Deepsearch Collaboration Meeting

1999 June 11, 9:00 AM or something


Saul says that we may not have to do this all day. People have been meeting informally all week. Mostly we need to hear from Chris Lidman, Alex Kim, and Carl, who weren't here last week.

We start with brief introductions so that Chris and Alex can meet everybody.

Chris Lidman going to tell us, with two transparencies, what's going on in the southern hemisphere. He shows a timeline with the facilities that are going to come into operation at ESO in Parenal [spelling?], and one in La Silla. There are four "UT"s coming, which stands for Unit Telescope, each is an 8m telescope. There are various instruments for each telescope. UT1, Antu, starts with FORS1 (an imaging spectrograph), which started running this april. It's a single slit imaging spectrograph, analogous to LRIS. (It also has controllable slitlets (19) on the focal plane, 18" long, with a tuneable width (slit jaws).) This is the instrument one would use for spectroscopy and followup. Chris doesn't think it will be as efficient in the red as LRIS. (FORS2, on UT2, will be using a slit mask instead of slitlets.)

ISAAC is a infrared (1-5micron) detector, imaging spectrograph I think. It has two arms, a Hawaii MgCdTl short wavelength detector (1-2.5micron), and an InSb array for longer wavelengths. Direct imaging, and spectroscopic resolutions up to R=10000. No AO on this instrument. This is the instrument that Chris runs. He claims you get down to seeing of 0.25" on ISAAC. Field size is 2.5'

CONICA is a newer upcoming IR imaging spectrograph with a finer pixel scale (.05", for a 50" field). It will use NAOS, the AO system to come online in 2000 or thereabouts. Chris thinks that CONICA, not ISAAC, would be the instrument to use to do J-band photometry of very high redshift supernovae. We might conceivably be able to use this by the end of year 2000, Chris says. (Chris is saying point sources down to magnitude 25 in K should be possible; ISAAC is 2 magnitudes worse.)

UT1 is the telescope where the laser guidestar will be available. That will be ready by the end of 2001. Chris is guessing that you'd probably need a 16th magnitude star within an arcminute, but he's pulling that off of the top of his head. (Chris points out that even though the seeing in Parenal is already good, UT1 already does better because of the tip-tilt secondary.)

SINFONI is an integrated field infrared spectrograph, to come online sometime in 2001 or 2002, for UT1. At that point UT1 will be entirely IR.

UT2, Kueyen, saw first light in March, and the first instruments are arriving now. The first one is UVES (Ultraviolet/Visual Echelle Spectrograph), for 3500 Angstroms to 1 micron. (Ultraviolet is a little bit of an exaggeration.) It works in double-beam mode, with a blue CCD and a red CCD; Chris isn't sure, but thinks, you will be able to observe the entire wavelength range at once. Saul wonders if we could use the high resolution to cut out sky lines. Chris doubts that the throughput will be as high as in FORS, as this is a cross-dispersed Echelle.

FORS2 is like FORS1, only with a punched slitmask. In 2001 will arrive FLAMES, which is a counterpart to the multifiber thingy on the AAT (2DF or whatever it is). It will have 400 fibres and will feed two instruments, GIRAFFE (optical detector), and FIRMOS (infrared detector). FLAMES will also feed UVES.

In 2000, UT3, Melipal, will come into operation. The mirror has already arrived. The first instrument will be VIMOS, a wide field (for an 8m) multiobject spectrograph. It's a little like LDSS on AAO, or (some instrument) on CFHT. 4 fields, each of 7'x7', and 4 separate cameras. Each is an imager, plus a multiobject spectrograph. It is essentially for galaxy surveys doing several hundred galaxies at once. This will be the instruments with the largest field of view on the VLT. If we want to do a search with an 8m class telescope, this may be an instrument to think about. Saul predicts half a dozen Albinonis a night.

VIZIR, 8-27micron camera, on UT3. (N and Q bands.) Our things are probably too dim for this. There is also a visitor focus on UT3, where people can bring their own instrument.

Saul would wonder if there would be any interest in putting some LBL chips in VIMOS. Chris says that some of the French are working on this one.

In 2001, UT4 (Yepun) comes into operation. All the mechanical work is in place on Parenal (however its spelled); UT4 just needs the mirror. FORS1 will be moved from UT1. NIMOS is the near infrared equivalent of VIMOS. It will have the same field of view, with four 2kx2k chips (which are under development somewhere, we guess). This will only be cooled with lN2, and will only go out to 1.8 microns (i.e. only J and H). The last instrument in 2002 will be CRIRES, an IR Echelle spectrograph.

Don questions modes that use multiple telescopes simultaneously. Chris responds that that is coming even later; perhaps sometime in 2000 UT1 and UT2 will be combined into a combined focus, but combining all of them won't happen until 2003 and 2004. The configuration is an L.

On La Silla, there are lots of telescopes there already. Chris notes that the La Sillaa site is being stepped down; in 4 years, ESO is planning only to operate the 3.6m and the NTT. Chris notes that there may be some opportunity.

The NTT is a 3.6m alt/az telescope. IT's a wonderful telescope, but not of much interest to us. SOFI (Son of ISAAC) is a 1-2.5 imaging spectrograph. It has a 5' field of view with good image quality and efficiency. Spectroscopically, it's faster, Chris claims, than the equivalent IR spectrograph currently on Keck. (Efficient grisms; 60-70%.) EMMI is an optical thing that does everything, and thus doesn't do anything that well. SUSI2 is an optical imager.

The 3.6m is where Chris has done most SCP observations. CASPEC, an Echelle spectrograph, has been decomissioned, as has TIMMI, a 5-20 micron camera. ADONIS is an AO IR camera. CES is an Echelle. EFOSCII is the imaging spectrograph (5' field of view) that Chris has used for the SCP observations. It's very good in the U and B bands, Chris says, and notes that it might be a good 'scope for followup of nearby supernovae. Saul and Chris wonder about trying to become a key project on this one. This would mean collaborating probably with the Italians, who are the European pundits of supernovae. Their primary interest is SNe II; Chris thinks maybe we could do a collaboration where we search, they chase the IIs, we chase the Ias

The 2.2m has the WFI, a 0.5 degree by 0.5 degree imager with 8 2kx4k chips with excellent image quality all through the field. The grav lensing people really like this telescope. It's the second best telescope in terms of image quality; it's not unreasonable to expect 0.5" seeing over the full field of view. The Italian group, Chris says, has a proposal in to look for SNe around z=0.3. It has fast readout- 20sec?- so you can potentially get huge data rates. WFI is only seen to last a couple of years, since on Parenal there will be a dedicated wide field imaging telescope, thereby making this one obselete. Chris notes that since there are no plans for this telescope after 2002, this might be a good one to think about putting an instrument on.

(Aside: La Silla median seeing 0.8", on Parenal, 0.6".)

Smaller telescopes; 1.54m is a joint ESO/Danish telescope. It has DFOSC, an imaging spectrograph. Chris doesn't think this will interest us; it's too small to do spectroscopy. (Carl notes that the first cosmological SN was found with this 'scope.)

The 1.52m has two spectrographs, B&C and FEROS. It's also too small to interest us, Chris says.

Over the last year, four telescopes have been closed: Schmidt, CAT (Cassegrain Auxiliary telescope), 90cm (not very good image quality, principally used for aperture photometry), and the 50cm. Sometime next year, the 1m telescope will become free. Currently, the DENIS survey is using it (IJK southern sky survey, equivalent to 2MASS). This one is better than the 90cm in terms of image quality, and thinks that this might be one that would be worth thinking about. Saul fantisizes about putting an instrument here and applying for a joint key project between this telescope and the 3.6m.

There ya go.

Chris notes that the conditions are very variable at Parenal. It depends on where the wind is coming from. He thinks that this will drive it more towards service observing, so that one can optimize the use of the telescope. He points out that there are a lot of nights that have light cirrus but excellent seeing; programs which specify that they require photometric conditions won't be done with this.

Carl Japan Satellite

Carl was in Japan, and waxes eloquent about Japanese hospitality.

He has a few transparencies on HOU, "Hands on Universe." HOU has developed a curriculum, image processing software, and a web site, for teacher support. They've used SCP BTC search images to look for asteroids. They just got 2.5 million dollars to scale this in a massive way. The goal of HOU is to enable students to do as much research and get as much joy out of astronomy as possible. The $2.5M is for figuring out how to train teachers over the web and figure out what works.

So far, about 1000 teachers have gone through some sort of HOU trainig. They have a subset of IDL-based software (Carl says that they have total distribution rights to the IDL subset they've done.)

They have a network of 30-inch class telescopes.

They have a paper from 1996 that had a couple of students from Oil City High School.

Carl says he's become convinced that they need the best images for this project, not throaway images. They want to get and create as many telescopes for themselves as they can, and not operate in parasitic mode. They also need to get scientists, time, data, collaborative skills, etc.

Re: internationally, Ariel's university was the first big push; about 165 teachers in Sweeden have gone through this. There are also teachers in other sundry countries.

Carl shows an overhead of the small telescope network, where they have a fraction of the time on, including the Berkeley automated 30inch telescope which is supposed to go to Chew's Ridge in -2 years or so.

There is a debate over whether it is economically reasonable to try to reclaim decomissioned 1m type telescopes... it may be cheaper and more effective to just build a whole new one rather than try to get the old ones up to spec and automated.

Carl shows another overhead of telescopes which are supposed to be coming online within the next year.

He shows another overhead with other plans. They want to have a wide-field imager on some 1m class telescope somewhere. (He has the text, "better than EROS??") Carl says that Peter Nugent is schmoozing Ken Nomoto about getting time on SuPrime Cam on Subaru (Mauna Kea 8m), which is some huge 8kx8k array or something where they get as good as 0.2" seeing. Then there's also this space station thing, which Ken Nomoto is involved in.

There's something about linking with the Air Force and Navajo nation and historically black colleges and getting telecopes and something or another.

He goes on to talking about the Space Station, which itself is a 70 billion dollar thing, which is probably the largest engineering project since the Pyramids. 2004 is the target for the thing to be completed. Carl says that they are looking at the Japanese Exposure Module, which consistes of a pressurized by and a thing where you hang instruments off it. The history of the space station is that they don't want to do astrophysics, but the Japanese managed to get around that and put EFU (Exposure Facility Unit), six of them, on the JEM, or something like that.

SHOUT, the Space Hands-On-Universe Telescope, is supposed to be a 1.x meter telescope to go on EFU 9, on the end of things. This is supposed to be some sort of prototype for some bigass Japanese space telescope (space-based Subaru), which may use JEM/ISS as a space factory. The lead collaborators are T. Ebisuzakie (RIKEN) and Y. Takahashi (somewhere associated vaglue with Marshall SFC). I. Mikami (Melco) was the program manager for ground-based Subaru, and wants to be the program manager for space-based Subaru.

There are some problems with the ISS as a site of the telelscope. Vibration and Contamination are a problem. Apparently, the Japanese want to fling it off after a while. (One contamination problem is that the shuttle, when around, once a day shoots out a huge plume of water.)

Etc. Insert claims about efficiencies of a 1m telescope being as good as HST+WFPC2.

Other possibilities with the Japanese; Saul says they've been talking to somebody else who's worked on the SuPrime camera. He is very interested in coming out to be with us on the next very high redshift run, and wants to send a student here to learn how our system works. (Are we going to be searching in September/October? I thought that was the French. Well, OK, I will admit that we haven't figured all that out yet.)

Saul and Carl debate what the Japanese are most interested in; science or engineering? With Subaru, does science drive it, or is it more the goal of building the best telescope? Saul thinks that they want to build the telescope for the sake of an engineering marvel, and then give away all of the telescope time on it. Carl disagrees, and think that the Japanese astronomers do have lots and lots of plans for Subaru.

Saul's SNAP Update

Saul says that the directors of LBNL are more interested in SNAP than we are.... They see NERSC as a big data center, the science as a LBL project, and they never got credit for Keck. It is a natural LBL thing to try to push for, and now the questions are where are the right political angles to push, etc. Saul thinks that we probably should try to quarantine our group from having to do too much on this over the next year, and see if we can get somebody from the lab to work on it. (This would be helpful, because we wouldn't be able to get all the rest of our stuff done.)

Carl thinks it's important to making it clear that the satellite fills a niche that HST and NGST doesn't come close to filling.

Alex Kim: the insider word from France

What do you call a young Girl Scout in a hurry? Brownian motion.

Most of what Alex is going to say is somewhat unofficial. There are three supernova searches in france. The first is a nearby search using a 55cm telescope, at the Pead de Midi (spelling?). Then there's EROS, and then there's FROGS. Alex was hired not to work in particular on supernova, but as a slave, at College de France for Yanik (spelling?). He was supposed to be working on the T55 and EROS collaborations. (Alex's boss, however, isn't interested in doing supernova anymore, so Alex hasn't been doing much on supernova at all recently.) The T55 search doesn't exist, Alex says, mostly due to hardware problems. (They haven't been able to get the telescope running.)

The EROS collaboration is a microlensing experiment that has their own 1m telelscope down at La Silla. For a while, it was the largest CCD array out there. You observe 1 square degree in two filters (B and R) simultaneously. (The B is more like a broad V, and the R is more like a broad I.) For each filter there are 8 CCDs, though not all of them are necessarily working at a given moment. Despite the fact that it's on La Silla, the average seeing is about 2". (Thought mostly to be dome seeing.) Another problem is the readout time, which isn't very good for a nearby search. The recent nearby search had 5m exposures, and they found a supernova at z=0.3. The mean was near around 0.13 or 0.15, which is a little further than what we wanted. It's hard to go any closer by because of the readout problem. There is thougt about trying to get around this with a single exposure, doing asteroid and cosmic ray rejection some other way.

So far, EROS has only used the B images for the search. You could use them for CR rejection, and also for more S/N. Another big problem with the nearby search is that it doesn't have a very good turnaround time; it takes several days between when the image is taken and when you decide there's a supernova there. Most of that is due to the software (Delphin's), Alex says.

The EROS collaboration is fairly large, mostly microlensing. About 10% of the time and 10% of the people are involved in the supernova search. There are about six permanent people spending a fraction of their time on the project; only one works fulltime on it (he's new). This does reflect that EROS isn't really dedicated to supernova. There are two graduate students, one of whom is working on the lightcurves (and is supposed to be done in 6-8 months, and will be using the data from the recent nearby run to see what he can do). The other is presumably going to be working on the rates of the EROS search.

Alex says that another problem with the software is that the efficiency is very human dependent.

Ariel wonders if the EROS group could use the FROGS software. Alex makes an aside on the FROGS. Alex says that EROS and the FROGS don't necessarily get along. Alex is not a member of the FROGS, for political reasons (even though Alex came up with the name).

For the future of EROS, they want to keep trying to do what they did in this last run. Saul wonders if the success of this last run will encourage them to keep going. Alex says they were very enthusiastic, and that Spiro in particular is interested in keeping going.

Re: microlensing, the view in Paris is that all of the MACHO objects seen so far were actually in the LMC, and not the halo, and as such the view is that the importance of microlensing searches is going down. The view of Spiro is that (something) (the Marley?) should be dedicated to supernovae in a couple of years. (Saul says that there is another microlensing group that is still more optimistic, and is trying to push do bigger experiments.)

Alex personally is a little bit skeptical as to whether the Marley is the ideal telescope to do a nearby search with. For instance, from his discussions with Spiro, putting in standard filters is impossible. (There is no filter wheel, and the filters are hardwired into the dichroic.) The seeing is irritating, but Greg points out that that doesn't matter because you already go too deep... and then there's the 2.5 minute overhead between fields. That includes 1.5 minutes of readout, and then a minute to lock on to a new guidestar.

Alex says that there has been no real discussion to organization if the whatchmadugit telescope goes over to almost fully supernova searching.

FROGS has quite a few people in them. Reynald and/or Sebastien told us a little bit about them. From Alex's counting, there are six scientsts really working on it, approaching four who are working on it full time. Four of those six are high energy physicists, and two are astronomers...

One of them a theorist, so, he's worthless.

-A. Kim

The other astronomer is an X-ray astronomer, so has no optical experience. Alex says that they could use some astronomical experience. The other student is the EROS one who's going to do the rates.

Alex, as an outsider, thinks that censored.

Alex thinks that one of the most interesting technical goals they have is to eliminate IDL. What they are doing is building a lot of pieces of sofware, tools that we have here. They are making a library of C++ tools. Alex thinks it would be good if we coordinated with what's going on there and here.

There's crap every time, and the crap's always different.

-A. Kim on the need for searchscan software