next previous
Up: Home Page

General Discussion



Note from the Editors:

This is the transcript of a recording of the general discussion that was held on May 19. The text has been slightly edited to make reading easier and understandable within a general context. The discussion was chaired by Gordon Walker and supported by an agenda presented on viewgraphs. Thus, the present discussion is divided in a few parts according to this agenda. Also, we have included our interpretation in the text when the speaker was referring to an item not explained before or presented only on a viewgraph, or when the tape recordings were not sufficiently clear. In most cases these editorial interventions are marked by square brackets.


The NGC Report and the Future of CFHT

Gordon Walker: This is the kind of agenda I set for discussion this afternoon, but... we don't have that much time. I've already gone over the Advisory Committee report. So I think I'd like to open up again discussions on the NGC report, the draft report then drop back to the 2003 scenario which Dennis [Crabtree] provided this morning and - then - the SAC is looking for input or reactions of this meeting to a whole series of issues - and I'm sure I've missed quite a few out.

I'd like to suggest that we might talk for about 15 - 20 minutes on the NGC draft report and similar sort of time on the what I call the 2003 scenario level and sort of midterm and that we make quite sure that we provide answers and reactions to these other items. Are there any reactions to this?

It seems to me that Harvey Richer and Claude Catala's committee has come up with just about everything one can think of and we've heard an eloquent plea for 4-meter telescope doing what it does best all the way through a telescope with a diameter of one hundred meters. I've yet to find out what the wind blowing on a hundred meter telescope on the top of Mauna Kea might be but maybe somebody can explain that to me.

I suspect that with the multi-element interferometer we are going to run into the land and natural resources people who say we got all the telescopes we're allowed up there...

John Hutchings: I'd like to start off just commenting on the NGC Report. I am very much in favor of a large telescope as our ultimate goal for the new CFHT. My feeling is that the 4-8m or the interferometer are interesting but the science is not compelling and in one case its not obvious just how much we can do at all.

I think that what we should be driven by is some telescope that has an unique capability that exceeds that of the 8 and 10-meter telescopes that are currently being put into use and that is either a backup to or complimentary to what the NGST will be achieving in that time frame. And so something of the order, whether it's 25 meters or 20 or somewhere in that ball park is what I think we should be aiming at.

In thinking through the science questions that are going to be up in 10-years time, this is obviously a matter of guess work and personal opinions. But it seems to me that many of our current major issues will be either solved or so different in 10-years from now that they're fairly hard to get at. But I would say that most of the cosmological constants are changing right now ... and will be either nailed down or impossible to take any further.

The question of dark matter is a major one right now. Whether we'll make significant progress by 10 years from now, I can't say, but its likely in my opinion to be the kind of thing that we should be chasing still. It will be a matter of quite a lot of activity.

The obvious other thing that's been talked about that needs time and needs major new facilities is search of the planets and study of planets once we've found them. And that's likely to be, in my opinion, a strong drive within 10 years from now.

The other comment that I have is that there are a lot of other new facilities going into service, not just on the ground; there's going to be AXAF, SIRTF, various UV sky surveys and NGST itself which almost certainly will discover new phenomena that we don't have any idea about right now. These will become major sources of action 10 years from now. So while we can build a telescope that will do things that we can foresee, we want to make sure it's versatile enough to do all sorts of things that we are not going to foresee. That's just the lesson of history.

Finally, in terms of down time and where we put this replacement, I feel very strongly that several years of down time at the CFHT would be quite unacceptable for most of us and that the attraction of perhaps building this new facility on the site of the 88-inch is one that we should strongly go after.

Harvey Richer: I have a comment, then a question for the NGC committee. The comment is that I completely agree with what John says. I think that we need to be ambitious here. When the CFHT was built it was arguably the best telescope in the world and I don't think we should settle for anything less for a replacement or upgrade to it.

The question is: So far all I've heard is everything that has to do with restrictions, environmental or space or otherwise on Mauna Kea and I'm wondering whether the committee is restricted to only considering projects that are located on Mauna Kea or whether alternate sites could be considered for a future telescope or a replacement for the CFHT? There are other sites that are probably or possibly as good as Mauna Kea such as for example the site for the LSA-MMA which will be at 5,000 meters in Northern Chile. It's a large area, very flat, so there should be plenty of room for delay lines.

Tim Davidge: I guess one thing I'm a little uncomfortable with is that the 25-meter option is kind of presented as either/or. You have to get rid of CFHT to have a 25-meter to kind of satisfy the scientific aspirations. I think we need both. I think we need to upgrade CFHT and get a 25-meter telescope. So, you're asking for a lot already - why not ask for even more of a lot! I'd like to see ideally an upgraded CFHT, that's a 6.5-meter say, that puts us within shouting distance of the 8-meter telescopes for spectroscopy and increases the diffraction limit or shrinks it by a factor of 50% which is good for adaptive optics work, etc. And it would support a 25-meter telescope.

G. Walker: I would like to inject a comment for a moment. I'm a little uncomfortable that apart from Bob [McLaren] we don't actually have anyone here from the University of Hawaii, I think. Is that correct? And oh!, I'm sorry Tim [Abbott]. I tend to think of you as CFH. I'm a little bothered if we keep saying that we're going to replace their telescope...

René Racine: On the committee's report, I'd like to make two comments. I would hope that the committee would consider appropriate to discuss and perhaps recommend to the Board, throughout the scheme of things that I was suggesting, that the CFHT really embarks into international or explore the opportunity of embarking into broader collaboration.

On a technical aspect for the 25 or 100-meter telescopes, history is not always a good guide. But if you study the history of the telescopes, you will see that the doubling time of aperture has been 42 years, plus or minus two, ever since 1609 and we're just building 8-meters. So in my book, I'm giving 10-to-1 odds that there will not be a 25-meter telescope on the ground before the middle of the next century.

G. Walker: We've heard very strong support for the very large telescope. We have at least one, probably more protagonists for maintaining the 4-meter, 6-meter level telescope. Is there anybody who feels strongly in favor of the interferometer that would like to speak to us? Unfortunately we don't have protagonists here and I think they actually had a very good case when they presented it. This is obviously a very expensive, technically highly challenging proposal, but I don't hear anything at the moment that suggests it would get significant support in the community. Anybody prepared to speak to the interferometer?

Isabelle Vauglin: I don't want to speak about the interferometer. But, if you are counting the number of people against or for that solution, I want to say that I am a protagonist. I'm strongly supporting the 4-meter telescopes because of the very special domain of mid-infrared astronomy which is particularly suited for a 4-meter class telescope at the precise place of the CFHT, due to the the high quality of atmosphere conditions there. We strongly support to keep the 4-meter telescope at this place.

Thierry Forveille: Maybe I could give an opinion about the interferometer. I am very much in favor of interferometers in general as a former and still radio astronomy interferometrist. I would advise against using five different telescopes, each differently optimized as an interferometer. People who have been doing the VLBI have had to do that for the last 20 or 40 years and they are finally extremely happy of no longer being in that situation of having to use different telescopes optimized for different things as an interferometer. Now that they have the VLBA, everything is so much simpler. So I don't think using existing telescopes as an interferometer is a very good option. Especially at least for the French community which will have access to VLT as an interferometer, which will only have two kinds of telescopes. I don't think this is an attractive option.

Roland Bacon: I would like to bring some information. As you know, the French community had a ``10-year prospect'' meeting one month ago. The output from the French community was that in the mid-term, the strong support for the VLT is a priority. The French interferometer community in France probably won't be able to support another project in this mid-time. But from the ``prospect'', there is a strong support for the NGST in the long term. Also there was a strong support for the CFHT facility in the mid-term. I remember well that the advice was that the French community would like it [to exist] at least until 2005, [and] that the CFHT has no longer any budget cut to maintain its facility at the highest level.

G. Walker: I think this is an important point on which I'm sure the committee wants some feedback. What everybody else will have to understand is the degree to which we are to be without a telescope while a new one is put in and how long - if one is going to this large telescope - we would like the current CFHT 3.6-meter to continue. I've heard a number of opinions about this.

Jean-Gabriel Cuby: I think for the large telescope project the question is more how long will it take to get the agreements between all the members who will contribute to this project, as Harvey Richer was saying. That would be an ``Earth'' [=global] project so we can expect that it will take years to get it funded, to get it approved, designed and then built. So then the mid-term for the CFHT will most likely be more than 10 years. Well, that would be my expectation at least.

Laurent Vigroux: I have one question and one comment. I first wonder what will be the cost of a very large telescope of 25-meter or larger compared to the cost of the NGST. And, as everybody knows, there is a very strong effort now in NASA and ESA to decrease the cost of space missions. The goal for the NGST and the cost for this kind of 8-meter telescope in space is something around $500 million dollars. This is about five times the cost of an 8-meter telescope on the ground and I just wonder if moving to a very large telescope on the ground will [not] make the project more expensive than a series of NGST's that can be specialized for different projects launched on a one or two year basis.

H. Richer: Just regarding the cost, Olivier LeFèvre presented some very very rough guideline on cost and he thought - this is based on the OWL project - that the hundred meter would cost about a billion dollars. And he thought that the 25-meter telescope could be a relatively inexpensive telescope of the order of a hundred million dollars. I think that we should appreciate that a hundred million dollars may not be outside of the budgetary considerations for a telescope funded by Canada and France.

When I was talking about the world telescope I was thinking of the 100-meter, but the 25-meter may not be unrealistic for the Canadian, French and Hawaiian communities.

G. Walker: Okay, I feel that we probably covered the range of opinion. I'm sorry, you want to ask a question or you make a comment?

T. Forveille: It's a question to the audience. Does anybody have a realistic feeling of the likeliness that the NGST will be an 8-meter telescope? Because as a radio astronomer I remember that it was first advertised as a 4-meter telescope. So has anybody any opinion on that?

Simon Lilly: I can just comment as a member of the NGST science working group. My feeling is in a sense somewhat ambivalent. Clearly the budget figure of $500 millions plus international contributions is widely recognized as being very tight. On the other hand, I would say that in all of the discussions that I've heard the aperture size which formally is still scoped as between 4 and 8 [meter] is always at the upper end of that range. The sort of trade-offs which are being much more seriously talked about are, say, the wavelength range and so on. One reason for that is as I think Harvey mentioned, NGST is seen very much in the context of the broader planetary ``Origins'' missions; NASA in particular sees NGST as much as the technology demonstrator for the subsequent interferometry missions as a science mission in its own right. That is clearly a major aspect of NASA's thinking. And so the whole aspect of deployable structures and deployable optics which is what you really need to get up to 8-meters on NGST is an aspect of the mission which NASA is very keen on; so my guess is that - assuming that it does go - it will be in the 8-meter class.

David Crampton: I guess even though I was advocating a 25-meter telescope just a short time ago, I actually wonder whether the committee should consider this hundred meter telescope for example. Because if you're asking me whether I want a hundred meter telescope or NGST, I'd vote for NGST every time. If you're talking about getting to a magnitude 38 in 10 hours - that's for adaptive optics. I've observed a lot with adaptive optics in the last years. Reality is you have the Sun during the day time, you have clouds getting in the way, you have the moon which does not affect your science, but affects the wavefront sensors; you got seeing on the ground. You're going to get much more with NGST and I would say 100 millions for a large telescope on the ground might be worth considering, but beyond that, we should go to space. My initial reaction to a 100 meter on Mauna Kea was: Don't put that on the ground, put it on the backside of the moon!

G. Walker: Prior to that, somebody who will remain nameless asked me to do something dangerous. I'd like to take a straw vote on [these three options], I don't approve of it; it is informal, carries no weight but it might be an useful indication in collection of highly dedicated, intelligent astronomers from at least two of the participating countries. Those who would like to see a 4 to 6-meter telescope maintained on the site, not necessarily CFHT as it now is, as opposed to a significantly enlarged facility...

[No reaction from the audience]

I knew that was going to happen. It's not my idea, I'm just trying ... give us some feeling, I mean... Are there those of you who haven't spoken who would prefer, given the choice, I'm one to item 2. Remembering that ... well in this case, we would be for a continuance of the existing telescope facility with a presumably minimum break for re-installation of and improved primary, secondary, etc. That's what I understand the proposal to be; the staff would have something to work on, we wouldn't be planning departure papers and so on for then. This is a complete replacement and a hiatus of I don't know many years. Okay, am I showing a bias there? All-right, if you would like to vote and if you want to vote for Item 1 [4-6m telescope], now is the time. Would you favor Item 1? In the same place ?

[Vote: About 50% of the audience raised their hands...]

And those of you despite the hiatus that might be involved sometime in the next ... not for 10-years, would prefer to see Item 3 [large telescope]?

[Vote: About 50% raised their hands...]

Now I don't care about numbers. We're not representative in any sense here. Okay? I've done my job.

Instrumentation

G. Walker Let's move on. All-right, this was put up by Dennis [Crabtree] this morning. Seems to me it was a rather useful place for discussion about the mid-term, just getting into the shorter term issues which I really want to get to within 20 minutes or so. The proposal is - or the extreme proposal as cited is - assuming we can afford these items: MegaCam is already in the works and is coming forward. There will be a proposal for a wide field infrared mosaic, we've heard on this time scale that Gemini North will be operating, so will the other 8 meter telescopes which will need almost certainly wide field in the infrared and there are many interesting programs which could be carried out in the infrared and wide field; AOB-IR which we currently have; MOS/OSIS-IR, that some say we have to discuss the OSIS plate scale; [New] GECKO and the idea of the cross-disperser echelle at coudé; dedicated detectors on all of the instruments and a quick change between focii. It's not clear at this point that we have to discuss MegaCam, unless somebody particularly wants to.

J. Hutchings There are things that are about to happen and things that are not at all funded. I'm not clear what it is we're trying to establish at this point.

G. Walker: As I pointed out, MegaCam is funded as far as I know. Is that a correct statement? Right. And therefore it didn't seem to me that it was appropriate to discuss it at this stage. I'd like to have heard what degree of support there is for the wide field infrared mosaic. It is unfunded. It needs a crusader and it needs a lot of effort from the various countries if it's ever going to happen. So that's the context in which I see this proposal.

R. Racine: I'd like to make a general comment about that list. I'd like to suggest that we give priorities to projects where the CFHT can really be the best. I've heard it this morning from Dave, I believe, that MOS is not competitive with what Keck can do. I don't think we can argue about that but it has to be argued and then I would plead to support only those things where CFHT is the leader. Otherwise, it is equivalent to sacrificing the agencies' funds, CFHT staff and the scientists, by doing science which they could do better somewhere and which someone else would do better somewhere and just kill them in press.

G. Walker: Well, the wide field IR is being one of those things. If anybody wants to speak to that I'd like to, I think what we don't have is the feeling for the level of support for the wide field IR.

S. Lilly: I would like to speak strongly in favor of it. I think it's extremely important that we have the capability of feeding the larger telescopes; MegaCam being restricted to wavelengths less than 1-micron is obviously very, very limited in that area. So I would strongly support it. If I understand what René has said, I would put it clearly in the category of a unique, potentially unique capability. So I would very strongly support it.

T. Forveille: I think there is a mixture in that form of instruments which will be just online like MegaCam, which will arrive on line in 2003, and instruments which will be at the end of their useful life in 2003. I would probably put the AOB in that category. AOB on that time scale is going to be something useful essentially to feed other instruments. It would be extremely useful to keep for a fiber-fed spectrograph, but I don't think its going to be competitive any longer as an imaging instrument. So I think we should make a difference on this between the top instruments which are MegaCam, the wide field infrared imager and most likely, I think, we need a fiber-fed spectrograph as a complement to use when the seeing is not good enough for wide field imaging and also when the conditions are not photometric. That's probably the future of the CFH. Essentially as a three instrument telescope, [with] very good instruments, which are the best in the world in that category.

G. Walker: Repeat the three.

T. Forveille: I would say MegaCam, the wide field infrared imager and a fiber-fed spectrograph.

G. Walker I'd just like to throw out a question to the SAC or the Executive of the CFH. Presumably somebody has looked at the question of laser guide star density or whatever on the mountain. MegaCam is going to cover a large area of sky. In the infrared, it is little less serious but with sodium lasers and so forth, is this going to be a hazard or not?

Christian Veillet: Well, there is a group which has been set between the laser observatories at the summit and there is a clear policy which has been defined on the way the lasers will be operated. The observations which could be impacted by laser beam will have always the priority. That's only the starting point of any laser activity at the summit so there is no problem with that, clearly.

G. Walker: Thank you; I guess that point is covered.

L. Vigroux: I want to make a comment on the last item on this list which is not an instrument, which will really save a lot of time and also will optimize the utilization of all these instruments: The ability to move from F/8 to prime focus without top end exchange. This is very important because you don't realize the size and the weight of MegaCam and it will make observation which would be spread over a long period of time, maybe for a few years and with different conditions and programs. We have to be very careful about systematic errors. Each time you remove the instrument and put it down, and you put it back on the telescope, you are not sure that you will put it exactly in the same position; and you will generate sources of error which are not very easy to track down. So possibility to leave the instrument on the telescope is very important but this means that you have to get at least a possibility to have both instruments working at prime focus and Cassegrain.

G. Walker: Okay, we had a specific suggestion and it's rather good that focusing our attention which set the priority instruments as MegaCam, wide field IR, a fiber-fed spectrograph plus these rapid switchings between focii. Does anybody see it as a difficulty with those as a level of priority for this mid-term?

Paul Felenbok: Probably you know that the 4-meter at Cerro Tololo is able to flip from the prime to the Cassegrain. You know they don't exchange the top end, they flip the top end, just flip it around. I don't know if this is something that can be foreseen with MegaCam...

G. Walker: I understood from the proposal that it will be the secondary that swings in and out.

P. Felenbok: So it is just to evacuate it?

G. Walker: It will be on an arm, it'll swing back up and swing out to the side.

P. Felenbok: But it will be out of balance, no?

G. Walker: I don't know.

Pierre Couturier: Just a comment. This transparency of Dennis has been prepared essentially to fix what would be the context of operation in 2003. We are not saying that AOB, MOS/OSIS will have the same status as other instruments. They will be going down more or less at that period. So I would want to strongly support the point of view which has been expressed by Thierry Forveille about having MegaCam, wide field infrared mosaic, fiber feed for spectroscopy and the possibility of flipping the F/8.

G. Walker: Yes, I should emphasize that I put this up here because it's a good basis for discussion.

T. Davidge: I'm just looking around to see if any [members] of the Altair team are here, I don't want to insult them, but I'd like to see how well the 8-meter AO systems are going to work before we kind of predict the demise of AOB-IR. I think it's unique, well - right now - it's unique, it may not be unique in 2003, but it's still a very powerful capability and I think this because some telescopes are larger and have smaller diffraction limits, we still shouldn't at this stage kind of say: Let's predict the future of AOB-IR on CFHT.

G. Walker: I don't think anybody's suggesting the de-commissioning of it. I think we're describing a derivative of it.

Jean-Pierre Maillard: In these three, there is something I don't see clearly about the wide field infrared mosaic. It's a very good case but when do you see it prepared? Because when I see the heavy load that MegaCam puts on CFH right now and for several years to come, I don't know when you place the work for a wide field infrared camera.

P. Couturier: Certainly a good question about what will be the near future of CFH... Clearly, we don't want to have a collision between MegaCam and the wide field infrared camera. The load of MegaCam is essentially in contracts to DAO, CEA, to all partners of MegaCam. Once this will be contracted, there would be a period when we could start working on the wide field infrared camera. And then the acceptance of the infrared camera will come after MegaCam, obviously.

R. Racine: It's too late to ask this question, like two years too late, but there has not been any consideration ever given to put MegaCam at the Cassegrain focus of a renewed telescope? Is it too late to think about that?

[Laughter]

G. Walker: No, no, no ... it's not too late.

R. Racine: I would argue that for the price that you are going to pay for the corrector you can get a new primary mirror for instance. With the price of the new upper end, you can get a new tube or of that order. Then you would only have one focus, Pierre...

L. Vigroux: At the time that we proposed MegaCam, we have eliminated the Cassegrain for field-of-view reasons and everybody, at that time, agreed that it was less expensive to make a new wide field corrector for the prime focus. That was the main reason and the main driver for the prime focus, maybe it was a mistake, but probably it's too late now.

Derrick Salmon: Just one comment on this switch between F/8 and prime focus and the ``easy mode". It's obvious, it's very appealing, but its also going to take a fair bit of capital investment to make the change in there and I think that this is a medium term, not really long term plan for CFHT. I think you might want to weigh that against other options for CFHT in terms of improving image quality. For example, a lot of observatories now have big gates on the side of the dome and they retrofit their domes to do that. It probably costs the same amount to improve image quality as to add flexibility to the telescope. I think you really have to decide which way to go there.

CFHT has put in an awful lot of work in defining the problem and it would be really nice if we can actually benefit from that work and get the improvement in the building structure.

G. Walker: That's a major suggestion. That almost has to come from the corporation itself.

D. Crampton: Dennis put this out as an extreme proposal for CFHT operations in 2003, I actually don't think it's extreme, I think perhaps it's not extreme enough! I would urge this meeting to consider it... Rather than putting it into this sort of mid-term... Rather than putting a lot of money into new instruments or whatever, to actually do the things which are necessary to make a proposal - like this extreme proposal - work. In other words, to put in something so that you can switch from prime focus to F/8. Make the operation of CFHT unique in some way so that they can carry out a unique set of programs. And I think you can do that when you are able to run a queue schedule mode and take advantage of the very best seeing conditions and the very best infrared emissivity or whatever. I think that when you do that we can get our 4-meter into a very competitive position. And, I think, we should be putting the money in the sort of mid-term into that rather than different instruments or whatever.

G. Walker: I think I might support your point of view but Derrick has raised this question of the seeing, which is something we used to be very proud of. What the concern is being is that the seeing has been deteriorating. You want to elaborate a little bit Derrick on what you think about putting in the gates for instance to provide better air flow and so forth? Are there proposals which could come forward from you or other people?

D. Salmon: I don't think we have anything on the books right now but, I think, that it would probably be roughly on the same order of magnitude as what you'd need to do structurally on the telescope to put in a flip F/8.

Putting in a flip F/8 is perhaps possible, I can't say whether it is or not, but it certainly isn't going to be cheap. Whereas I know CTIO and other observatories have done it [put in ventilation on the sides of the dome] for amazingly little money, and Kitt Peak is doing the same thing.

The other thing perhaps that I should address a little bit here for just a few seconds is, and I hope it's a perception only, that seeing at CFHT is starting to degrade. I agree with that we are not putting in the great efforts that we did when René was there, but I obviously don't think we're back sliding. I think the community perhaps isn't aware of the efforts that have been going on in the last couple of years to keep up and improve the seeing. There are things like going through the building and changing heat sources from convective heaters to radiative heaters, changing the kind of lighting and control we had so that the lights don't get left on. There is primary mirror air cooling that is now in place every day, for the next night - that wasn't there. And there's really an ongoing but relatively quiet development in the telescope and I don't think we are slipping back. We're maybe not advancing as much as we'd like, but I think we are addressing problems slowly.

P. Couturier: I just want to continue on that point. We have made a mistake not to provide statistics of what we did with UH8k. The last run of UH8k was almost a continuous run of 0.5 to 0.6 arc-second for the seeing, for the whole field.

G. Walker: How much do I have to pay for it?

P. Couturier: No, but the point is that it's the only instrument with which we could make an evaluation of the seeing now. For all the other ones, we cannot do it.

D. Crabtree: I would just like to make one comment in response to Derrick's proposal about flushing the dome. The idea of moving from prime focus to F/8 and back and forth is to give flexibility to take advantage of the seeing conditions... Whereas this proposal, which I think we should also do, is to improve the seeing overall. But it wouldn't, if we got 1-arc-second seeing, the dome isn't going to improve it significantly to allow you to use AOB for instance.

J. Hutchings: I'd like to try to focus us back onto the wide field infrared red camera. This stands out in my mind as a major and expensive and unfunded new instrument. We need to decide now, or very soon anyway, whether we want to put our money into this thing, whether that's what we're going to be doing as the major new instrumental development. Because, as I understand it, we have to go into it soon to take advantage of being in the line of detectors and Pierre has to go beg for money with very strong backing from the scientists to do it. If we don't speak up now, it's not going to happen!

G. Walker: Well, I think we got that message earlier as well and that's why I want us to do this. Maybe, David: You can respond? Do we have a group and a crusader to that?

D. Crampton: I certainly had a lot of positive feedback from my presentation yesterday so I suggest we just simply take a vote on it!

G. Walker: You may not be used to Canadian politics... You can never take a vote until you know you're going to get there. Any other quick comments before we call the question?

T. Davidge: I too favor the wide field infrared mosaic. I'm just thinking out loud here ... I see the most important thing I could do may not be the broad band imaging, but perhaps narrow band or perhaps to have some really basic simple spectroscopic mode. Something that would give it a means for studying the spectral energy distribution of galaxies and stars over a large field. Or, if you can't do that, put in a wheel that would hold something 30 narrow band filters.

G. Walker: I think we're getting somewhat into detail there and the understanding is that the wavelength and area on the sky are real attractive elements at this moment. Does anybody else want to talk to the infrared wide field?

R. Bacon: Just to say that during the ``10-years prospect'' [meeting] of the French community, we expressed a good, strong interest in the infrared camera that was being proposed. Also there was no clear idea about the funding problem.

G. Walker: I don't think we have to worry about funding. That's the agencies problem; it's what they get paid for.

H. Richer: Before we vote, I'd really like to be clear on what we're voting on here. If we're voting for the wide field infrared mosaic does that mean we're voting against the cross-disperser echelle for example?

G. Walker: No, no, no, no.

H. Richer: Well, we're going to have limited resources, so we may not be able to do both for example.

G. Walker: This group is not concerned about funding, at this moment.

H. Richer: Than we should build a hundred meter telescope!

G. Walker: No, no, no. We're concerned about whether scientifically, for the mid-term, on CFHT, a wide field infrared mosaic, would put it in a very unique position because of where it is and what it can do and the coming on line of the narrow field 8-meter telescopes on Mauna Kea. That's what I interpret the feeling to be. And we probably realize there isn't enough money.

P. Felenbok: I have no personal knowledge in the wide field infrared imaging, but the NGST is the landscape of that too, so how do you compare all that?

J.-G. Cuby: Well, this is just a comment that the VLT will have a NIRMOS which will have basically four 2k chips like this proposal has and the field of view will be 14x14 arc-min. It is slightly different from the Canadian situation with the Gemini which will have a much smaller field of view. To me, this project is only interesting at this point for the French community, but also in the context of international competition, if it has the wide field and ``wide'' means at least 20x20 arc-mins. And, it should be clear that it's at the expense of the spatial sampling.

R. Racine: I'm going to put in my two words. I strongly support this for many reasons but I don't think we should support this before we have some assurance that the CFH telescope itself will be brought up to snuff. To be very specific about what I was saying before, I think we need to replace the primary mirror, the secondary mirror, the tube, to open the gates and the dome and make the CFH telescope a state of the art facility. Then you put an infrared array on it and then you get an one-eight to a quarter of an arc-second images over half a degree. And I think this is worth it. But we shouldn't do that before we fix the telescope.

G. Walker: I respect your opinion and I understand why you put it, but I think the proposal - as we have it - is with the existing CFHT. For this date of 2003, give or take, we're talking about the existing telescope.

P. Couturier: Just a comment on what has been said by René. I know presently a lot of people who would want to build wide field infrared camera for CFH in exchange of telescope time. They would go for it very fast. Changing the telescope is another major issue which would also need funding. But it's really another issue from my point of view. Wide field infrared camera has to be treated at this users' meeting, has to be decided or rejected at this users' meeting and SAC meeting.

G. Walker: I would like you to endorse or otherwise the concept of the wide field infrared mosaic camera as a mid-term initiative. But we're talking something of the order of 20x20 arc-mins. So those of you who are in favor raise your hands...

[Vote: A large fraction ($\sim$80%) of the audience raise their hands...]

G. Walker: Those of you who are against it?
[Vote: Very few hands ...]

G. Walker: Thank you. Now we can move on from that topic which is what I think is up to the SAC now to find the appropriate way to carry the banner and take it to the Board and the agencies to see where the funding comes from.

G. Walker: I'd like to discuss the coudé spectrograph. There's quite a large number of people here, including myself, who have considerable interest in maintaining the high resolution spectrograph capability on the telescope. And the fiber feed looks like a very viable way of using the telescope in a versatile mode.

So, I think when I look at this list and, we had the proposal already that ... the third item or instrument capability is the coudé spectrograph. And, I think, we can certainly look forward - if the high level of transmission which Paul suggested to us for fibers could be realized and we can demonstrate it - that we might well be able to lead light over from Gemini. I think their attitude at the moment has shown me that they're not going to go any further until we can demonstrate that we really could get sufficient light across. But I think that there is a wide potential there and that this does come under the heading of sharing facilities and so on. I'm not quite sure how it would work at the moment, but I just wondered if you want to say anything more about the cost of the cross dispersing echelle or an estimation of GECKO. Do you want to speak to that?

J. Hutchings: There was some discussion that got cut short yesterday as to the scientific value of measuring velocity to 1 m/s accuracy. I'd like to hear the end of that.

G. Walker: Well, I could probably summarize what I'm aware of. We've certainly known for a very long time that if you're going to measure radial velocities to the order of a meter per second, you then face the noise of the source itself, which is the star, which has magnetic cycle modulation and the whole series of other things, some of which cannot be calibrated. You're basically looking at a noisy source. And so I would suggest that a few meters per second is not an unrealistic figure but much below that is basically hopeless. Well not hopeless, but probably not going to provide very much new information.

I think which is the broader issue is whether CFH is worth considering for looking at radial velocities of nearby stars for the planet search, whether that's still a viable, long term program for CFH given all of the other activities, particularly with the other large telescopes.

I would be cautious about that one, but I think there's a broader issue of looking at the precise radial velocities; other things should be examined: surface circulation, turbulent velocities, rotational velocities, resolution of lines in late-type stars... the whole business of mapping surfaces in stars which we heard about yesterday carries over a wide spectral range. These studies have only really been scratched on the surface... On a long term, it takes a lot of work. So I personally would be a bit cautious about supporting or justifying high resolution spectrograph on the planet search, but I think that there are these other areas within the French and Canadian communities that we could basically be the leaders in this work.

Brett Gladman: I have seen claims that if you're looking for Jupiter mass planets around stars, you can use seasonal averages. Since you're looking for decade periods you can use seasonal averages to beat down the stellar noise.

G. Walker: That's right, providing you get enough observing time. This point is if you want to get a large enough sample; I'm not saying one wouldn't do it, but I'm not sure whether is it would be a driver for what we are dealing with at the moment?

P. Couturier: I would like to make a comment about prioritization of this kind of project. You have seen this morning my last transparency. There is more than just one project at this level of funding which are more or less in competition and prioritization by SAC. There is the medium resolution cross dispersing echelle. There is a cool Fabry-Perot in front of KIR. There is ESPADONS. There is a list of projects and I'm sure that we will not do all of them. So, probably, if we are discussing that, we would have also to discuss the others.

J. Hutchings: Well, it's what I was going to say. It would be helpful to get a sense of whether there is a preference among these things, or whether we need them all, how many people want what?

Jean-Francois Donati: One may think about how to really be the best, which spectrograph is going to be the unique one in the world. For very high resolution, many 8-meter telescopes are going to be equipped with very high resolution spectrographs and CFH is not going to be very competitive with a one-order high resolution spectrograph.

The AAT has also a high resolution spectrograph, a one-million-resolution spectrograph. So I'm not sure that once more CFHT will be competitive in that field, particularly. If we want to have a particular niche, [let us think about] either spectro-polarimetry - because that exists no where in the world - or, the full spectral domain in a single exposure - that doesn't exist very much in the world as well. Maybe we can decide high resolution of the full spectral domain, but then we need funding for that as well. So that's another decision.

T. Forveille: From this project, the project outside the wide field imaging, I see a difference between the fiber fed spectrograph and the other instruments that have been discussed, such as OSIS-IR. In that, I see a longer term future for a high resolution spectrograph. I think, all this IR has a niche, but that's basically 3-4 years niche which will be finishing by the year 2003. That's an instrument which is worth building but it's certainly not worth building if it's going to slide on schedule much beyond the official value [07/01/99] which is quoted now. That's my basic statement about that particular instrument. So, I think, high resolution spectroscopy has a future as a complement to the wide field imaging beyond 2003. That's the main difference that I make between these instruments.

G. Walker: Any other spectroscopists want to speak to this?

If I understood your comment correctly, were you suggesting that it was an issue between [resolutions of] 50,000 and 150,000, or it's not quite clear. The idea of using three or more 2k by 4k CCDs in the focal plane should be fairly straightforward.

There is an issue I think at the level of [resolutions of] 50k to 150k. I think some of us, certainly I, feel very strongly about the higher resolution level. If one's going to do precise radial velocities, it is well demonstrated that one has to be able to resolve the iodine lines otherwise the point spread function varies too much. And that's the level at which we should be aiming if we truly believe in this level of high precision. That's a personal opinion and that's what I would have gone for.

John Rice: I was just going to comment that I think we have a situation here and which is somewhat similar to the debate started over the scale on the wide field IR imager. We have to resolve this within our own spectrographic community. And, I think, we will do that. We started to talk about getting together on this issue a day ago. So I don't think we can resolve this right at this meeting, I think. We need to have a discussion among the spectroscopists and we'll put that forward to the SAC.

D. Crampton: Can I make a provocative statement? This sounds like the same discussion we had at SAC in Meudon. Quite a long time ago... [This was] ... when we decided to go with the present coudé rather than a cross dispersed echelle. And ever since I have been told many times that we made the wrong decision. I'm intentionally being provocative, Gordon.

G. Walker: That's fine. I'm not disagreeing with you.

P. Felenbok: I would like to ask a question. A little bit aside of that but it's still included in the fiber feed landscape, as I presented yesterday. The request was only to couple the Cassegrain to the GECKO. I showed you the possibility to do more than that and I explained to our colleagues at CFHT that whatever would not be decided at the beginning of the design, it will be extremely difficult to incorporate later. So, I would like to hear from you: Are you interested to save the ultraviolet capacity? And do you think that for the future it's a good idea to be able to flip from the prime focus to the GECKO? What is your message to SAC? And after, from SAC to the corporation?

G. Walker: If you're referring to me, I'm in favor of both of those, UV and flipping prime focus.

D. Crabtree: Yes, but if you can go from prime focus to F/8, we already would have the fiber feed at F/8, so you don't need a fiber feed from prime focus.

P. Couturier: I think Paul is somewhat right, you know we have to make a decision now. Because - clearly - the fiber feed will have to be included in the design of the present prime focus environment. If you are not making this decision now, we will be unable to come back later. And, it is presently part of the design. I want to confirm to Paul that it is part of the design presently. The problem to implement this project is something else. Making the design and implementing the project is different planning. I would want also to comment about this project which is [to cost] about $100K - 150K, or something like that... If CFH has an alive budget, that could be covered.

G. Walker: Okay. Don't say anymore....

P. Couturier: But, clearly, there will be an issue with the planning as I have shown this morning. We need to plan these things and not to put into parallel all these projects. So I'm not saying that CFH could make that just now.

G. Walker: I think in terms of the cross-dispersed echelle. There is the implication that we're dealing with something very expensive. I think, you presented a cost that was not very high for your proposal, is that right, Paul? I've forgotten the number now, $70,000?

P. Felenbok: In fact, we would not be building a spectrograph, we would be exchanging a grating in the spectrograph that exists and that has a large pupil which is 30 centimeters. There's no question [=need] to build now a spectrograph with a 30 cm pupil. The main cost is the grating and it seems not to be very expensive, it's $75,000, plus some ``side things''... some very small size optics which would cost about $15k. So it is around 90 kilo dollars.

G. Walker: So, we've just heard from the director that you can fund ...

[Laughter from the audience]

P. Felenbok: So, I don't want to go farther on this discussion.

J.F. Donati: But I did want to emphasize that it doesn't cover the whole spectral coverage.

G. Walker: I don't want to get into a loop on this, but I'd like to get closure, because - as David has pointed out - we're likely to start chasing our tail again. That will not necessarily lead to good decisions.

Claude Catala: I have just one comment. It seems to me that, and I can hear it from the discussions, that we need both: full spectral coverage and a high resolution. By high, I mean above 100,000. It seems to me also that we have heard about two projects that [would] actually [result in two] facilities at a very reasonable price. So I don't see why we could not have both. Both, a fiber link to GECKO with the extension of GECKO to cross dispersion that would give us something above 100,000 resolution with a significant spectral coverage even though it is not complete and, at the same time, we could have ESPADONS which would give us the full spectral coverage at 60,000 resolution. We actually made a list of all the fascinating scientific goals of both facilities and - given the cost that's represented - I don't see why we should cut ourselves and just choose between the two. Because really, they are very complimentary.

G. Walker: Okay. Any more discussion on this issue? All right, I think that's a very reasonable point of view, but it may not be the only one... We, spectroscopists, may be forced to make the decision in the last.

Is there any feeling that people would like to express their view about the level of resolution? It's the resolution versus spectral coverage [issue] as presented at the moment. I feel that's a slightly artificial position in my view. I don't think we have any designs in front of us... Maybe we could get full spectral coverage in something of in excess of 100,000 than it seems to me it would satisfy everybody. Is that a correct statement?

Or is that not a statement that we should carry forward and ask?

Okay. I think I'd ask the spectroscopists to at least make one more stab at getting this resolved before the end of this week.

D. Crampton: It seems to me that both are cheap, that capital costs are negligible, just forget about it. The real question though is the complexity of CFHT operations. Then it really comes down to the operational cost of this. Now Pierre told us earlier that if this is a visitor instrument, it doesn't cost very much at all. So it seems to me that's one of the solutions: Perhaps to make this a visitor instrument?

D. Crabtree: Just one comment on that, because - I think - one of the advantages to having it fiber-fed is the ability to switch to it from other instruments. It's hard to see how a visitor instrument is queue scheduled.

C. Catala: I would also like to mention that ESPADONS, which is a fixed spectrograph with a fixed configuration, will necessitate only very minor operations from the CFHT staff. Also when you think of flexibility and being able to switch from AOB observations to spectroscopic observations it's much easier to switch from an instrument which is always in the same configuration because then you don't have to go to the dome and set it up for a given particular program: It's there, it's already set up and this is almost zero operation.

G. Walker: All-right. We should try to move on from this and I think we've probably exhausted what we want to do on this list. [As to] questions dedicated to detectors; that's more technical consideration.

Let's go to something easy. We might get an answer right away. What about the AOB? What about the AOB splitters?

D. Salmon: Is this to do with the OASIS beam splitters?

D. Crampton: The problem in the infrared beam splitter is as follows: We have a beam splitter which sends visible light to the wavefront sensor and the infrared light goes on to the detectors. When we specified that beam splitter, we didn't put up proper specification on the AR coating, we just specified it in a visible range. As a result of that, in the infrared, the AR coating is not very good. In other words, about - I can't recall - maybe 10% of the light goes some place else.

Not only does it do that though, but it obviously must increase the background level in your infrared detector as its got to go somewhere. The adaptive optics is generally used in the infrared; you get the best results in the infrared. If you want to get the best out of the infrared, if you want to get the best out of AOB, $10,000 will be the costs for the beam splitter. I imagine if you could ever get it off the telescope for long enough and polish off the AR coating, it might be worth it.

J. Hutchings: Just another thing, mainly for general information. There's I think a bit of misconception about using the AO in the visible. People think: It doesn't work well... you loose half the signal... In fact, we have new beam splitters which are more efficient for visible signal and, as we saw yesterday, you can achieve 0.2 - 0.3 arc-second images in the visible. I think there's a great deal of time-value in doing that, so I just want to make sure people - including members of the TAC - realize this.

D. Salmon: Just one point, we're talking up to 10% level in coating beam splitters... We have a couple of beam splitters right now, but have no coating on them and we've had a hard time getting them. The glass is available, but I think that this is something both - time-wise and financially - we can cover just inside CFHT operations.

Gregory Barrick: Can I just make one comment? We have a couple of beam splitters [at CFHT] that actually nobody uses at this point, that we could strip the coatings off and put on a new coating; we have actually considered that in the past. So, I don't think it's really a problem. We could probably do it for a couple thousand dollars.

Observing modes

G. Walker: There's my suggestion of an initiative or at least something be done about it: an objective seeing monitor, possibly in collaboration with Gemini. This may already be happening, I'm not sure. I don't know if there's general support for this. I feel that it's something that may become needed at some stage - in connection with the service and other observing.

D. Salmon: Right now we can tell the seeing pretty well by looking at the TV monitors on the bonnettes.

G. Walker: Right, but that tells you what the seeing is, at the focus of the telescope; but what are the conditions outside? It tells you what the seeing is in the direction of the telescope is looking.., I think there's a broader issue here in relation to what are the conditions over the site...

D. Salmon: ... and how they change at different locations in the sky.

G. Walker: Anyway, I don't want to prolong the discussion on this. Is anybody against this idea? ... any initiative that you think the Board should consider?

T. Forveille: There are two issues here. One is that the seeing you get at the focus of the telescope - that you probably can get from the guiding camera. Another issue, which I think is important to the context of the seeing improvement, is the seeing outside of the dome. I would like to know which fraction of the actual seeing at the focus comes from the telescope itself and its enclosure and which fraction is intrinsic to the site. If most of the seeing is intrinsic to the site than there is no point in trying to improve it. If three quarters of it still comes from the telescope and the dome... I don't think we have this information at the moment.

Yvan Dutil: Is there any plan for a transparency monitor or an infrared brightness monitor?

D. Crabtree: What we're planning on with the TCS system will be a particular digital readout from the guiding camera. This will be a way of monitoring transparency changes during exposure by looking at 10-second averages of the guider counts and somehow getting that into the data. And then there's also a Mauna Kea wide facility, an infrared camera, which will be provided by the Subaru people, but - of course - it is another observatory. So, just for CFH - which I know I can speak about - we plan to have some capability of tracking the sky transparency through averaging guider counts over every 30-seconds every minute or something like that, and then including that with the FITS header of data.

G. Walker: Okay, now the 12k camera... And the way we use it. We heard quite a bit about the queue and service observing today. And adaptive scheduling ... Are there any comments about this? I have the feeling that this raised concerns, but it didn't necessarily raise hostility to the idea...

S. Lilly: I think a lot of my concerns would be in the details and the particular set of rules which were bounced... clearly would lead to problems that I think other people pointed out. But I think that we should focus on the concept and the detail rules can be dealt with in due course.

G. Walker: We are dealing here with a single instrument which has got fairly limited number of modes - so that [service observing] is not a big challenge in that respect. Does anybody want to speak to this? Is there a general sense of - a positive sense - that this should go ahead? ...

Gregory Fahlman: I think adaptive scheduling as Dennis [Crabtree] presented it has one element which I think is fairly difficult to adapt to. And that is that you can submit proposals at any time; I kind of wonder whether you will ever find people that would be willing to serve on TAC, if they knew they were going to be faced with proposals at any random time over the course of the year. I think that the virtues of the adaptive scheduling - as Dennis presented - could also be achieved within the semester system that we have, or perhaps a quarterly system, or some other. But I don't think that it would be particularly appealing from a practical point of view to be able to accept and evaluate proposals just at any time of the year. I think that's the only real problem that I see at the moment.

S. Lilly: I agree completely with Greg, but for slightly different reasons. I think one should distinguish between the adaptive scheduling aspect which is - as I mentioned [in the context of] JCMT, a very attractive idea - and the adaptive submission of proposals, which I think would raise all sorts of problems, including what Greg mentioned. In particular, I don't see how it's really possible to evaluate competitive proposals in an absolute sense when you have finite resources. If we had an infinite amount of telescope time then - sure - you can decide on an ad-hoc basis, but since it's always going to be competitive, I cannot see how you would manage that competition if the proposals were coming in drifts and drafts. It seems to me you'd have to have a formalized competitive process.

D. Crampton: Actually the adaptive scheduling, as proposed by Dennis, has an enormous advantage: It is well-known that everybody puts in a proposal the night before the proposal's deadline. So this means, there would be much less pressure on the telescope time!

T. Davidge: I just have one concern. I think adaptive scheduling is good in principle, but the scenario that Dennis presented this morning showed a lot of queue observing, so I suspect the adaptive scheduling would fall into the queue regime. The observing queues work best and most efficiently (eg. the case of WIYN) if proposals are in there for long enough to have a chance to percolate to the surface. So, if you're continually throwing proposals in and pulling other ones out, it could be that the queue scheduling will not reach the benefits that it is advertised to. So, the adaptive scheduling may not be a good idea for queue observing.

P. Couturier: I have a proposal which I think is a merging of different proposals we have until now. I would like to keep the ideas that Dennis has presented. There is an easy way to do it. It would have a six-months review by as many TAC's as you want. And after that the [CFHT] Executive will make adaptive scheduling. At the end of the 6-month period, TAC will review the results of the adaptive scheduling made by the Executive without any new input. And then they will review and they would say if they are happy or not with what has been done. But the way we are presently working is totally stupid!

G. Walker: I realize we can all take a light hearted view of this [matter of scheduling], but it's definitely fundamental for the way we run the telescope and a lot of us - when you are back at your home institution, or get rejected, or you are at 1 am before the deadline or whatever - we also probably agree with you about the need of a satisfactory solution ...

T. Forveille: I want to comment that what Pierre described is actually the way IRAM schedules proposals. I would say - on average - proposers are probably slightly happier about that than about CFH. They are not terribly happy about it because the pressure is significant enough that some proposals - majority of proposals - get rejected... And people complain about that... But that's life.

G. Walker: It sounds to me as if this is something which should be under review and clearly discussed further, both at the level of SAC and the Board. But we must say that we are getting a very reserved response from a lot of people.

Y. Dutil: A comment from someone who has recently finished his PhD. Adaptive scheduling may be be very helpful for students. Because, now if you have a 15-months delay, and that is if you get the first proposal right, you still have 15 months between your proposal and your observation. You need generally more than one observation... You know it does not work the first time; I had to try four times and got the observations only 6 months before the end of my PhD thesis. If you could have a way to shorten to 6 months in this, it would be very good for students.

G. Walker: I think that's an extremely important point actually, and I'm very sympathetic to that. Any more questions; otherwise we move on to the scale of the OSIS-IR.

The pixel scale of the OSIS-IR camera

D. Crabtree: I have a proposal, a suggestion, that we hand out slips of paper to everybody. Everybody writes down what plate scale they want. Then we plot it, and we take the mean or median, or something.

G. Walker: David [Crampton], would you want to repeat, the argument was survey coverage versus ...

D. Crampton: The OSIS spectrograph can use a 5 arc-min field. And its a tight scale, its 0.2 arc-sec per pixel, I guess, or 0.21 - whatever it is. This way, it turns out to be a 3-minute field so you're limited in a spectral dimension particularly. I guess my own personal view is that even in the direct mode, I would be happier with a larger field of view rather than the finer plate scale to properly sample the images. Mostly because there are other ways we can look at detail, with AOB or with Gemini, or anything else ... I myself prefer something like 0.3 arc-sec - but I will compromise somewhere.

Dennis Crabtree: I just want to remind people that the camera is meant to be used for direct imaging as well, and not just for a spectral use; although that was the priority given by SAC.

J. Hutchings: I just want to be clear that there isn't some technical problem with having the larger pixels, the sky brightness or some other thing that we'll be sorry about if we go now. Has this been sorted out? Is everybody quite clear and understands what we're doing?

J.-G. Cuby: There might be one point which deals with sampling of the OH lines and that was raised by René Doyon yesterday. If you observe at high resolution in spectroscopy, you mainly observe between the OH lines and maybe the OH lines will not be properly sampled and may not subtract very well out. But I think it's not an important issue because - at high resolution - what we want to do is to observe in between the lines. So personally I fully support the 0.3 arc-sec scale which provides both the larger field of view and the larger spectral coverage.

G. Walker: All right, we have a proposal for the 0.3 arc-sec pixel.

René Doyon: I'd like to see a proposal that will permit to get spectra across a galaxy at a resolution of 2000... That's very hard. I agree with David [Crampton] that you're going to have less spectral coverage... But you're not going to have less spectra, you're just going to have less spectral coverage. I mean, it all comes down to what exactly you want to do. I mean, [the camera] has been delayed long enough and right now we have a design which is ready to be built... There are constraints on the size of the filters you want to use in the camera ... and so on. Technically, to go from 0.2 to to 0.3, it is not much of a deal... The degradation goes as the square of the field - roughly... You may be just going to make all things bigger... But you know that this will take longer... Besides, there are financial constraints on the project and all the delays. So this is where you must strike a balance.

G. Walker: Do I take that as a NO vote? Who is in favor of the Point 3 [= an increase of the pixel size] amongst the infrared people?

D. Crampton: I'd say, why don't we ask whose in favor of Point 3 [increase of the pixel size] provided that it doesn't drive the technology? I don't want to delay this subject at all. And so, I mean I'm prepared to negotiate... It's ``Point 3'', provided that we don't identify any serious problems in the next two weeks, it seems to me.

G. Walker: I'm not sure anybody is saying that, but that's your proposal.

P. Couturier: I remember being a Board member, the decision of GECKO, when the Board was almost voting on the high dispersion grating. So I would want to make a decision as an Executive [Director] and to say: It will be 0.211 arc-second per pixel! Because we have received three successive resolution from SAC about that. And that's it! You know we are now starting design... I understand the argument of David, but at some point we have to be consistent with what had been discussed... It's very, very difficult now to change our mind. We have changed our minds three times on this OSIS-IR spectrograph.

D. Crabtree: I would like to add that [this matter] is a real incentive for us not to delay these projects... Because then, we don't give people a chance to change their minds!

J.-G. Cuby: Just a reply to your comment. Yes of course, I agree. But I want to point out that if we [wanted to] discuss Point 3, it is because the project is two years late. And this is the only reason why... So if you tell us that the design already exists and is almost ready, and you are absolutely ready to go tomorrow for it, then it's absolutely fine! [Applause]

Joint proposals

G. Walker: Let's move on to this question of joint proposals; we heard a bit about that just before lunch. Does anybody have opinions about how joint proposals should be handled? If I heard correctly, it seems to me that somebody or some people were suggesting reviving the international TAC (time allocation committee). I do seem to remember that we moved away from an international TAC because it didn't work. We've heard from Jacques Vallée that there were problems with international TAC's at other institutions, and I'm concerned of that it's not a solution ... That it may amplify problems.

Does anybody wish to speak to this? To provide some guidance - to the SAC - on this issue?

G. Fahlman: I think that I am a supporter of having separate TAC's, because I think that they simplify the management of the observing time at the telescope over the years. I think that the present situation is changing and that some new kind of response is absolutely necessary at this time. Particularly, the joint proposals. There was a mechanism in place in which the CFHT itself had a TAC and this TAC was made up of, I think, two Canadians, two French and one Hawaiian. The proposal deadlines are generally, sufficiently in advance so that the joint proposals could be identified and perhaps the members of the official CFHT TAC's could be asked to find a special way of dealing with the joint proposals. Such proposals would be identified as proposals which were sent to both agencies; it might not even be necessary to send them to both agencies, but there could be a category of joint proposals and these would have to be dealt with by the CFHT staff ... I'm not sure how they would do that... We probably need a new mechanism for dealing with those particular proposals. This whole area is sufficiently complex that it's probably not decidable by a group this large... It is going to require a smaller group to come up with some ideas that can be then discussed in a reasonable way.

P. Couturier: I think it's so an important an issue, that I would want to say what I was thinking of when I signed this letter of intention [concerning the MEGACAM usage]. It's really something which has to be coordinated with the agencies. I was really thinking that because the contract with the CEA people would take place in November or December. That the next SAC meeting will have something in writing and will have probably a chance to discuss that by e-mail before; and these [decisions must] have the support and the agreement of the agency. Because the agencies are the owners of the telescope time - that's where the discussion has to take place. It's not to say that we are not interested of the point of view of the users. But it's so difficult an issue to solve and to fix, [even] in a very stable environment, that I think the agencies have to make a point of that. We have to be consistent in the way we are making investments like MEGACAM. And this consistency means that we have to have a policy for time assignment or management regulations on the use of telescope time, prep-time - and all of these kinds of things.

G. Walker: So, you are suggesting that this is an agency issue?

S. Lilly: I do think, it is important - as I mentioned this morning - to distinguish, though they're connected issues - to distinguish between the joint proposal issue, which could be for any science instrument at all, and the large-scale survey program issue which might well be a joint program. They are distinct. I think - they both are problems and they both need solutions. What I'd say on the joint proposal issue is: If we have another semester where six out of seven joint proposals are awarded time out of phase, [then] - you can forget it - for any future [such] type activities. It's just a disaster for people planning collaborative programs.

Howard Yee: I was going to say that the joint proposals and MEGACAM should be separated. They're not even the same issue. And, secondly, I am also a little bit worried how the tail is wagging the dog. In a sense of saying that: If - because the agency has invested a lot of money in an instrument - that instrument should drive the policy of how scientists operate... I'm not entirely agreeing with that.

G. Walker: What do you want? What would you propose?

H. Yee: I'm proposing that the co-agencies should not tell, or put down, the law on how the scientists should operate. That is what I'm proposing.

G. Walker: Do we want to go on with this issue or is this an issue that the SAC should handle?

Pierre Bastien: A quick comment: [a proposal] that we share the referee reports... The two or three reports for a given proposal... So that they have the same information.

Alain Mazure: [agrees with Simon Lilly]

G. Walker: Let's move on and the suggestions of some sort of intermediary ... and a better exchange of information ... should be studied. Now - MEGACAM. This is probably an issue on which everybody has an opinion and it won't be the same. This is obviously a major investment: time, effort, money and fairly long-term. And now this question of the time assignment: how things will be managed and regulated. Do you have any comments?

G. Fahlman: I have been a member of the Board for the last three years and so I have a slightly different point of view from what I had before being on the Board. And I understand what Pierre [Couturier] was saying: That perhaps not everybody appreciates that the agencies do regard ownership of the time allocated on the telescopes as perhaps the most precious resource that they do have connected with the telescope. And that, right now - as I'm sure we're all aware of - great pains [are] taken to assign the time in proportion to the budgetary contributions made by the different agencies. This is absolute and it's checked and rigorously adhered to each and every semester. When we get into instruments like MEGACAM and into large joint programs - certainly programs which will involve different elements within the two communities, or the three communities - then the idea of assigning nights to different countries becomes itself an issue and I would say that it's very much a question of whether, as a user community, we want to continue to adhere to this strict division, strict allocation of nights. I think this community will have to give guidance and an opinion as to how they want the time counted for these large joint projects. So I think this really is an issue and I think it's the issue that Pierre brought up just a few minutes ago.

G. Walker: Is there a sense in which we want to guarantee time for, at least initially, for MEGACAM?

P. Couturier: The word ``guarantees'' is always dangerous. Clearly, we need scientific control in the use of telescope time even when it is a consortium ... and probably even more when it's a consortium. So we have to define new ways of review, scientific review, peer review for this consortium survey time. And there will exist target-oriented proposals which would be selected the same way it has been in the past by TAC's. What I would want to see for large joint proposals ... would be really a kind of achievement [goal] of the joint proposals for the CFH12k to [lead to formation of] a large consortium of people working along the guidelines which have been shown this morning by Laurent Vigroux. This has to be prepared in advance as a policy. And I cannot see this policy defined without a strong commitment from the agencies because - at the end - the agencies are the owners of the telescope time.

L. Vigroux: I just want to make a comment about the time scale to make decisions. You might think that we are not in a hurry because this has to be in place by 2001. That's not the case. This is because the definition of operations has a very strong impact on the pipeline design. We want to have the scheme of the operation - not frozen, but a guideline on the way we will operate this instrument - quite soon to define the pipeline. So we will have a review by November, the critical design review, of all the different items of the camera, and also of the MegaPrime, I think, it would be wise to have at the same time also a first draft of the final agreement that would be signed by the different supporting agencies.

G. Walker: Any further comments? That has to start tomorrow, okay...

[laughter from the audience].

All-right, unless there are other comments this is what SAC will hear!

AOB and KIR

G. Walker: We talked about the wide-field infrared camera. We heard yesterday from Jean-Luc [Beuzit] that there would be a proposal - if it hasn't actually appeared yet - of the coronographic mode of KIR, that many of us have almost a fanatical desire to see happen. The [excellent quality of] KIR optics, coupled with the intrinsically low scatter PSF of CFHT give us a clear advantage for the immediate future... So I propose, as an impartial chairman, that we endorse a coronographic mode to KIR... [laughter from the audience]

H. Richer: I would certainly support that. I'd just like to ask Gordon or whoever is proposing this ... There is a range of hole sizes in such an operation; is that correct?

G. Walker: Yeah... I don't think we have a hole... But...

H. Richer: And the design will obviously not preclude the light from the object driving the wavefront sensor?

J.-P. Maillard: This is a question more toward the AOB rather than as KIR as the detector for AOB... I remind you that there is another AOB on the CFH which is the one from UH, which has actually better response than PUEO. So I would like to ask a question, is there some plan to upgrade PUEO to the level obtained with the UH AO system?

G. Walker: Not that I know of ...

Jean-Luc Beuzit: We have not yet studied this possibility with Roddier's team. It's possible to foresee [an upgrade] for the future, but we have first to look into the design and see if it's possible at all or not.

G. Walker: What would you judge the financial impact, or impacts in any way, of initiating and completing coronographic mode?

J.-L. Beuzit: KIR has been designed with the coronographic mode in mind, so it's possible [to implement such a mode] without modifying completely the design of KIR. Basically, it would [require] adding two wheels and a few optics [items] inside. We didn't look at the budget exactly, but it would probably be [about] $20,000 or maybe [only] $15,000, or something like that. And the only impact on KIR would be that we would need to remove it from the telescope for about 1 - 2 months to integrate and test [the new mode], before putting it back in operation.

G. Walker: There is the cold Fabry-Perot to go inside KIR. Somebody wants to speak to that?

J.-L. Beuzit: I didn't have time to elaborate a lot on that yesterday when I did the presentation on KIR, but I point to a poster from Daniel Rouan. There were in fact two groups proposing the same [concept which] is doing integral spectroscopy with adaptive optics. The first one was Daniel Rouan and some other people proposing to use a cold grism inside KIR, in the pupil plane, [that is] using the [current] filter wheel, and putting a low resolution Fabry-Perot in front of KIR - outside - and slits in the focal plane, inside KIR also, and doing spectroscopy with resolution of about 1200. That would be a low cost system and could be coupled with the coronographic mode easily. In fact, the coronograph would need a wheel for putting masks in the focal plane. Daniel did some calculations and showed that it was possible to mount grisms inside the filter wheel. So that would be the first step. There was also the proposal - not a formal proposal yet, but a discussion with the Grenoble team of 2D-GraF, an ADONIS instrument at ESO, for doing spectroscopy at resolutions of 5,000 to 25,000 - 30,000, by adding a spectrograph outside of KIR, actually a Fabry-Perot with a cross-disperser.

G. Walker: So what is the proposal?

J.-L. Beuzit: The first proposal by Daniel Rouan is to start with the low-resolution Fabry-Perot in front of KIR, a grism in the filter wheel and a slit [in the focal plane].

G. Walker: And what's the impact as far as SAC or this group is concerned? Are you asking for funding? Manpower?

J.-L. Beuzit: I don't know because Daniel is not here [today] and we did not have a lot of time for discussing that [as[ect]. I think that Daniel was looking into a source of funding [in France] for the grism and the slits... What Daniel is asking [from CFHT] is to have a wheel in the focal plane to be able to install that slit and to have access to the CFHT Fabry-Perot controller. That's the idea at the moment.

D. Crampton: With the coronographic mode for KIR: Obviously the intention is to look for faint companions, planets, or whatever. Are you competitive with coronographic mode on the Hawaii 36 element AO system? Because, they've got a better, higher Strehl and they also use the F/35 upper end - you don't have as much light in the diffraction ring... So this is a question: If you really intend on going to the limit, aren't you better to do it with the UH systems?

P. Couturier: We don't have the technical document describing the results, the final results of François [Roddier], but - clearly - the visible will be more competitive ...

G. Walker: I should stress, the key trick in our technique is to do dual imaging. Anyone can put a disk in the focal plane. But without the dual imaging, there would not be the speckle pattern - which is the limitation.... With theirs or anybody else's system. Our experience is that we have - at least in part - made a significant improvement over [speckles?] by taking simultaneous images...

T. Davidge: Are there plans to use appodized mask... Maybe this was mentioned and I missed it?

J.-L. Beuzit: Yes, there could be also an appodized mask.

T. Davidge: Okay. And the UH coronograph, is it an appodized mask or is that just a piece of metal? ... I guess, you could replace it ...

J.-L. Beuzit: I didn't see the [detailed] design of François' [Roddier] coronograph yet. He's planning to test his own idea, the phase mask, but I don't think that this has been implemented yet.

G. Walker: I think we can talk about this later, but now that would probably get us into details. Put it this way, the mask is a damn nuisance... There are all kinds of uncertainties here...

Okay, there's a lonesome item at the bottom here called ``archives''. I didn't know there was a problem here... Is there anything we're supposed to address?

CFHT archive

David Bohlender: I just wanted to gauge the interest in the CFHT archive. Daniel [Durand] gave a talk yesterday describing the current status, but I don't think he got across the frustration we sometimes feel within the CADC in our attempts to make the archive usable or more user friendly. We were encouraged by the CFHT12k... But I was a little disappointed when Dennis [Crabtree] said that, two years after it is on the telescope, it would be archivable - I'm hoping we can do better than that! But what is the interest in the archive and making it more useful? There's some minor things... What we see as minor things that have been problems for... - well - since I started as a resident astronomer [at CFHT] in 1992 that are still there: run ID's, and other things. We'd love to see LAMA-file archives so that we could have some hope of identifying objects on MOS frames, ... And perhaps provide some sort of quick look facilities. So I'd like to just gauge the level of support in the archive or are we better off spending our time doing other things.

G. Walker: Did I hear a number yesterday, of about once a week, that people call [CADC] ...

D. Bohlender: I'd say on average about once a week ... maybe twice a week ... we have to put some files on line. The most recent survey that we did, about 15% of the users have made use. That is, 15% of the CADC registered users have made use of the CFHT archives, that is about 120 people.

J. Hutchings: I commented on that yesterday and I'll say it again today: The archives are really much more valuable when we, when you, can search them and use them properly. So getting how useful it is by how much it gets used now is not going to achieve much... I intend to have the SAC push this thing through and recommend what is being done to the CFH12k, and to get it on line as fast as we can.

G. Walker: At the moment, the people coming into the archive are from both countries or is it mostly from mainland?

D. Bohlender: It's just registered users... So we can't tell you whether it's Canadian or French or elsewhere.

G. Walker: It sounds as if you want to encourage people to use the archives ... And I guess it's a matter of education.

Are there other issues which you would like to have discussed?

Multi-ARGUS

Jean-Pierre Picat: Yesterday you had a proposal to use the ``multi-ARGUS'' spectroscopy and I would like to have the feeling of the community on this proposal... This is an existing instrument ... [that] we ask to be modified a little. We think that we would like to use this instrument as a visitor instrument on CFHT. We think that it would be a small load to the CFHT staff. I know also that most of the programs one can think of on this instrument will be better made on a 8-10 meter telescope. But we have two to three years before having such an instrument on a 8-10 meter telescope... So we can have some information how to deal with these kind of data... And also the advantage of already having 2-D spectroscopy [to do] measurements of rotation curves and so on. So I would like to have the feeling of the community on that proposal. I don't know if its a SAC issue now? I don't think so...

G. Walker: Well, remind me how many channels ...

J.-P. Picat: How many? Ten. And each multi-ARGUS will be a field of about 6 arc-sec.

G. Walker: Well, we certainly heard people talk about this at this meeting.

D. Crampton: I think [the situation] is a bit like with MOS/OSIS-IR. People should come forward and say whether they are really interested in using it. I think this is the same case... You should ask for expressions of interest from other people. And, I mean ... in a realistic case. One can always say: ``Yes, I am interested in it, I think this is really good idea.'' But, will I use it in the next year? I would say likely: NOT. But that's the kind of decision you have to make - I think - but maybe not here in real time.

G. Walker: Any expressions of strong interest in this? By not expressing strong interest in this does not mean you're not interested ... Anybody who wants to express support or otherwise on this one?

[silence] Okay, its a gallant shrug at this moment.

S. Lilly: I would be exactly at David's level, I think it's an interesting capability. I think there many things on MOS to do [what] one's going to do on an 8-10meter. Whether I would propose to do it in the next year - probably not.

G. Walker: Any other issues that people would like raised which are appropriate for the CFHT users?

De-commissioning of instruments and foci

J.-G. Cuby: Just an ``easy'' one. What about the de-commissioning? De-commissioning of instruments and a focus, as Pierre has presented this morning...

G. Walker: The proposal, as I heard it, there should be an annual or bi-annual review of scheduling of these things so that it can be done on an orderly fashion, so that the scientific community knows what time scale is for each of these. Some criticism has been in the past when it was discovered that an instrument is being decommissioned without having been properly discussed. But I'm not sure what you would like to do... To repeat your proposition, or have I just done so?

P. Couturier: You know... I dislike to be negative... To make negative statements when there is no scientific support for the kind of statement I'm making... And the provocative presentation I made this morning was not the first one... I know that there are some people who are complaining about the fact that I'm presenting that like... just like a bureaucratic decision. It is not a bureaucratic decision. It is a decision which has to be made because even if science is good, at some point we have to decide to place a new instrument by decommissioning an older instrument. In order to do that, I agree that the provocative approach is not the best one.

We have to go through SAC appraisal to make a decision about the instrument and the focus which will be decommissioned. I've made a suggestion ... because I want to be transparently clear about that... As I'm analyzing the statistics of what has been done. But you could say that it's ``just statistics''... So I'm asking the SAC to make the recommendation. We need to have the recommendation to prepare [ourselves] for [the year] 2000, semester two (2000II). Because after that, it will be 2001 and 2001... And there will be a new plan, and I don't know about the constraints that the agencies will put finally... I suspect that they will not propose to increase the budget of CFHT... With what I've shown this morning, I hope that you are convinced that there is a problem: to operate all these instruments and to keep on development [of new instruments], to make preventive maintenance in advance of a major failure... All that takes time, manpower and all that... So we have to make a decision about what we are decommissioning! I've made a proposal, it's probably a wrong proposal... I'm prepared to change my mind, but I want a proposal in November!

G. Walker: Okay, what do you say to that?

J.-G. Cuby: I mean, this is clear! If it's going to be addressed by SAC in October, I think it's an ideal place here to discuss it. I would like to hear opinions from the community.

G. Walker: Well we're not being overwhelmed by them.

[voices from the audience]

Well, you've got full support from the Advisory Committee on that: The Advisory Committee recognizes the corporation cannot continue to support all instruments beyond a certain amount of time and effort...

Closing

T. Abbott: I'd just like to remind everyone that the deadline for bulletins is the middle of next month. And I have no offers of any articles yet. And also, I've had one comment on the online-only version of the bulletin... If anyone else has any comment of whether or not they like it, please let me know.

G. Walker: I like it. But then I'm in a similar position to you.

I realize that we move into a joint session tomorrow but I'd like to take this occasion to thank Gilles Joncas for the organization - I think, this was excellent... Just the right place. I'd also have to say I've learned a great deal at this meeting, it was extremely interesting. And it's going to be a very interesting time at the CFHT in the next few years... So unless somebody has anything else to say, I close...

[Clapping from the audience]


next up previous
Up: Home Page
Pierre Martin
10/28/1998