In a paper published in Science in 1999, A. Townsend Peterson, Jorge Soberón and Víctor Sánchez-Cordero presented strong evidence for niche conservatism in 37 sister taxon pairs of mammals, birds and butterflies on either side of the Isthmus of Tehuantepec in Mexico. These results suggested that speciation initally took place along geographic dimensions and ecological separation evolved later. Eighteen years after the paper was published, I spoke to Townsend Peterson about his motivation to do this study, memories of field work and what we have learnt since about niche conservatism.
Citation: Peterson, A. T., Soberón, J., & Sánchez-Cordero, V. (1999). Conservatism of ecological niches in evolutionary time. Science, 285(5431), 1265-1267.
Date of interview: 6 January 2017 (via Skype)
Hari Sridhar: I wanted to start by asking you bit your motivation to do the work presented in this paper. A lot of your early work was on birds, and you’ve also done work on systematics and evolutionary relationships of individual species. What was your motivation, in relation to your previous research, to do this particular piece of work?
A. Townsend Peterson: Well, there’s a long evolution. Yeah, you’re right; my base of experience is in birds. My doctoral dissertation was on the systematics of jays. And that really got me involved in looking at use of environments and use of habitats across a species’ range. Once I got past the PhD, and into a job here at the University of Kansas, those geographic interests got stronger and stronger. I became very interested in the question of whether I can anticipate the distribution of a species from the points at which it has been recorded. That’s really what started what eventually became the paper we’re talking about? A colleague and I were sitting with a very, very early version of a geographic information system. And we were playing with points, which are the places where a particular species had occurred, and we were relating those not just to the geography but also to the environments. This paper is actually the second time I used the word niche in a in a publication. The first was in a very obscure book chapter, but it was about the same sort of thing, looking at the geography of the use of environments. That really was the genesis, where we started to look at why a species was where it was and why it wasn’t where it wasn’t, if that’s not too complicated.
HS: What’s the name of the colleague you just mentioned?
ATP: That was Jorge Soberón, who’s a prominent Mexican ecologist, and actually is now a professor here at the University of Kansas. Later in the process, when we were assembling data for the project, we brought in Víctor Sánchez-Cordero, as well, who’s another prominent Mexican scientist.
HS: Do you remember when this playing around with the date with Jorge Soberón happened?
ATP: It would have been in 1998, maybe six months earlier. Science is quite expedient about publishing once you get through their review process.
HS: Before I ask you more about the genesis of this piece of work, I’d like to step back a bit and ask you how you got interested in birds and also in topics of distributions and niche conservatism.
ATP: The interest in birds goes back to prehistory. Apparently, I was three or four and already fascinated with them. I had an older sister who was interested in birds, but perhaps not given as much freedom to be out in the field as I was. But she got me interested. The interest in geography comes from my dissertation, which was geographic sampling of a single species during my doctoral dissertation research. It’s an exercise that very few people do. You travel inch by inch across the range of a species and you look at what habitats it’s using and what is its environmental range, at one site, and another and another and another. It was like 66 sites where I worked with a single species. And that really teaches you some very exciting lessons.
HS: Were you working on the scrub jay?
ATP: Yes, exactly.
HS: How did you get interested in working in Mexico?
ATP: Partly because the jays that I got interested in for my dissertation were present across much of Mexico. And partly because I ended up falling in love with the country, and I made some very good friends in the country, who are my colleagues to this day, 30 years later, and ended up even marrying into a Mexican family. It’s kind of come to be a second home.
HS: You cite an earlier paper in Ibis for which you had collected a lot of the data that went into this paper. Was the work for both of these done around the same time?
ATP: Um, it was all kind of part of the same bits of thinking. The Ibis paper was essentially asking, where, across Mexico, do we have a lot of data, and where do we have a little data or no data? That was kind of the compilation phase where we were getting these occurrence data together. Later, we were trying to put those data to good use. And that’s when we started into the GIS work.
HS: Give us a sense of how this work actually happened? Did all three authors get together in the same place to discuss and do the analysis or was it mostly done remotely? How often did you meet when this work was being done?
ATP: A lot of the early work that I did in Mexico was kind of, painfully, at distance. Even long distance phone calls were expensive. We would make lists of questions and points to discuss and we’d have these speed chats by phone. But, actually, that year, we had quite the luxury that both Victor and Jorge and I were together here at the University of Kansas, because both of them were doing sabbaticals. And so we had more opportunity than usual to be together.
HS: This study was done in the Isthmus of Tehuantepec in southern Mexico. Is this an area you continue to work in? When did you last visit this place?
ATP: Well, the data on which the paper was based actually came from across all of Mexico. The barrier across which we were doing our analyses was the Isthmus of Tehuantepec. I worked there quite a bit in the 1990s and the 2000s. There’s a fascinating region called the Chimalapas. It’s a beautiful, beautiful half-million square kilometer region with quite intact rain forest and montane forest. It’s just a wonderful region to work. In recent years, I haven’t been working in the field as much in Mexico, simply because the security situation has gotten more and more difficult.
HS: At one point in the paper you say that you apply new tools and approaches to examine the question in birds, mammals and butterflies. What was the big advance, at that point in time, which allowed you to do this piece of work, if you had to pick one that really made a difference?
ATP: I’ll pick two. One was that we visualized the distribution of a species, not just in geographic dimensions, but also in environmental dimensions. This was not for the first time, but we were one of the very early groups to look at species distributions in two spaces that are linked. That’s one, but then the real insight was, oh, we’re not making a model of the distribution of the species, but rather we’re making a model of aspects of its ecological niche. That’s really the point that made the difference and made it such an exciting project.
HS: What about in terms of the methods used? Was this one of the first studies to use GARP (Genetic Algorithm for Rule‐set Prediction)?
ATP: It wasn’t the first. Actually, GARP goes all the way back to the 1980s, when the algorithm was developed, but not really applied to enough situations. That early phase was one of, kind of, taking this tool that showed some potential and applying it to a bunch of situations and testing it and exploring it.
HS: Had GARP been used to address these sorts of questions earlier?
ATP: It was developed by a man named David Stockwell. David was, I believe, at that point, associated with entities within Australia. His work, I think, was more applied to land use planning and conservation. What we did was one of the very first that that used the same tools but oriented it towards questions of ecology and evolution, rather than, I don’t want to say just management questions, but we were really looking at questions of ecology and evolution.
HS: Can we go over the acknowledgements to get a sense of how you knew these people and how they helped?
ATP: Sure.
HS: You thank R. Holt and E. Wiley for discussion.
ATP: Bob Holt and Ed Wiley were senior colleagues at the University of Kansas. I was just finishing up my assistant professor stage. These were two older colleagues with whom I consulted and debated. That’s why we don’t all just work at home these days. We come to the office because we can talk to our colleagues and debate and discuss ideas.
HS: The next person is A. Navarro-Sigüenza
ATP: Yes, that is Adolfo G. Navarro-Sigüenza. Adolfo is my oldest and closest colleague in Mexico. I’ve worked with him since 1988, so we’re at 29 years working together. He’s been a companion in working with Mexican birds from the very beginning.
HS: J. Llorente Bousquets
ATP: Jorge Llorente Bousquets is a close colleague of Jorge Soberón. They work together on questions of butterfly distributions across Mexico.
HS: The next name is DRB Stockwell. Did you get in touch with him to discuss these ideas?
ATP: I don’t know that we discuss these ideas specifically with him. He was very, very helpful in those early years in getting GARP running. In fact, he mounted it on a web application specifically for us, which really made the tool usable. But I’m not sure that we checked these ideas with him.
HS: RM Timm
ATP: Bob Timm was curator of mammals at the natural history museum here at the University of Kansas. Bob was another senior colleague and knew quite a bit about neotropical mammals. And so, again, a person with whom we consulted and discussed.
HS: H Benitez Diaz
ATP: Hesiquio Benitez Diaz. Hesiquio was the first Mexican student that I advised. I advised him to his to his undergraduate degree at a university in the state of Michoacán. He had been with me on a lot of the fieldwork that I was doing. And again, another colleague with whom we discussed ideas.
HS: Would you remember, you know, how long it took you to write the paper, and when and where you did most of the writing?
ATP: I definitely remember where; it was in my office complex here at the University of Kansas. I think getting from insight to data and analysis probably took us on the order of a couple of months; it was a fair number of species to analyse. But writing the paper was quite efficient, because Science papers are short, and once you have the ideas clear in your mind, they really don’t take very long to write.
HS: Were all the authors in the same place when the writing was happening? Did you take the lead in writing a first draft?
ATP: Yeah, we were all together, which made things quite a bit easier. I think, probably, we outlined the paper all together, but then I would imagine that I wrote a first draft. I don’t remember specifically.
HS: Was Science the first place you submitted this paper to?
ATP: It was.
HS: Did it have a smooth ride through peer review?
ATP: I remember we got a good set of challenging reviews, but it wasn’t months and months of work to take them into account. It was kind of ideal especially that early in my career. I was certainly super, super motivated, and the idea of getting a paper into Science was very attractive. But it wasn’t a terrible amount of work either. It was a good balance.
HS: How was the paper received in academia when it was published?
ATP: Academia moves slowly. Over the succeeding 2-3-4 years, it provoked quite a bit of debate. There were research groups that agreed and got excited about those ideas, and there were research groups that were seeing other things and disagreed. That’s how science advances, so it’s all good.
HS: Are you surprised by the kind of impact it’s had on the field? At the time when you were doing the work did you anticipate its impact?
ATP: You don’t waste people’s time sending on a manuscript to Science if you’re not pretty excited about it. As soon as we came to the basic insight, and the insight led us to structure a test, and the test was affirmative that that we were seeing niche conservatism between sister species, we were pretty excited about it. I don’t know that we could have put a bet on how many citations it would have in 2017, but I think we were at least convinced that it was an important contribution.
HS: Do you have a sense of what it mostly gets cited for?
ATP: This field that we can now call ecological niche modeling or species distribution modeling depends on the assumption that these niches that are being modeled are relatively constant. If the physiological characteristics and, therefore, the environmental distributional potential of a species are very plastic, then we’ll get no useful predictions out of these models. And so, I think a lot of the citations come from the need of people to bolster the idea that, well, I have this set of populations or species that are closely related, and I really would like to trust that they will share similar niche characteristics.
HS: Did this paper have any kind of direct impact on your career?
ATP: Well, um, let’s see. I had a permanent job. That’s very fortunate in academia. I started my job here at the University of Kansas in 1993. And then we have a six or seven year period, which is your assistant professorship, and at the end of those six or seven years you were evaluated for tenure. I was not yet given a permanent position when we were writing this paper. I think this paper came out about four months before I came up for tenure. A paper in Science certainly helps; it doesn’t hurt, right?
HS: In hindsight, would you say that paper marked an important moment in your career, in terms of the impact it had on the future course of your research?
ATP: Without a doubt, which is to say I was very interested in, and still am very interested in, lots of aspects of ecology and biogeography and evolution, but this paper laid the basis for a whole new set of insights about, what we could call “distributional ecology”. This definitely opened doors to a whole new set of insights for me, regardless of for the rest of the community.
HS: Today, would you say that the main conclusions of this paper still hold true, more or less?
ATP: That’s an interesting question. As I said earlier, some of the people in the broader community liked those ideas, and got very interested in those ideas, and some people were getting the opposite results. That led to quite a series of debates with these other research groups. The interesting thing came about six years after we published the paper. A research group from California – Michael Turelli, who’s a population geneticist, Richard Glor and Daniel Warren – first asked me and my colleagues for the data that had gone into the paper and did the same with one of the research labs that was getting the opposite results. And they pointed out – it’s almost a bit embarrassing – but they pointed out that, let’s distill it down and simplify: these two groups were arguing, I see niche conservatism, and the other was saying, I never see niche conservatism, right. Well, it turns out, we were arguing about different points. It’s like my saying, the sky is blue, and you’re saying, no, the tree is green. This group took data from each of the two research groups and pointed out that the two research groups that were getting opposite conclusions were testing different null hypotheses. Basically, my group was testing the idea that niches were surprisingly similar between closely related species, and the other research group was testing the idea that niches were identical. But niches could be significantly non-identical but very similar. That then led to the question of which of those two things is more meaningful? And, you know, some of my colleagues might disagree, but I would argue pretty strenuously that surprising similarity is more important than being identical. And so I think, with those modifications, the ideas are still quite current and still supported quite well, with much better tools and much better insight and a much improved conceptual framework.
HS: If you were to redo this study today, what would you do differently?
ATP: That’s an interesting question. I think that study was very much a product of its moment, For example, we didn’t have many, many robust phylogenetic hypotheses or genealogies of bird species. So we used the trick of looking at sister species. We didn’t have many algorithms for estimating niches, and we didn’t have many environmental dimensions. If that study had never existed, and I were to be doing it right now, it would probably take me about five times longer to do it, because the methodology has gotten so complex and nuanced. But the basic question of comparing ecological niches between closely related species wouldn’t change very much.
HS: Have you gone back to this particular system and looked at this question again with more data? Do you now have more data on other species from this particular area?
ATP: I think it was 37 pairs of species that we looked at. I don’t think that that we could add many species to that list. There’s certainly more data in existence about their occurrences. But really, to be honest, the Warren group, when they did that reanalysis, they did a very current and very deep and thorough reanalysis of our data. I really haven’t gone back to this system. I’ve certainly replicated the study in other systems and other taxa. But, again, a lot of credit to the Warren, Glor and Turelli group for doing that work.
HS: When you when you look at other systems and other taxa, in general, would you say that there is support for niche conservatism, from the work that you have done and the work that others have done?
ATP: Yeah, definitely, there’s been, in the broadest sense, in general, there’s been quite a bit of support for what we found in 1998 and 1999. We also have very much focused on finding the exceptions. Those, I would say, have been quite the exciting things, to find the exceptions and, and focus more on where is conservatism not dominant.
HS: Towards the end of the paper, you say, “An untested question is whether the observed conservatism results from active constraint (stabilizing selection) or whether it reflects the absence of additive genetic variation in niche-related traits []. Similarly, our analysis does not eliminate the possibility of niches of both members of species pairs responding in parallel to broad-scale environmental changes. Subsequent to this paper, were these ideas tested further?
ATP: Certainly. They’ve been visited and revisited, perhaps not in big overviews, but more either in theoretical studies or in individual taxa. I’d say there’s support for perhaps all of those ideas. Certainly, one of the people who led this bit of theory was Bob Holt. He and colleagues looked at, in theory, if a species niche outlines its distributional potential, and outside of its niche its reproduction is not sufficient to maintain populations, then how does the species explore outside of its niche? And they came to a very interesting insight that it’s very difficult for a niche to expand. So, that was interesting. But then there have been some experimental studies, especially with plants, where they’ve shown that that the evolutionary potential – this is the question of additive genetic variation – the exclusionary potential of a species is not sufficient to allow the species to adapt to climate change as climate change is happening. So, I’d say, all of those ideas have been looked at, and I’d also say we’re probably not yet to a consensus about what is a general property.
HS: What about for the specific pairs of species that you tested? Have any of these species been studied in relation to these aspects, in terms of the availability of additive genetic variation or whether there has been environmental change in ranges of both members of a pair?
ATP: I don’t think with those specific taxa, simply because those are tropical taxa, not the sort of species on which you usually see detailed, intensive study.
HS: In the last lines of the paper, you say, “Finding general conservatism in ecological niches opens the door to phylogenetic studies of niche evolution, comparative evaluations of conditions under which niche conservatism breaks down, construction of predictive distributional models, and numerous other applications to questions in biogeography (estimates of α and β diversity, centers of endemism), biodiversity (foci of species diversity),and conservation biology (development of conservation prioritizations). Today, 17 years later, could you reflect on how much these questions have formed the focus of your own research, and in general, how much attention has been paid to these aspects?
ATP: Well, I’d say that was a pretty good prediction, which is to say this general framework of niche modeling has become quite a burgeoning area of research. It’s become very, very active, specifically, the phylogenetic work. That’s a big area of concentration for me right now. I’d say that is probably the area where there are some of the bigger challenges waiting still. I think a lot of the work going on in that area right now is actually not correct. We’re still kind of maneuvering around and trying to get it right before we settle on a final methodology.
HS: Have you ever read the paper after it was published?
ATP: Yes, of course, especially when the Warren et al. work was going on. You really do have to go back to your old work because your own thinking evolves, and you don’t always remember, you know, what was I thinking and writing in 1998-1999, as opposed to 2000 or 2004.
HS: When was the last time you visited this paper?
ATP: I’ve certainly pulled it up on my screen and checked some detail in the last couple of months. But reading it in great detail, reading it till the end, was probably four or five years ago.
HS: Most of the people I’ve interviewed have said that they’ve never read the paper from start to end, after it was published.
ATP: Maybe, I’m just afraid that I’m not stating accurately what I’m remembering. If I’m going to say something, I want to say the right thing.
HS: Is the way you write today very different from this paper?
ATP: Probably not markedly different in style. But I think that, certainly, I am less focused on publishing in Science! Simply because, at this stage in my career, my focus is more on publishing bigger comprehensive papers rather than the stripped down sort of papers that you do when for journals like Science.
HS: Would you consider this as one of your favorite papers?
ATP: Um, it was certainly a fun period of time in my career, simply because I was learning a lot. I’d say the first paper that you do in an area, or in a field, is far more fun than the 20th or 30th, just because you’re learning. And that makes a big, big difference and in making things exciting and getting you animated about a paper.
HS: What would you say to a student who is about to read this paper today? Would you guide the reading in any way? Would you point them to other papers that they should read along with this? Would you add any caveats to keep in mind when reading this paper?
ATP: I don’t know so much about caveats. But I think that paper was definitely a product of its time context. And so, there are pretty clear references in there to different processes of speciation – vicariant speciation versus ecological speciation. I think, sometimes that, those debates, which were very current at the time, may not be so current now. So, it’d be very useful for somebody reading this paper, to be familiar with, you know, what the cladistic biogeographic school was doing and what the Endler Group, which was pushing ecological speciation at that time, was doing. It would help to have that context of what were the big issues and the big discussions at that time. That would be the main point.
0 Comments