee pot left on.... Additional security issues were discussed during the SAC meeting such as intrusive entry to the CFHT.

AO activities

Christian Veillet presented the results of the VASAO two-color experiment. A test on the sky carried out in November 2007 had shown no apparent correlation with atmospheric tip-tilt. Another experiment took place in June 2008 with a larger detector. The continued lack of evidence for a correlation may be due to insufficient signal to noise. A paper to be published in a refereed review will close this VASAO experiment.

Christian Veillet informed SAC that the modeless laser experiment, supported by CNRS and of interest for other observatories on Mauna Kea, will be pursued in 2009 though with a low priority. The laser was ordered in August and has not yet been delivered. The SAC was not provided a detailed report on this experiment. CFHT is planning a couple of engineering nights in 2009A for the FlyEyes on-sky qualification.

SAC noted that the level of AO activities (VASAO, experience with the modeless laser) planned for 2009 has decreased and that they are no longer mentioned in the list of priorities. The AO effort from now on should be related to the new generation of instruments, in particular `IMAKA.

4. Telescope use and scheduling

Queue Service Observing

After the departure of Pierre Martin, Nadine Manset has taken over the responsibility of QSO operations. She presented a detailed report to SAC on the queue operations for semester 2008A.

2008A was the first semester for which all the three major instruments were in queue mode. The transition between visitor and queue mode for ESPaDOns programs went smoothly despite a few initial problems. Following a SAC recommendation, a survey was sent in August to all ESPaDOns PIs to inquire about the phase 2 process, data acquisition and data retrieval. Almost all PIs who replied (half of them) were satisfied with the QSO mode.

SAC thanks the QSO team for their dedication and careful management. The transition of ESPaDOns programs to queue has not been particularly painful and damaging to PIs. SAC is impressed by the quality of the web interface developed for the data retrieval.

SAC heard that about half of the ESPaDOnsprograms that were started in 2008A could not be completed due, in particular, to the presence of many time-constrained proposals. Only four of seven A programs were completed.

SAC understands that ESPaDOns programs are more risky and that a lower level of completeness compared to other instruments is expected. However, SAC is concerned about possible negative scientific impacts of having too many uncompleted time-constrained proposals accepted for the queue. Indeed, for programs consisting of stellar phase monitoring, an interrupted program may have a poor scientific value or be useless. More statistics and a careful monitoring of the situation are required before any useful recommendation can be considered on this matter.

SAC learned that one MegaCam program that could not be performed for technical reasons was carried out with WIRCam.

SAC believes that the TACS should be notified of such important changes in strategy.

During its meeting of May 2008, SAC recommended that CFHT inform the chairs of the national TACs about the status of the PI programs after the end of each semester. SAC further recommends that the information include details of the incomplete programs (e.g. for ESPaDOns programs indicate if a time sequence run had to be interrupted) and information about significant changes of observing strategy.

In 2008A, MegaPrime programs of priority A were completed with 85% efficiency while B programs achieved 73%. The MegaPrime validated data averaged about 5.3 hours per night, higher than previous A semesters but lower than the average number of validated hours in the B-semester (because of bad weather affecting mostly the winter runs) and also lower than the number of validated hours per nights for WIRCam and ESPaDOnS. In 2008A, all A-queue WIRCam programs and 97% of B-queue programs were completed. The average number of validated hours was as high as 6.2 hours per night. For ESPaDOns, the average number of validated hours was 6.4 hours. For reference, during the last few semesters the conversion factor used for scheduling the observing runs were 5.5, 6 and 7.5 hours per night for MegaCam, WIRCam and ESPaDOns respectively.

SAC discussed whether the conversion factors between hours and nights should be updated more frequently. In particular, B semesters seem more favorable for MegaCam runs. But given the lack of statistics and the need to have a reference, SAC believes it is premature to change the conversion factors.

The telescope time between the different agencies is well balanced. CFHTLS did not request all its allocated time (see below). One Canadian ESPaDOns B program could not be scheduled due to a direct conflict with a more highly ranked program. Time was given back to the Canadian agency in 2009A and CFHT recognizes that the problem was caught too late in the process.

Status of C programs

C programs are those below the cut-off line that are used to overfill the queue. Depending on the instruments and semester, the number of C programs started and completed may not be negligible. The decision to allow a C program to enter the queue is made by CFHT and not by TACs. This could be a concern. C programs are also used to ensure the agency balance on a per instrument basis. Since this could have an impact on the science carried out at CFHT, SAC requested during its May meeting a detailed report on the status of C programs.

The report provided to SAC clearly indicates that in terms of hours, C-programs represent only a small fraction of about 5% of A+B programs. SAC was also told that the per instrument agency balance was, from an operations point of view, the most practical solution to ensure the agency balance at the end of each semester. CFHT is otherwise limited by the fact that instruments cannot be switched during observing runs. The only other degree of freedom is changing the length of an observing run.

SAC took note of these conclusions.

Discretionary time in 2008

SAC was presented the list of proposals that obtained Director Discretionary Time in 2008. The F-TAC president noted that a couple of them from external PIs were rejected by the F-TAC. Christian Veillet replied that proposals that present a scientific or technological interest to CFHT and its staff may be accepted in the DDT framework. Proposals that require too many operations from CFHT and affect regular PI programs are rejected.

The D-time benefits the CFHT staff. It is used when there is little pressure on RAs and, as a result, not all the available D-time is used and some nights are returned to PIs.

Telescope scheduling in 2009A.

Christian Veillet presented SAC the results of the CFHT TAC meeting as well as the agency and instrument trends. MegaCam was once again the instrument most requested by PIs and the one allocated most of the observing time in 2009A - about 52 nights. WIRCam received 37 nights and ESPaDOnS 24 nights. MegaCam and WIRCam are popular with the Canadian community while the French community often requests ESPaDOnS. The proposal success rate per instrument varies strongly from one national TAC to the other: it is the highest for ESPaDOnS on the F-TAC and for MegaCam on the C-TAC. No AOB-IR PI time is scheduled for 2009A. In 2009A, the semester was made 3.2 nights longer for each agency to incorporate a day-time program. The pressure on the Canadian agency (1.85) remained constant despite the increase in the number of available nights, while the pressure on the French agency dropped to 1.6.

SAC discussed various hypotheses to explain the decrease of telescope pressure on the French agency that started a few semesters ago but accelerated in 2008B. It is surely partly due to the fact that many previous French users of ESPaDOnS are now collaborating in one or both of the approved ESPaDOnS LPs and are therefore not applying for PI time. SAC further speculated that the additional pressure introduced by the LPs in the 2008A semester (combined with the traditionally high pressure, about 4, from the French side) might have further discouraged MegaPrime or WIRCAM PIs from applying for time. Before the call for the 2009B semester, SAC encourages CFHT to post on its website a detailed breakdown of the pressure from the LPs as a function of RA and, if known, observing conditions (as recommended in the May report).

The end of the CFHTLS observations

The chair of the CFHTLS Steering Group, Jean-Charles Cuillandre, presented SAC a detailed report on the CFHTLS observations in 2008B and on the status of the data release. The Legacy Survey has benefited recently from particularly good weather conditions. Thus, PI programs as well as the PAndAS LP did not suffer from the higher priority given to CFHTLS in 2008B. About 20 hours of a total of 108 hours allocated to the LS-Wide were returned to PIs.

The strategy to acquire the new i’ SNLS references has slightly changed with respect to that presented during the last meeting to maximize its scientific return (The SAC chair was informed of the change beforehand). As recommended by SAC, the SNLS team has submitted a PI proposal to obtain the references on the D3 field in 2009A.

The official release of T0004 was announced during the SAC meeting even though the stacks have been available since late August. A few minor problems identified in the preliminary release were corrected in the official release.

Due to the load generated by the production and quality control of the T0005 release at Terapix no time was left to handle the planned release of the Very Wide data set. Given that the T0006 release will likely have to focus on the Deep and Wide data (since it will be based on fully reprocessed Elixir images), the CFHTLS Steering Group recommended that the fully reprocessed Very Wide data should only be part of the T0007 final release (expected in 2010). Terapix could in principle release stacks (without full quality control) of the Very Wide with data based on the currently active Elixir recipe as part of T0006.

The workload on Terapix required by a partial T0006 release of the Very Wide is not considered necessary in view of the fact that 1) the main science goal of the Very Wide does not make use of stacked data 2) there has been no pressure from the community to access Very Wide stacks and 3) most of the Very Wide data already exists in stacked form with acceptable astrometric and photometric precision through a MegaPipe release.

Recommendation 3 – CFHTLS; Data Quality, Very Wide

SAC recommends deferring the data quality control and stacking of the Very Wide data until the final Terapix T0007 release (2010) as proposed by the CFHTLS Steering Group.

Back to top

The SG asked the SAC for guidance in ensuring the legacy value of the CFHTLS. While the SG has plans for data handling, archiving, and documentation on the various sites involved in the CFHTLS (CFHT, CADC, Terapix), the SG is wondering if more effort should be undertaken.

Recommendation 4 – CFHTLS; “Google Sky” graphical interfaced

To help generate a broader access to CFHTLS data SAC recommends that the steering group works with the CADC to implement a "Google Sky" graphical interface for the Terapix release of the CFHTLS similar to what already exists for the MegaPipe Release.


Back to top

Large Program implementation

Daniel Devost presented to SAC a report on the implementation of the Large Programs selected by the LPTAC and SAC last May. The Pan-Andromeda survey (PAndAS) was allocated 75.3 hours with MegaCam in 2008B. At the time of the meeting, 80% of the images were validated. The other MegaCam Large Program, the NGVS, will start in 2009A.  Contrary to the MegaCam LPs that have observing runs spread over different semesters, the two ESPaDOns programs, MaPP (Magnetic Protostars and Planets) and MiMeS (Magnetism in Massive Stars) compete for telescope time. Each proposal was given A and B time but MaPP, ranked higher by the LPTAC, gets priority in each category. In practice, PIs of both proposals have come to an agreement to avoid overlap as much as possible. At time of the meeting only the MiMeS observations had started.

Before its meeting, SAC had asked the PIs for feedback on the LP implementation. All of them have declared satisfaction with the process and praised the help of the CFHT staff. The MiMeS PI expressed concerns about the data transfer to CADC where the LP database is installed. CFHT will investigate why it had not yet occurred.

Next call for Large Programs

SAC was requested by the Board to consider the case for another call for Large Programs (sooner than the two-year interval envisioned in the LP implementation document at the time of the LPTAC). SAC reviewed the amount of time currently allocated to Large Programs and the results of the PI competition for 2009A. The Canadian pressure (1.85) has remained constant despite the increase in the number of nights while the French TAC pressure dropped to 1.6.

As this reflects only a single semester of data and as it was felt by TAC members that the scientific merit of the programs allocated time for 2009A was high despite the lower pressure, SAC believes there is no strong argument for a premature call for another round of LPs. SAC will consider the results of the 2009B PI competition and the LP impact on the PI programs in 2009A and then re-consider the value of a new call for LPs.

Recommendation 5 – Large Programs, next call

CFHT should not consider another call for LPs starting before 2010B.

Back to top

5. Future of CFHT

Christian Veillet presented a quick update on the metrics of the CFHT's Golden Age plan. He informed SAC that Brazil is to become a new partner of CFHT with a status similar to Taiwan.  Brazil agreed to buy 10 nights per year starting in 2009B. The MoU is valid for three years but may be extended. The relative share of telescope time between C, F and H remains unchanged.

The SAC members believe that the arrival of new partners with the corollary loss of a reasonable number of nights to the current agencies will not have a strong impact on the existing communities. The next generation of instruments will benefit from the extra funds received by CFHT.

6. Next SAC meeting

Next SAC meeting will take place 18-19 May 2009 in Paris at Observatoire de Paris.


Board Resolution on the New Instrumentation Proposals

The Board expresses its thanks to the Corporation and to the SAC for the very useful summaries and recommendations concerning the new instrument proposals. The Board found them helpful in guiding discussion and reaching the conclusions summarized below.


The Board thanks the team for presenting a very interesting and innovative concept. The Board encourages the team to continue to develop their concept with a view to developing a visitor instrument for use at the CFHT.

2. MegaMOS:

The Board recognizes that this is a very interesting instrument concept most relevant to the long-term future of the CFHT. The Board asks the CFHT and the SAC to maintain contact with the team and to monitor activity related to this concept and similar instruments on other telescopes.


The Board thanks the team for presenting a well-developed and viable concept for a future CFHT instrument. The Board encourages the team to continue to develop the concept with a view to positioning SITELLE as a Guest Instrument. The requirement is that the Corporation participate in the definition of the project at a level to be reviewed by the SAC when considering the priorities of the Corporation.

4. ‘Imaka:

The Board agrees that ‘Imaka presents a highly promising vision for the long-term scientific future of CFHT. However, the concept is ambitious and has a number of high-risk areas that require continued study and definition. The Board recommends that the high-risk areas be systematically addressed through an R&D study program supported by CFHT. The Corporation is directed to work with the ‘Imaka team to define initial work areas. The studies are to be carried out under the management of the Corporation.


The Board recommends that this concept proceed to a Phase A study.

  • The Board asks the Corporation to prepare a Phase A requirements document in consultation with the instrument team.
  • The Board seeks to have the Phase A study completed at a level suitable for a Preliminary Design Review in March 2010, with the results available to SAC at their Spring 2010 Meeting normally held in May.
  • The Board agrees with SAC that a mid-term review of progress should be held in October of 2009 with the results available to SAC for their Fall Meeting in November 2009.
  • The Phase A study is to be managed by the Corporation through a formal agreement with specified deliverables, such as technical requirements, and under the authority of a responsible individual at CFHT (e.g., a designated Contract Officer).

The Board further notes:

  • That the rough cost estimated in the concept study is consistent with the envisioned resources of the Corporation and of the funding Agencies. Consequently we believe that the Phase A study should also present a prioritized list of options to assist SAC and the Board in reaching a decision on the scope of the instrument should costs expand beyond available resources.
  • That the science case should be strengthened with a clear statement of requirements, including those relevant to the instrument options. We note that instruments addressing similar science goals have been proposed for other telescopes and we encourage the team to make use of existing studies where appropriate.
  • That the instrument concept team may need strengthening in order to meet the intent of the Board to proceed quickly through Phase A and, if warranted, into the construction phase.

Back to top