| The LIGO
LIGO Scientific Collaboration NewsResonant Bar Detector Dedicated at Hanford
An Interview with LSC Spokesperson Peter Saulson
First LSC Meeting in Germany
During the November LSC meeting at the LIGO Hanford Observatory, a historical exhibit of one of the original resonant bar gravitational-wave detectors was dedicated. The bar was one of three used by Joseph Weber of the University of Maryland in his pioneering experiments to search for gravitational waves from astrophysical sources. Another of the bars is on exhibit at the Smithsonian Institution in Washington DC.
LIGO Director Barry Barish presided over the dedication which included brief talks from Peter Saulson, Rai Weiss and Professor Virginia Trimble. Virginia was married to Joseph Weber and facilitated the transfer of the resonant bar from the University of Maryland to LIGO.
Peter described the imaginative and maverick role Weber played in initiating the field of gravitational-wave astronomy during the 1960's, an era when General Relativity became coupled to experimental research. LIGO is a descendent of that period and, particularly, of the spirit that led Weber to attempt an actual experiment.
Rai spoke of his personal recollections of Weber, an association that extended from the middle 1950's, when Weber was carrying out microwave spectroscopy on molecules, through the time of the stunning and controversial initial ventures to observe gravitational waves. The relationship continued into the 1980's when Weber became interested in detecting neutrinos. Rai described his personal interactions with Weber as full of debate.
Lastly, Professor Trimble gave a brief vignette of the courtship and intellectual life she shared with Weber. Later that evening, she gave a public lecture at the Battelle Auditorium in Richland, entitled "Astrophysics Meets the Millennium."
This summer, Peter Saulson, of Syracuse University, succeeded Rai Weiss as spokesperson of the LIGO Scientific Collaboration (LSC). With each new gathering, the LSC grows in importance to the overall functioning of the LIGO Laboratory, and we were interested to hear Peter's thoughts on the status of the Collaboration and his vision of its future. We sent our roving correspondent, Linda Turner, to catch up with Peter to find out a little more about the man and the pivotal organization he now helms.
Can we start with a bit about your background and your current position at Syracuse? And then maybe you'll bring us up to speed on how long you have been involved with LIGO...
I got my Ph.D. from the Gravity Group in the Princeton Physics Department in 1981. My adviser was Dave Wilkinson, who just passed away about a year ago. I went there to work with him because I was interested in measurements of the cosmic microwave background radiation. My first year in grad school I went with Dave to the National Scientific Balloon Facility in Palestine, Texas, with a microwave anistropy instrument. Everyone on the staff there looked at it and said, "That looks like Rai Weiss's package. Do you know Rai? He's a great guy." That was my first knowledge of Rai.
I ended up getting my degree for an infrared astronomy measurement that looked for dark matter in the form of dim red low-mass stars surrounding spiral galaxies. (Norna Robertson also did similar work around the same time.) We didn't find dark matter, but it was fun.
Then I had to figure out what to do for a living. Several things influenced my choice. The Gravity Group's belief was that you would do the best physics if you thought of something important to look for, designed a new kind of instrument to look for it better than had been done before, and then made the instrument work. It was less important that you actually find it than that you be looking for the most interesting possible thing. That idea was reinforced in a 1981 book written by Martin Harwit, called Cosmic Discovery. He said that almost all important astronomical discoveries had been made by people who built new kinds of instruments. The assertion made a lot of astronomers mad but it was just the Gravity Group's philosophy. Harwit drew a list of new possible discoveries to be made with new instruments, and gravity wave detection was right at the top.
During that year my adviser Dave Wilkinson said that Rai had applied to the National Science Foundation (NSF) for funds to see if a big gravity wave interferometer could be made which would be sensitive enough to actually detect gravity waves. Dave said that he thought Rai was just the right guy to figure out if the gravity wave field could succeed--or whether it should be put out of its misery.
So when I learned that Rai had received the funds from the NSF to do the design study for what eventually would become LIGO, I was ready to be very interested. I hadn't been all that interested in gravity waves before, in spite of the fact that one of my best mentors at Princeton was Steve Boughn, who had gotten his Ph.D. helping to build the Stanford cryogenic bar with Bill Fairbank. But Rai's plan seemed to be the way to just barely get in on the ground floor of something big. Rai thought that by the early '90s, we'd be in business.
I spent the next eight years working at MIT with Rai and a lot of other good people. (David Shoemaker was a student there for part of that time.) It was a great time to learn things from the bottom up. We had a tiny prototype interferometer, just 1.5 meter on a side, that fit on a single granite table. It used delay lines in the arms and an Argon-ion laser. It was hard to make it work, and did best at night.
The grant that paid my salary was for the design study of large interferometers. We worked with Arthur D. Little, Inc. on the vacuum system conceptual design (they recommended ion pumps,) and Stone & Webster Engineering on the civil construction conceptual design. I remember one day standing at a blackboard with a S&W engineer when he sketched out a concrete pipe arch for a protective cover over the beam pipe. Years later, when I saw it actually built at Hanford, it was an eerie feeling.
I also helped out with the 1.5 meter interferometer, but mainly my "physics" work was to think about vibration isolation. I worked on active vibration isolation, building a very simple early prototype of what will go into the Livingston Observatory retrofit and into Advanced LIGO. Later, I realized there was physics that we didn't understand about thermal noise, and started trying to learn more about that.
In 1989, I went for a visit to JILA at the University of Colorado, Boulder. Peter Bender, Jim Faller, and Tuck Stebbins were interested in understanding how low in frequency could earth-based interferometers go, which grew out of their work on LISA. So I went there to work more on active isolation system design. I also started following up my interest in thermal noise with some simple experiments.
During my time at JILA, I read an ad saying that there were five faculty openings at the Syracuse University Physics Department. The Chair there was Marv Goldberg, now at the NSF. He was very successful at convincing the University to support his initiatives. Another striking thing about the ad was that the openings were in any field of physics. That was crucial because no one was hiring specifically for LIGO in those days. I was lucky that Marv was interested, as was the prominent relativity theorist Abhay Ashtekar, who was also at Syracuse then. They made me an offer, and I started there in January 1991.
In those days, there was no LSC. The only people working on LIGO were at Caltech and MIT, and we didn't even have a context for thinking about how one might work on LIGO from somewhere else. Some creative thinking by LIGO's then-Director Robbie Vogt and by Rich Isaacson of the NSF made it clear that it would be possible as long as my work wasn't being relied on to actually build LIGO. I wanted to follow up on thermal noise issues, and that seemed sufficiently "off to the side" to make us all comfortable with the arrangement. I got NSF funding the next year, and started building a lab. Gabriela Gonzalez was my first graduate student at Syracuse. She did a great job in building up the lab, as well as doing a very nice thesis on thermal noise.
The thermal noise research went very well, and I enjoyed teaching at Syracuse. But I was pretty isolated from all the hard work going into actually designing LIGO. That changed when I took a sabbatical from Syracuse to work at the LIGO Livingston Observatory (LLO) in 2000. This had been a suggestion of Barry Barish, LIGO's present Director, as a way to get me and the Syracuse group involved in Initial LIGO. It was a great year. I had the title of LLO Interferometer Commissioning Leader. That was a bit premature, since a lot of installation was still going on during 2000. I helped to commission the LLO mode cleaner, but spent a lot of time just trying to catch up on what the LIGO interferometers actually were. I was incredibly impressed (and still am!) with the design and the execution of it. Since then, the commissioning has gone really well.
When did you take on your new role with the LSC? And can you tell us a little about the responsibilities you now have?
Rai Weiss has been the LSC Spokesperson since it was first organized in 1997. He was instrumental in building it from nothing into a thriving organization--with, of course, a lot of help from others.) But after five-plus years at the helm, Rai decided he was ready for a change. So he asked the LSC Nominating Committee to run an election to choose a successor. The results of the election were announced this past March, just before the LSC meeting in Livingston. Rai wisely proposed a "soft start" for my term; I first chaired a meeting of the LSC Executive Committee back in May.
The LSC is the organization that unifies scientists, both within the LIGO Lab and from other institutions around the world, to carry out the scientific program of LIGO. Of course, the people of the LIGO Lab at Caltech, MIT, Hanford, and Livingston, play a central and irreplaceable role. But several aspects of LIGO research make sense to share more broadly. Historically, this first included research on how to build a more sensitive interferometer; out of this research came the ideas that were brought together in the Advanced LIGO proposal. Not long after the LSC was launched, though, Rai got various data analysis working groups going to prepare for the stage where we are now--actually looking for gravitational waves in the data from LIGO (along with the data furnished from our strong partnership with GEO as well).
The LSC Spokesperson's job is to coordinate the work of the LSC scientists. Since the organization has become so strong, that means mainly letting people do the work they know how to do, and letting the various Working Group leaders, Search Group Chairs, and others handle their responsibilities. Of course, LIGO is moving into new ground, with the release of our first set of papers on gravity wave observations, the rapidly developing sophistication of the analyses for the data runs following our first Science Run, the continued intensive commissioning pushing us through the last order of magnitude we need to reach Initial LIGO's sensitivity goals, and the rapid progress toward the hoped-for approval of Advanced LIGO, and then the wonderful challenge of making Advanced LIGO work. So the Collaboration is continuing to evolve, and the spokesperson needs to help that process along.
What does this role mean to you professionally?
Many people have devoted huge chunks of their professional lives to shaping LIGO into a reality, and now to making it work. I'm proud to have been able to help out in a variety of ways over the years. Spending a term as LSC Spokesperson is another way to be of service to this great endeavor.
How about personally?
One of the great pleasures of life as a scientist is to be able to share in a great quest with colleagues one respects and admires. The LSC is a wonderful body of colleagues who share in the quest to discover gravitational waves, and then to use them to learn more about the Universe. I can't imagine a more satisfying way to have a career in science.
What would you consider to be among the more exciting elements of the collaboration?
Harry Collins has taught a lot of us how to think about the organization of our work from a sociological point of view. I think back to before we had the LSC, and to when it was just getting started due to suggestions by Barry, [LIGO Deputy Director] Gary Sanders, Rai, and our friends at the NSF. A lot of us made cynical jokes about the impossibility of "herding cats" (or less flattering metaphors.) It was a real struggle to figure out what we wanted the LSC to be, and then to make it happen. It could easily have failed. But we are all better off for having persevered through that struggle. So for me, the most exciting thing about the collaboration is that it exists, and that it does its job so well. You could say that many of us learned a new way of life, and made that way of life a reality for ourselves. We had good role models from other branches of physics, but we had to build our own organization in the style that suited us and our work.
How about the most challenging aspect?
Our greatest challenge as a collaboration is to continue to grow into our new responsibilities. We've struggled through commissioning Initial LIGO, and we're getting very close to our goal. We struggled to invent a coherent plan for going beyond the goals of Initial LIGO, and now we've made the proposal for Advanced LIGO, which has been very well received. We struggled to get the first Science Run (S1) data analyzed and to write clear papers describing those analyses, and now we've succeeded at that. Just around the corner are new challenges, to fulfill the promise of our past successes. I'm not very worried about continued success in commissioning (although I'm sure the work will be very hard), nor am I very worried about making Advanced LIGO work, although there are very real technical challenges there, too. We've proved to the world, and to ourselves, that we know how to meet those kinds of challenges.
Where I think we have the most growing to do is in the continued development of our program of searching for gravitational-wave signals. Getting the S1 papers out is a wonderful thing, but we went into the S1 analyses planning to set upper limits. Before the LSC can be said to have matured, we'll have to be as comfortable with a possible discovery as we are with setting an upper limit. Stan Whitcomb, head of the Detector Support Group, led a very interesting discussion on this topic at the LSC meeting that took place in Hannover this summer. Having that discussion was a great step forward. But we are certainly far from having reached a consensus as a collaboration on how we'll become confident we've seen gravitational waves. We need to keep working on this. It will be another struggle, but one we should all relish.
How many groups are currently represented in the collaboration?
Over thirty groups, presently. (Exactly how many depends on how you define a group.)
Do you foresee the collaboration as continuing to grow--and, if so, by how much and in which areas?
It will almost certainly continue to grow--though by how much, and in which areas, are both hard to say. It will depend on who gets interested, and on what interests and skills they bring. Certainly people in the LSC feel swamped by the amount of work it takes to do a good job, so I can't see turning away good, interested people.
In the past, the LSC has met twice a year. That format recently changed and the LSC now meets more frequently. Why?
For the first several years of its existence, LSC meetings focused on the activities of the Working Groups devoted to research on advanced interferometer technology and on data analysis techniques. But once we started gearing up for the data analysis that is now under way, we needed to spend additional time for meetings of the Upper Limit Groups (now changing their identity into Search Groups.) The idea behind their structure was that they shouldn't consist entirely of the people who had originally devoted all of their time to planning the data analysis. Those folks are still key leaders of data analysis, but their skills and interests needed to be complemented by those of people who had been building the interferometers and were making them work. We wanted people to be able to be active in a Working Group (say, in Lasers or in Suspensions) and also to participate in the actual search for gravity waves using LIGO data. To do this required not only new organizations within the LSC but also new meeting time within the LSC meeting. The twice-yearly meetings added an extra day (usually the Monday, called Upper Limits Day.) It isn't clear that that is enough time; at the past several meetings, many of the Upper Limit Groups have added separate unofficial face-to-face meetings onto the official four-day LSC meeting.
The change to the current three meeting per year pace arose out of a different, but related, need. Now that we are writing papers based on our analyses, it has become crucially important for them to be discussed carefully by the whole LSC. We first did this last March. For a couple of hours each, the four Upper Limit Groups presented their S1 results and led discussions of those results both from their officially-appointed LSC reviewers and also from many interested LSC members. I was tremendously impressed by the depth and the seriousness with which the LSC took its responsibilities. It is clear that the papers were much improved for having had those discussions. It is also true that only by having discussions like that can the work of a part of the LSC come to be truly a product of the LSC as a whole. I don't think it is possible to do that just by written reports and email exchanges, and even telecons don't seem to work nearly as well as intense face-to- face discussion.
The S1 experience was exciting and new, but it was also a prototype for a process that will have to become routine for the LSC. We'll soon be discussing results from the second Science Run. (Preliminary results were presented and discussed at the November meeting in Hanford.) The third Science Run is already well underway. And after the commissioning period of the first part of 2004, we plan to launch a much longer run. So we have to build in a way for the LSC to routinely have the same kind of discussions that we had last March.
We also need to be able to have them more often if LSC deliberations aren't to turn into a roadblock to publication of our results. This was recognized by the Upper Limit Chairs, who got together at O'Hare airport last May for a one-day retreat organized by Alan Wiseman. One key idea that emerged from that discussion was that we needed to increase the frequency of LSC meetings to keep pace with the progress of data analysis. In particular, we need to be able to review new results in advance of major outside meetings where we'd like to present papers. We considered the idea of moving to four LSC meetings per year, but in the end we decided that switching from two to three was a big enough change for the time being. It turns out that we have three major outside meetings each year to prepare for: the Gravitational Wave Data Analysis Workshop (GWDAW) in December, the APS April meeting (that sometimes happens in early May), and in July the Amaldi meeting (or in off years the GR meeting). We're wrestling with setting up the new pattern of LSC meetings, but fitting it to the pattern of outside meetings is one important consideration.
So this change in LSC meeting frequency was driven by the needs of the data analysis process. It is an interesting question how this change fits the needs of the other crucial activities of the LSC, in particular the groups involved in technological work leading to Advanced LIGO and beyond. There was some discussion that three meetings per year may be too often for the Working Groups to want to schedule full meetings at each one. I don't know if that's true, and the answer may be different for different Working Groups. But if it is, it might not be a bad thing. One thing I've often felt was that there were too many interesting discussions happening simultaneously at an LSC meeting. Maybe we'll move to a situation where LSC meetings specialize, with fewer technical sessions but with greater chance for all LSC members to be involved. This could also be an opportunity for people to feel more free to participate in data analysis discussions, when in the past they may have felt fully committed to meetings of their primary Working Group.
To wind up, how about cluing us in to some of the important milestones coming up for the LSC that we should all be on watch for?
There's quite a large number of big milestones coming up. Currently we are underway with the third Science Run, and we are working to collect the best data we can. We are still in the process of getting our S1 papers published. Our instrument paper will have been sent to Nuclear Instruments and Methods by the time you read this, and the Inspiral and Pulsar papers, now posted on gr-qc, will soon go to Physical Review D. The Stochastic paper will be posted soon, and after a three-week comment period will go to PRD as well. The Burst paper has been presented to the LSC Executive Committee for approval. Hard on the heels of all this will be completion of S2 analyses, and moving on to S3 data. The S2 work includes a joint analysis with the TAMA project in Japan. S3 will return to the pattern of joint data collection and analysis between LIGO and GEO.
There will be an intensive commissioning period after S3. At Livingston, the HEPI active seismic isolation system will be installed, with the goal of enabling the interferometer there to work around the clock. At Hanford, commissioners will be pushing hard to reach the sensitivity goal of the Initial LIGO Science Requirements Document.
By late 2004, we hope to start a long Science Run at good sensitivity, perhaps six months duration. That will constitute several milestones at once. It will be a substantial change in the way LIGO scientific work is carried out. Commissioning will take a long hiatus, freeing commissioners to be more involved in data analysis. Monitoring of data collection will need to become routine if we are to maintain it for so long a period. And our data analysis will have to be better organized--that is, we won't plan on waiting for the end of a Science Run to start analyzing the data. We'll have to find ways to keep up.
In parallel with all of this, work continues on preparing to build Advanced LIGO. The key first milestone there will be approval of the proposal. Beyond that are the processes of final design, installation, commissioning, and operation.
The big milestone we are all anticipating is the detection of gravitational waves. It is hard to predict when that will occur, but it will certainly be a big moment when it does. It may not happen until Advanced LIGO is running, but we will all be working hard to take advantage of whatever Nature sends us in the meantime.