|Bob Serafin. (Photo by Carlye Calvin.)|
Bob Serafin's directorship of NCAR began on 1 May 1989. For the ten-year anniversary of his appointment, Staff Notes Monthly asked Bob to reflect on some of the pleasures and perils of directing NCAR during a time in which the scientific landscape he navigates has grown far more complex. (Rick Anthes completed a decade as UCAR president last fall; his retrospective "Ten Years of Change" appears in the Winter 1998 issue of the UCAR Quarterly.
SNM: You began as director of NCAR not long after the summer of 1988, when the greenhouse effect and global warming had just broken into public awareness in a big way. How has NCAR's role in global climate research evolved since then?
We were engaged internally in very significant scientific debates among the staff on whether NCAR should even participate in this thing called the U.S. Global Change Research Program [USGCRP]. There was a lot of skepticism among the scientific staff about the strong programmatic focus. One of the things that has happened over the last decade is that funding for science has become increasingly programmatic in nature. An increasingly large fraction of the NSF budget is assigned to programmatic research, while the core or basic research component is diminishing.
The decision we had to make, and I certainly didn't make it unilaterally, was whether we wanted to participate in these more programmatic activities or whether we wanted to focus only on basic research, come what may, and accept the consequences. And the consequences would be a diminishing of NCAR's funding, program, size, and so forth.
The staff and the institution eventually adopted the position that NCAR had much to contribute and that we should participate. We hoped that these strategic opportunities within the USGCRP and in other areas would be sufficiently broad that they could benefit the full spectrum of geosciences at NSF and therefore at NCAR. This has, over time, come to pass.
We also formally acknowledged that the NSF programs themselves did not afford us all the opportunities that were afforded by the [other funding] agencies collectively. And so we had to look to other agencies for support of certain research. This led to great debates over how much of NCAR's budget should be NSF-funded and how much should be funded by other agencies.
So that dialogue really took off in the nineties?
Cost-recovery and other-agency funding were issues even when Francis Bretherton was director [in the late 1970s]. It's not a new thing, but I think the debate became somewhat more intense during the nineties.
Another thing that we did in the NCAR director's office--Steve Dickson [then budget director], Peter Gilman [then associate director], and I--was to think differently about planning for the future. We built, for the first time in many years, a strategic plan that identified who we essentially were--call it our mission, including our close association with the universities--and also talked about what we wanted to do. This was debated much with the [UCAR] Board of Trustees and internally. Eventually, the board fully endorsed it. The message was that as long as the support we were seeking from other agencies was consistent with our strategic goals and objectives, then we should use that support to advance those goals and objectives. In very simple terms, we ought to do the right things, regardless of the source of the money.
Today we are encouraged by NSF not just to look to NSF for support but also to consider other sources of funding, because the objectives that each agency has may be somewhat different and no single agency may be able to provide for all of what we think is necessary. That's a tremendous change, a different paradigm. I don't know if this is really the best way for science to be funded, because it involves the writing of a lot of proposals to accomplish something that, 20 years ago, would have been handled through the submission of a single initiative to NSF.
So the landscape has changed a lot in these ten years, both for the good and for the bad.
What else is being emphasized in NCAR's current research strategy?
Our research has become more interdisciplinary. We try to involve universities in our programs and projects to the greatest extent possible. In fact, we've been very careful to be sure that, when we seek funds from other agencies, the research satisfies criteria established by NSF, UCAR, and the University Relations Committee so that we avoid unfair competition with the universities.
Another thing that's happened--a very important trend in this decade--has been more attention to, and awareness of, NCAR's role as a focal point and a leader for community programs. The recent review by NSF of all of our programs placed great emphasis on how we interact with our university partners in the community. The level of interaction is not only very substantive and very broad, but it now represents a large fraction of NCAR's activities. This is appropriate for a national center. NCAR is taking the scientific lead in many areas. The CSM [climate system model], the lead scientist role in the USWRP [U.S. Weather Research Program], the Solar Magnetism Initiative, programs such as MLOPEX [Mauna Loa Observatory Photochemistry Experiment], and new programs such as MIRAGE [Megacity Impact on Regional and Global Environments] are examples of this leadership.
I think the human-dimensions crosscut should be an integral part of most research efforts. An excellent example of this is in the USWRP, as exemplified by the work in ESIG. This work has been embraced by the weather research community, which five years ago wouldn't have understood why human dimensions should be a part of its research program or how that research could contribute.
NCAR and UCAR's role in education has been broadened. At one time we were discouraged from considering getting involved below the undergraduate level. Now there are good reasons for centers such as NCAR to be involved in K-12 education. Even if all these students don't become scientists, better appreciation of science by society is really important.
It sounds like all these things are generally consistent with the NCAR strategic plan and the UCAR 2001 planning document.
UCAR 2001 followed our strategic plan and broadened it somewhat, but at least four of the UCAR goal areas are exactly the same as ours: research, major facilities, technology transfer, and education.
On balance, tech transfer has been very positive for us because it has demonstrated that atmospheric research can benefit society. Sometimes we forget that we learn a lot from doing applied research that feeds back into basic research. One illustration is the work we began on convective storm initiation [in the 1980s]. Because of very applied work--testing and forecasting techniques for microburst warnings--a whole new area of basic research has opened up that's paying tremendous dividends.
What projects or philosophies have you had to push the hardest for?
One is that we needed to think strategically about what we were doing. We had to look at opportunities and could not expect a single source--NSF--to continue to meet all of our needs.
We have pushed for more interdivisional activities, and I think we are succeeding. It takes time to make these things happen. The most important element is staff buy-in. The staff has to be persuaded, and, in the end, it's the staff who do the work. I don't believe that anything gets done by directives out of this office. I do think that I can take leadership, and that others can take leadership positions, and that by discussing issues with segments of the staff, we can get new and exciting research under way.
I'm curious about the tradeoffs the organization has had to make. One of them seems to be between the largest programmatic projects and "small science." Do we really have to choose one or the other, or can we accommodate both?
I hope that we're the kind of organization where people can and should be involved in both kinds of activities. We try to provide an environment that allows them to do both. I know that in some cases it's not been possible.
Early on, when Rick Anthes first talked about a program called the CSMP [Climate System Modeling Program, an attempt at integrated modeling in the early 1990s], it was not embraced across the community, nor within NCAR, for many reasons. One of the criticsms was that a climate modeling program cannot and should not be viewed as a Manhattan Project, that the Manhattan Project was engineering, and science can't be engineered.
I view it somewhat differently. I think that institutions like NCAR have to set ambitious goals and take some risks. With CSM, we did set some goals. We said that the first version of the model would be ready within two or two and a half years. We said that we would use no flux corrections--we would just try to get the physics right in both the atmosphere and oceans so that we could demonstrate the simulation of a stable and reliable climate. And we achieved those goals. We put ourselves in the hot seat, no doubt about it, but the fact that we succeeded has been a very positive one for NCAR overall. I think we have now answered a lot of the criticisms that were then rather prevalent about NCAR having not addressed the issue of coupled model development.
A program is more than just setting a single goal and accomplishing it. The CSM continues to grow. It's a process that has engaged close to 200 scientists around this country and around the world. We are now acknowledged as having [one] among the best climate models in the world. We're taking a lead on improvements to our models, and the community is contributing substantively to those improvements. We have made some very important runs--the climate of the 20th century and the simulations of the 21st century--that are going to play a role in the U.S. National Assessment of climate change.
On balance, this large program has been very beneficial to NCAR and CGD and to the scientists who primarily did all of the work, who were willing to take the lead, willing to submit themselves to greater scrutiny and bureaucratic control, but I think who individually have benefited also. Because now they're acknowledged as leaders.
If NCAR suddenly had unlimited resources, what would you like to see us focus on?
We'd do everything, and we would do it with the university communities, and presumably there would be no debate over the lack of funding, and the competition issue would be resolved, but that's utopian, and this won't happen. Therefore, it is important to set or identify major priority areas. This has been done in our proposal to renew the cooperative agreement with NSF.
In addition, every year in June we conduct a retreat with about 100 participants--senior scientists, scientists III, senior technical staff, managers, sections heads, some administrative people. Out of last year's retreat we got a number of suggestions for things that NCAR should do, many of them being interdisciplinary in nature and all consistent with our five-year proposal. For about two days, we debated and finally agreed on what we would do if we had a $4 million increase in the NSF budget, which we thought was possible for fiscal year 1999.
We divided the $4 million into 10 units and then assigned those units. We assigned 5 to the first priority, which was fundamental research. We assigned 3.5 units to understanding and predicting the earth system, 0.5 to human dimensions, and 1 to undefined opportunities to allow some high-risk research. If we received less than $4 million, we would not necessarily allocate resources the same way.
We also said there were some unmet needs that were "off budget" in the sense that they were broad community needs that ought to be addressed at the NSF level. We certainly couldn't accommodate them within the $4 million hypothetical increment. These included the HIAPER aircraft and supercomputing at NCAR, along with enhancement of support for some basic ATD functions that had been affected by shrinkage of the NSF budget.
I think these NCAR-wide retreats are absolutely essential. It's also essential that each division have a retreat every year and that these retreats include some people from outside the division. Some divisions had not had a retreat for five years, but they're all doing this now. Call it "out-of-the-box thinking"--it's a term NSF and a lot of people in Washington are using now. It really pays off.
If you were to relive the last ten years, what would you do differently?
We've not succeeded in hiring a woman division director. We've tried hard and recruited widely. In order to increase diversity in our staff it is absolutely critical that we not only pay competitive salaries and try to hire well-qualified women, but also that we create a working environment that will nurture their careers and allow them to succeed here. I'm going to place additional emphasis on this in the coming year. I'd like it to become more a part of our culture. We've made a lot of progress. Rick Anthes has had a strong influence, and Susan Solomon, when she was acting director of ACD, also had a lot to do with moving us along the right directions. I think the SOARS program is a terrific opportunity for underrepresented minorities, because one of the critical issues is the pool of people available.
What have you learned about the nature of directing a place like NCAR?
I think you learn a lot through experience, and you take a lot of lumps because of that. Everyone I know who's taken on a major leadership role--division director, section head--they all take their lumps. I'm sure if you asked Rick [Anthes] this question, he'd tell you the same thing.
One of the things you learn early is that you have to be careful about what you say. Anything one says can be interpreted many different ways, and often in ways that are not intended. You can find examples of this not just in things I say or Rick says but in things that review panels say. I've looked at reviews of a division or a group within a division and found them to be generally favorable and nonthreatening, and yet individuals will read these reviews or a portion of them and take them very personally.
So you do have to be careful about what you say, but you also have to speak what you believe to be the truth and accept the consequences.
Is there any secret to picking yourself up and going on when you've made a mistake?
You just learn from what you've done. I try not to fret about the past. Sometimes we worry too much about mistakes, and we try to pass rules and regulations that will ensure that we never do "that" again. I'm not sure this is the best way to proceed. It results in a lot of unnecessary bureaucracy. At the individual level, learn from what you've done and go on.
Maybe it's characteristic of my style, but I try not to react initially to any crisis. I would bet that more than 50 percent of crises solve themselves within a few days, or at least there's enough additional information available that makes the crisis easier to solve. Problems need to be dealt with in a forthright manner that's fair to the individuals and also fair to the institution. Being excessively compassionate to individuals can be very damaging to the institution. Being excessively rigorous in the strict interpretation of policy and laying it on people can be damaging to individuals.
No matter how credible one side's view might be, the other side has something to say that needs to be heard. I think that's probably the secret of great diplomats--they listen to all the sides and serve as effective communication channels. I think all people in management face this problem to some extent.
You're now the longest-serving NCAR director on record.
(Laughs) Well, it obviously must mean something. It must mean that I am happy--I am--and also that the board and the [UCAR] president are happy with me. And I hope I've done a good job. That's really in the eyes of the people who judge my contributions.
I think this is one of the best jobs, if not the best job, in atmospheric science. I don't think there is another institution like NCAR. We have a lot of credibility outside these walls. We are a very visible organization and one that has been very effective, and I think the fact that we've been in existence 39 years now is pretty significant.
I never expected to be NCAR's director. I'm certainly happy that I've had the opportunity. I've enjoyed it a lot. It's a tremendously challenging job. It isn't unique in that respect--lots of jobs are challenging--but I think it's good for us to challenge ourselves. Each year I am awed by the tremendous work that NCAR does, by our accomplishments and by our vision for the future. After 25 years here I still relish the opportunity to go into the field and interact with the many scientists and support staff--always trying to learn more about the earth system. I like to meet with our postdocs and see their enthusiasm. They represent the future.
It's when you stop improving, stop trying to do things better, stop trying to do something new and exciting, when you're no longer interested in learning, then you've been in the job too long and you should do something else.
I think NCAR has always been a good place for people. Salaries are fair, benefits are good, people pay a lot of attention to individuals. I felt that when I walked in the door 25 years ago. While a lot of things have changed--many of them have been imposed upon us by rules and regulations of the federal government and so forth--this is still a very good place that allows us to do some excellent research and make technological advances and also for people to enjoy their careers. I don't know if we could do more to build a greater sense of family and belonging. Maybe we need to give more attention to this.
It's impossible to maintain the status quo. That can never happen. We can go forward or go back, and either one will represent change. Change is generally something people are concerned about, but it's inevitable. We had about 90 scientists on staff [in 1989]. If we hadn't made some strategic decisions, and the staff had not embraced them, we'd probably have 75 or 80 scientists now. Instead we have about 130 scientists and another 100 long-term scientific visitors, including our postdocs. I'm sure that some elements of the program have suffered, but I think they'd have suffered anyway because funding from a single source during this whole era would not have been sustainable. It's not NSF's fault--they would have liked to have done more. I'm very grateful, as all of us are, for the NSF's commitment and its continuous support of this institution in now its fourth decade.
I think other disciplines could benefit from something like NCAR. It's been an interesting and unique experiment and one that has succeeded. But I think we also have to expect that it will change. We have to be adaptable to change, and often we must be the agents of change. No director, no president of UCAR, can do this alone. They can only be a part of it, as I said earlier. The idea is to be open to debate and discuss and take the right steps and hope that staff will pick up on some of it, and also tell us when we're dead wrong and get us off some of the dumb things that we might propose.