Well, it’s only the institute’s second decade. So I was impressed.

  Tell us about IES and its history.

We’re an unusual institution. We are a not-for-profit, free-standing research and education institution. That’s a very odd animal these days. We’re in Millbrook, New York, which is not far from the city of Poughkeepsie.

IES was started in 1983 when Gene Likens was hired to be its director. What had existed at this location before was called the Cary Arboretum, which was a botanical and horticultural institution concern. It was very small with a visitation program and a large tract of land, originally from the bequest of a wealthy donor, whose name was Mary Flagler Cary. The Mary Flagler Cary Charitable Trust was established after her death to manage her assets. Because Mary Flager Cary had a life-long interest in plants and especially trees, the Trust transferred the ownership of the property to the New York Botanical Garden as its steward. Eventually the Botanical Garden and the Mary Flager Cary Charitable Trust decided to change course and wanted to set up some a more ecologically-oriented institute. In 1983, Dr. Gene E. Likens, then a professor at Cornell, accepted the position as Director. In 1993 the Institute of Ecosystem Studies separated its ties with The New York Botanical Garden and became an independent institution. Likens has won every award ecologists can possibly win and a few that ecologists can't win. He recently got the National Medal of Science, which is very rare for ecologists. He was one of the people who discovered the phenomenon of acid rain and its causes. Upon his arrival as director he began to hire new staff and we are now 15 permanent Ph.D. staff. We function like professors in a university department. We don’t have tenure but we have the expectation of continuing. We’re reviewed like tenured professors every 5 years. We have 8 to 12 post-docs, depending on funding. We have graduate programs with several different universities—240 employees total. Our scientific staff is roughly the size of a large ecology department at a university.

  And is most of your funding still from the Cary Trust?

Most of the science is funded from competitive grants from the National Science Foundation, or other federal and state agencies, and grants from private foundations. The overall budget is about 30% from endowment money that comes ultimately from the Mary Flager Cary Trust and the other 70% is from grants and other sources. The research program runs on grants. The permanent scientific staff is expected to recover a portion of their salaries from grants.

  How do you account for the institute’s dramatic increase in citations from the early to late 1990s?

Well, we came into existence in 1983 and really didn’t have a full staff on board until probably 1990, and the staff is mostly of similar age. Most of us are from our late 30s to early 50s, so we’re at the point in our careers where we’re really putting out a lot of work. In effect, we came of age with the institution. The Institute succeeded by growing, but the increase in citations is in part the result of the maturation of the staff. That sounds simple. But when Gene Likens arrived, he started populating the institute with young scientists who had done postdoctoral work elsewhere and with other collaborators. So it took some years for that group to grow into interesting scientists and into interesting collaborations with each other. What we’re seeing now to some extent is a result of that growth and that maturation process.

  Would you attribute any of the citation increase to deliberate planning or just this serendipitous development?

We do a lot of planning, but the director’s sense is that scientists function best when they’re just interested in the topic. That has been the major model for us. Get interested in something; go out and find funding for it and do it. We do, however, spend a lot of time talking about larger science initiatives and whether we should get together as a whole group and chase them.

  And do you usually decide to go after them?

Only if we’re truly interested in the topic.

  You mentioned before that the institute is very collaborative. How does that pay off in citations?

We are very collaborative here, and many of the highly cited papers reflect that. Although not all are what I would call home-grown collaborations. Take our #1 paper from Nature in 1997 ("The value of the world’s ecosystem and natural capital," Nature 387[6630]: 253-60, 15 May 1997), which already has 346 citations. It has some 30 coauthors—one is from IES. It’s cited very widely within science and within management. That’s not really what I would consider a home-grown paper, but it’s characteristic of this institute that the people collaborate widely. You go down the list further and you find one by Likens et al. in 1996, which is more home-grown ("Long-term effects of acid rain: response and recovery of a forest ecosystem," Science 272[5259]: 244-6, 12 April 1996). The first author is our director and it’s on the long-term effects of acid rain, which has been a major research interest of Gene Likens and a number of other people who work here. For almost 40 years, Likens and his colleagues have worked at the Hubbard Brook Experimental Forest in New Hampshire. It’s called the Hubbard Brook Ecosystem Study. It looks both at long-term records of material budgets of small watersheds as well as experimental manipulations of those watersheds to understand the basic biogeochemistry. This paper is a review of the history of the effects of acid deposition in these forest ecosystems, and I think it’s highly cited, first of all, because it’s an excellent long-term record and long-term records are rare and, secondly, the topic is both of scientific interest and of management interest.

The next one down is a very different story. It’s by Pacala and Canham. Canham is at IES. The title is "Forest models defined by field measurements: estimation, error analysis and dynamics," (Ecol. Monogr., 66[1]: 1-43, Feb 1996). Canham and Pacala have worked for a long time on a model of forest dynamics, and this paper explains the model and how the model interacts with real data. Some models are built to make predictions, but they’re not nearly as tightly connected with the real data as this one is. The paper has been widely picked up by both forest ecologists and forest managers. Canham travels all around the world to help forest ecologists parameterize this model to work in their forests. This has led to some very interesting things. We have a series here, for instance, called Cary Conferences. Every other year we sponsor this international conference and the last topic we did was on models in ecology.

  Are there other factors that might explain the Institute’s upswing in citations?

In terms of citations, one thing is that we publish a lot. We basically add more than 100 papers a year to our publication list. One reason we can do that is we have light teaching responsibilities. We do close to full-time research here. So having experienced scientists publishing means you’re going to hit the mark with some frequency. Another factor is that the group here is quite willing to put their data in the context of other people’s data. We do a lot of synthesis, rather than just publishing a paper that says, "here are my data and only my data." The group here is quite willing to use other published information and data, and pull it together to make a more coherent story. Those papers combining primary data and data from the literature have a review-like quality and they get cited more and are also trusted more. A third factor is that the group is also quite collaborative with people from outside this institution. If something big is going on in ecology, it’s likely that one of our staff members is involved. And the last factor might be these long-term studies, like this Hubbard Brook Ecosystems Study. That was really the prototype for the National Science Foundation program on long-term ecological research. Hubbard Brook started without that program in mind, but its success gave NSF the idea to launch a network of long-term studies. At least, that was one impetus for NSF to do this. And long-term data in ecology are extremely valuable. If you have time-series data, you can do things no one else can do.

Let me describe it this way. When I was a graduate student, more than a decade ago, applied ecology was sort of a negative thing. It was what people did when they couldn’t get funding. Now applications of ecology are frontiers of the field. I think this has come about for a few reasons. For starters, ecology has demonstrated it can actually be useful sometimes, rather than just being an ivory tower type of science. And, secondly, the mood of the public has changed. There’s a lot more accountability, as with everywhere in science.

  Is there one over-riding philosophy that might describe how the Institute approaches the science of ecology?

Well, I think the science that stands the test for time is science that uses more than one approach. A colleague of mine, Steve Carpenter, made an analogy that ecosystem science is like the surface of a table and it’s supported by four legs. This is published as a chapter in one of our nine Cary Conferences Books (Pace and Groffman [eds.] 1998). You can argue about what these legs are and you can say that maybe there should be five. But each is a different approach. You have experimental studies. That’s pretty obvious: you jerk the system around and see what happens. The now-classic work on the effects of clear-cutting at Hubbard Brook falls into this category (Likens et al., 1970) as do some of the more recent experimental work here with lyme disease (Ostfeld et al., Science 1998). You also have comparative studies: you look at processes across systems and see how they differ. You can learn things about ecosystems, for instance, by plotting nitrate export among watersheds of the world (e.g. Caraco and Cole, Ambio 1999). Another leg is long-term studies: you follow a system or more than one system over time, as in the case of the Hubbard Brook Ecosystem Study or the work on the zebra mussel on the highly-cited list (Caraco et al., 1997). Those allow you to do analyses that you couldn’t otherwise do. The fourth leg would be modeling and theory. That could be one leg or two. But those models allow you to ask what the consequences are of your current assumptions and to ask about the trajectory of the system or about patterns among systems. It’s probably true for science in general and certainly for this science, that it benefits when at least two and hopefully more of those legs are being used at the same time. IES scientists have written a book about theory in ecology that is widely used (Pickett et al., 1994, Academic Press); the most recent Cary Conference examined the role of models in ecosystem science.

Well, Clive Jones has several papers on our highly-cited list for this decade. He and Steward Pickett stand out as the most theoretical spoke of our group. Rather than being really fundamentally interested in data collection or even reviewing existing data, those guys like to think about how things are put together. Clive is involved in the idea of organisms as ecosystem engineers. It is a very compelling concept, and his paper in Ecology ("Positive and negative effects of organisms as physical ecosystem engineers," Ecology 78[7]: 1946-57, October 1997) has been very, very widely picked up. I think people who are more bent on collecting data are now starting to think not just about the organism and what it needs, but about what it builds and how that has an effect on the landscape and on biogeochemical processes. It’s become a very hot topic.

  What is your prediction for how your field is going to change in the next 10 years?

I think funding for what you would call investigator-initiated science is going to get tighter and more competitive. Part of that is simply driven by the economy. Part is driven by a more centralized mission on the basis of the funding agencies, on trying to direct science from Washington rather than from scientists. You can argue about whether it’s good or bad. But it’s happening and it will continue to happen. I think there will probably be more push for work that cuts across traditional academic disciplines, because that trend has been going on for a while. But in all this, I really hope that there’s room for investigator-initiated science, because it’s what scientists do best, and what appears to be most cited.