July 22, 2014 ::
by Wallace Tucker
On April 20, 2014, Harvey Tananbaum stepped down after 23 years as director of the Chandra X-ray Center (CXC). This event was duly noted in various press releases, but its significance may not have been widely appreciated.
By almost any measure Harvey has been one of the most outstanding directors ever of a major astrophysical observatory. He was director of Chandra for longer than any other director of a world-class telescope. In fact, he became director of Chandra in 1991, 8 years before it was renamed from AXAF (Advanced X-ray Astrophysics Facility) to Chandra upon its successful launch in 1999.
In 1976, Harvey had joined with his mentor, Riccardo Giacconi, to write a proposal for AXAF based on Giacconi’s original concept dating back to 1963. During the late 70’s and through the 80’s, Harvey led the SAO mission support team, which worked with NASA’s Marshall Space Flight Center to develop and demonstrate critical technology for the mirrors at the heart of the Chandra Observatory. The MSFC-SAO team then over-saw the building of the observatory. SAO was competitively selected to develop the AXAF Science Center (with Harvey as Director) in 1991 to plan for science operations and to assist the science community in using the observatory. In the mid-1990’s responsibility for operations and the observatory health and safety was assigned to the Science Center which was renamed the Chandra X-ray Center shortly before launch.
Throughout the course of the program, Harvey had worked tirelessly to get support from the science community, priority within NASA, and final approval by the Executive branch and Congress. Then when final approval turned out not to be final approval because of a budget crisis at NASA not related to AXAF, Harvey along with MSFC Project Scientist Martin Weisskopf, was in the middle of the effort to restructure the program in ways that ensured not just a successful mission, but one of the most powerful and versatile tools for doing astrophysical research in the 21st century.
The Chandra mission has exceeded its original planned lifetime by ten years, and it consistently ranks first or second with Hubble in measures of scientific productivity such as citations in the scientific literature and the number of research papers (about 6,000 at present for Chandra) published by scientists using the observatories. As Harvey will be the first to tell you, Chandra’s success is due to the efforts of not just one, or ten, or even a hundred, but literally thousands of scientists, engineers, managers, programmers, technicians, etc. That is certainly true, but just as certainly, Harvey played a critical role for more than three decades.
In terms of taking the credit for Chandra’s success, Harvey is content to fly under the radar. In fact, as a colleague once mentioned, Harvey was for all this time like radar, continually scanning the horizon for problems –and opportunities– to ensure the success of the mission. This has involved looking at abstracts for virtually all of those 6,000 papers and reading a significant subset of them, reviewing technical and engineering reports, interacting and staying informed through countless conversations with the scientists and engineers involved, all while being available 24/7 should a problem or a scientific Target of Opportunity arise.
This is nothing new for Harvey. He did similar work for a shorter time for the Einstein Observatory before Chandra, and before that, for the Uhuru X-ray satellite. It was Harvey’s work on Uhuru that caught Giacconi’s attention, and he was quickly given increased responsibility, as project scientist for Uhuru, then Scientific Program Manager for the Einstein Observatory, and in 1981 he succeeded Giacconi (who had left to become Director of the Hubble Space Telescope Science Institute) as director of the High Energy Astrophysics division at the Harvard-Smithsonian Center for Astrophysics.
On April 30, I talked to Harvey about his career before, during, and after Chandra. Excerpts of this conversation follow.
WT: Are you now officially retired as director of the CXC?
HT: Yes. Belinda Wilkes started as director on April 20.
WT: How does that feel?
HT: It feels different, that’s for sure. We had a meeting at NASA headquarters and 3 of us made presentations on Chandra and on a future project. It was a test, with a transition to the “new normal.” Last week we had a CXC senior staff meeting, and I didn’t attend. Having those couple of hours was useful, but not being at the meeting was a change.
WT: What were some of the high points of your 38 years working on AXAF/Chandra?
HT: From 1976 to 1991, at first, I had no real title. I was PI (principal investigator) for the mission support contract, so I could poke into anything that was going on. Being CXC director for the last 23 years was more formal.
As they say, when you bake a cake or pie, the most satisfying part is in the eating. Sometimes it is also fun to bake it together, and do things with colleagues and friends.
Clearly the launch, and initial deployment from the Shuttle, and getting to the final orbit, and opening all the doors, and actually seeing the first X-ray source before the formal first light, and knowing that we were going to image and locate sources to arc second precision - that was very satisfying.
A week later, when we got the official first light and watched the image of the Cas A (supernova remnant) build up on the screen, and saw the little dot in the center after the first 15 to 20 minutes, that was a pretty powerful indicator that Chandra was going to deliver the goods scientifically. At that point we didn’tknow it would work for 15 years, but now the assessment of the engineers is that there are no showstoppers for at least another ten years. The community is not running out of ideas, the proposal oversubscription requesting observing time is still about 5 to 1, and the number of papers in refereed journals is about 450 per year. Seeing that kind of productivity is very satisfying.
Going back in time to other highlights - seeing the first pair of optics polished and the first X-ray test in 1991 with images not quite what we expected, then fixing the holders to better account for the effects of gravity, and a few weeks later seeing that the full-width half maximum of the image was about 0.2 arc sec, much better than the 1 arc second which we had promised to Congress --that was a very high point in the program--. The optics were better than those Congress had required.
The actual launch postponements were disappointing, and heart stoppers. The first countdown went down to 6 seconds. I kept waiting for launch, and my younger son Greg nudged me and said, “Dad, the clock has stopped.“
It turned out that the launch had been aborted because of a spurious reading on one of the sensors of the main engines on the shuttle. Two nights later, on the next launch attempt, the weather forecast was that there was a zero-percent chance that weather would be a problem, then a lightning storm came up and the launch was scrubbed, and re-scheduled for the next night.
The third attempt was a success, even though there were some problems that we weren’t aware of at the time. A computer controlling the engines went out and the back-up switched on to control things.
The launch was surely a high point. I stayed out in front of the bleachers. I was one of the last people to leave. You could see the spot of light that got fainter and fainter. You could track it for several minutes, until it was fainter than one of the faintest stars. My wife, Rona asked me what I was thinking. I had seen AXAF/Chandra from the drawing board, to the parts level, until it was fully assembled. In some sense, it was like sending a kid off to college and saying goodbye. We were definitely letting go of the hardware and could never touch it again, but of course we expected to communicate regularly.
I think I was also contemplating that, whereas everyone else was cheering, and had gone back to the bus and hotels to celebrate a successful mission, there were still more hurdles to clear. Launch was a big, big step, but over roughly 3 weeks, the inertial upper stage and other engines had to burn 5 times to get into the proper orbit, and the various doors had to open, the mirrors had to work, we had to lock on stars. A lot of things had to work before it was a successful mission, and although that moment at launch was a high, I knew that I would feel a lot better in the next 3 weeks when we knew we were going to have a successful mission.
WT: What were the scientific high points for you?
HT: I am very attached to the science even though most of it is not my own. My science role has been to review requests for Director’s Discretionary Time, to work on the Chandra proposal submitted every 2 years for NASA’s senior review for extended missions, and to co-author occasional review articles. Some of the exciting Chandra results that stick in my mind are related to images of extended objects such as clusters of galaxies or supernova remnants. In clusters, we see cavities or regions of greatly reduced X-ray brightness in the hot gas filling the cluster volume. These cavities are produced by jets from black holes that give us a sort of fossil record of past explosive activity. The evidence for dark matter as clearly distinct from ordinary matter based on the Bullet Cluster is one of Chandra’s best results.
Finding the point source in Cas A, almost on Day 1, was very satisfying and a demonstration of Chandra’s power. That source has gotten fainter over the past decade because the neutron star has cooled much more rapidly than may have been anticipated. The rate of cooling indicates that the interior is in a superfluid state, allowing us to probe at matter at super high density.
WT: What were some of the most emotional moments?
HT: Two or three examples beyond those already discussed. We had worked for more than a decade on the technology, trying to make the most precise mirrors ever, and had struggled to get into the NASA new start budget, even though we had the 1980 endorsement of the Field Committee (a National Academy of Sciences committee of distinguished scientists chaired by George Field that recommended priorities for the next decade). In the mid-80’s, Len Fisk became NASA Associate Administrator for Science and Applications, and picked Chandra. The OMB (Office of Management and Budget) kicked it out of the budget plan for FY1989, and then Fletcher (James Fletcher, NASA Administrator) asked to put it back in.
OMB kicked it back out again. Fletcher planned to appeal to the President – Ronald Reagan. Art Fuchs (the AXAF manager at NASA headquarters) and Charlie Pellerin (Astrophysics Division Director at HQ) wanted a compelling, brief argument – it had to fit onto 1 page and had to convince the President to over-rule OMB. I went out running, something that often opened my mind up to new ideas, and when I came back, I suggested to Art that we make a presentation showing 3 flags – Russian, Japanese and U.S. – on the bottom, we say “who shall be first in X-ray astronomy?” As unusual as it was, we were somewhat desperate and needed an “out-of-the-box” approach, so it was approved by Pellerin, Fisk, and Fletcher. Fletcher went over to the Executive Office Building as a prelude to meeting with the President and talked to Jim Baker, Howard Baker, Donald Regan, and maybe William Miller, showing them the chart. They asked, “Why didn’t you tell us this project was so important? Of course we will approve it.”So our chart never went to President Reagan.
A second crisis came after we showed that the first pair of mirrors performed as promised in the fall of 1991. The following January we were called to NASA headquarters, where Charlie Pellerin told us we had to downsize, since the budget wouldn’t accommodate Chandra as planned.
We were aghast, and opposed the decision. Chandra, like Hubble, was supposed to go in low-Earth orbit, and be serviced. . . . The budget people were opposed to that. This led to a 5-6 month period of restructuring (or arguing about how to do so) that was one of the lowest points in the program. We simply couldn’t converge on a design. Cutting the number of mirrors would not save that much money, although Charlie Pellerin argued otherwise.
Claude Canizares (MIT and one of the instrument PIs) and I finally set up a meeting with Len Fisk, and asked for the boundary conditions on the costs. There were yearly caps on costs, and then on the total cost. It was clear from the Hubble experience that servicing was a huge cost driver. So NASA had decided to put Chandra in a higher orbit, where no servicing would be possible. We were able to get 2/3 the number of mirrors (4 out of 6 pairs), but the higher orbit increased the available observing time. Ultimately, it turned out to be a great choice, because the higher orbit is a more benign environment without continual temperature changes due to going into and out of the Earth’s shadow. That orbit actually had a huge benefit.
A series of lows came the last year or two before launch, when we had a few delays. The ACIS door failed to open during the thermal vacuum test at TRW. We never figured out why, so we made some changes and had to launch with a little bit of uncertainty. There were delays due to issues with the instruments and then with reliability of various parts – mostly related to spacecraft units which led to a dozen alerts, further checks and sometimes reworking and replacement of components or electronics boards. Up until 4 or 5 days before launch, we didn’t know if we were actually gong to launch because of a final issue involving possible cracks in some capacitors.
The whole program had a lot of highs and lows. In any case, it was a voyage.
WT: Was there ever a moment when you thought it wasn’t going to happen?
HT: I think realistically, no. There was never a time when I thought that it wasn’t going to happen. I surely didn’t think it would take 23 years from the time Riccardo and I wrote the 1976 proposal until launch of the observatory, which became AXAF and then Chandra. I was 34 years old at the time of the proposal, and I was coming into the prime of my career. I am not sure I would have done it if I knew it was going to take 23 years . . ..We thought it would launch in 7-8 years, but the time just gradually stretched out . . ..We kept making progress. We did not get all of the money in many of the years, but we got some of it and kept making progress. What we ultimately built and flew as Chandra is much, much better than what we could have done in the late 70’s and early 80’s.
WT: So, like in a voyage, you never thought of turning around and going back home, because you thought there would be something at the end of the trip.
HT: The other thing is, you have a certain investment, and you always think the investment is going to pay off. It is very difficult to walk away. Several years after Riccardo left to become director of Hubble, he told me directly that he thought AXAF was taking too long. Maybe he could have made things happen faster, but many things were beyond anyone’s control, such as the explosion of Challenger.
WT: Anything you would have done differently? If you could rewind the clock?
HT: You can’t really say. Maybe if we could do it over, the six agonizing months of restructuring could have been avoided. If we would have known how it would play out, and gone to Len Fisk in the first month, and found out the boundary conditions, maybe we could have settled on the final solution in one month rather than six.
I think collectively, we have made a lot of great decisions through the course of the program. I bet there were tens of thousands of decisions made, and very few of them were “wrong”.
WT: So, knowing what you know today, you would get back on board again?
HT: Definitely. With the success of Chandra to date, and the bright future still ahead, I think all of the work over the years has led to a great “return on investment,” for NASA and the science community, and for me personally. So yes, I would do it all over again.
WT: There has been a great return on investment for the public as well. What are your plans for the future?
HT: I plan on continuing to work, partly on Chandra. I do hope to put time in on planning for the successor for Chandra –SMARTX – 30 times the area with comparable angular resolution, and much better off-axis performance. Hope to launch around 2030.
WT: Good luck with that, and thank you for leading us all on an exciting voyage with Chandra!
The planning and building of Chandra are discussed in:
W. Tucker and K. Tucker, Revealing the Universe: The Making of the Chandra X-ray Observatory (Cambridge: Harvard University Press, 2001)
The transcript of an extensive interview of Harvey Tananbaum conducted in 2002 by Patrick McCray for the American Institute of Physics Oral History Project can be found at: