Reflections on Papers Past: Revisiting Roemer et al. 2002

In 2002, Gary RoemerJosh Donlan and Franck Courchamp published a paper in PNAS showing how the introduction of pigs had reshaped an island’s foodweb and caused the decline of an endemic fox. Fourteen years after the study was published, I spoke to Gary Roemer about the making of this paper, and what has happened to the foodweb since.

Citation: Roemer, G. W., Donlan, C. J., & Courchamp, F. (2002). Golden eagles, feral pigs, and insular carnivores: how exotic species turn native predators into prey. Proceedings of the National Academy of Sciences 99: 791-796.

Date of interview:  21st July 2016 (on Skype)


Gary Roemer rappelling to a Golden Eagle nest (© Brian Latta)

Hari Sridhar: What motivated you to do this study?

Gary Roemer: Well, actually, it was serendipitous. I went out to the islands to study the behavioural ecology and genetic structure of the island fox. And while I was there, I started to see some mortalities of foxes. I had worked with bald eagles in the past, and had figured that golden eagles, which were irregular visitors to the island, were taking foxes every now and then. And then I started seeing a steep decline in the foxes while I was there, so I just happened to be at the right place, at the right time, to be able to identify what was going on.


A talon hole in the skull of an eagle-killed fox (© Gary Roemer)

HS: So your PhD, to start with, didn’t have anything to do with this set of interactions? It was focused only on the island fox?

GR: Exactly. It was going to be a study on their social organisation, dispersal and genetic patterns, foraging ecology etc., you know, sort of an autecological view of the island fox. And all this happened while I was there.

HS: How did this group of three authors come together?

GR: Well, I had published a paper in 2001 in Animal Conservation based on this. I had used Franck Courchamp’s model in that paper.  And then – I think it was either 2000 or 2001, I have got to look back – I gave a presentation on that work at the Society for Conservation Biology meeting in Missoula. Josh came up after the talk and, you know, started rapping with me and we kind of hit it off and became friends. And then a little bit later I was talking to him over the phone about an aspect of the research that I wanted to try and get a handle on, which was to evaluate the food habits of the golden eagles. Josh said: “well, we could use stable isotopes to get at a broad-scale look at their food habits”. So then, he and I started conversing about that and felt Franck was a better modeller than either one of us, so we asked him to be involved. That’s how the team came together.

HS: Did you do all the fieldwork and the data collection?

GR: Yeah, that’s correct, myself and various technicians. Josh came out once to help me collect tissue samples from feral pigs.

HS: How did you build upon the Animal Conservation paper for this paper? Was it through the modelling that Franck Courchamp did?

GR: Yes. There was some modelling in the Animal Conservation paper as well, but the model was developed further here. The structure of the models were similar, but the model in the PNAS paper was more elaborate, and we included additional parameters that would make the model more realistic. We also included skunks, which were a competitor of the fox. Not just the pigs. The original model in the Animal Conservation paper was just trying to show that pigs were probably one of the more important prey animals that was driving this interaction.  Then, the stable isotope information was also helpful in determining certain species interaction factors that we used in the PNASmodel.

HS: Did the collaboration for this paper start after the Animal Conservation paper was published?

GR: Yes, I think the Animal Conservation paper was already published or at least in review. I had already put that paper together, because that’s also what my presentation at the Society for Conservation Biology meeting was based on. And as you know, the whole process takes a long time, so that process had been ongoing for probably a couple of years before we got the PNAS paper together.

HS: How did the collaboration actually work? Did the three of you ever meet? Was it mostly over email?

GR: Yeah, it was mostly over email and on Skype a couple of occasions. There were a couple of phone calls as well. Josh and I got together more frequently, but Franck was over in France, so with him it was always remotely. As a matter of fact, I had never met Franck till about – I am trying to remember the year – I think it was 2004 or 2005 at the Society for Conservation Biology meeting down in Brazil.

HS: Was most of the writing for this paper done by you?

GR: Yes, Josh and I were the principal writers.

HS: Do you remember how long the writing took?

GR: Originally, we put a paper together for Science. That was a shorter version but that didn’t make it. And then we communicated with Dr. Jim Estes. I don’t know if you are familiar with Dr. Estes; he just wrote a book called Serendipity about his experiences in biology. But anyway, Jim suggested that we submit to PNAS and offered to help us communicate it, if you will.

The whole process went pretty quickly actually. I would say within six months we had contacted each other, done the analysis, written the paper, got bounced, talked to Jim – it was all relatively quick. We submitted the original manuscript on August 11, 2001, received comments back and submitted a revision on October 16, 2001, and the paper was published on January 22, 2002.

HS: But on the paper it says this was a direct submission to PNAS?

GR: What happened was Jim communicated with Dr. Harold Mooney who is a National Academy Member; Dr. Mooney gave us some more advice on how to craft the paper. We had submitted the paper directly, but Jim had suggested that Dr. Mooney take a look at the paper because he thought it was a really nice study and that Dr. Mooney could give us some good advice. Jim also did a real nice job helping us edit the paper and giving us suggestions on how to improve it.

HS: The paper is full of really interesting natural history information. Was all that already known or did it come from your PhD?

GR: Well, certainly for the foxes, a lot of that was from my PhD work as well as work that I had done previously. I had been working with foxes since 1988. About the eagle and the skunk, a lot of that was just from my experiences working with the particular species or related species. I guess I would consider myself more of a scientific naturalist, in general. Natural history is important to me.

HS: Yes, that is so evident. The paper really stands out because of the natural history in it. Did you have help in field or did you do most of this work on your own?

GR: No, I always had one technician at any time during the project. There were various technicians that worked on the project, and one or another one of them worked with me during the entire project, they did everything that I did.  The technicians included my cousin, Tom Roemer; a really good friend, Dr. Paula White; Rachel Wolstenholme, Dr. Debra Woollett, and Dr. Jeff Howarth. However, Paula, Deb and Jeff were not doctors at the time! I taught most of them how to handle the foxes, how to track the foxes etc. Except for Paula, who had as much or more experience than I, primarily with Arctic foxes. So, we worked as a 2-person team.

HS: Was most of your work on Santa Cruz?

GR: Well, my dissertation focused on Santa Cruz, but I worked on all the islands where there were foxes, from 1988 to about 2001. That was the last time I was in the field working on foxes specifically.


Santa Cruz Island (© Gary Roemer)

HS: You cite a personal communication from T. Coonan, for some information from Santa Rosa. Was T. Coonan also working on foxes on other islands?

GR: Yeah, so what happened was, originally I was going to be doing my work on Santa Rosa Island, and I was partially funded by the National Park Service. But because of political situations on Santa Rosa Island the park superintendent sort of nixed that project. I had applied for money and got funding through the park service to do the work, Coonan was one of the park service personnel who really championed the project. He convinced his administration to allow the transfer of the funds to Santa Cruz Island, because the work on Santa Cruz was germane to the rest of the island fox populations. And at that time, the National Park Service oversaw San Miguel Island, Santa Rosa Island and about a third of Santa Cruz Island. So when I started the project on Santa Cruz, I trained two biologists from the National Park Service – Tim Coonan and Cathy Schwemm to start a grid-based mark-recapture analysis on San Miguel Island. Shortly after beginning my work on Santa Cruz, I started seeing declines in the fox population there, this trend was based on population estimates derived from two grids and various transects where we were trapping foxes. So my estimates of population size were declining, and then of course I was also seeing all these mortalities with radio-collared foxes. Over on San Miguel Island, where they were running three trapping grids, they started to see declines in foxes there as well. But we had no information from Santa Rosa Island. And so I convinced Tim to send me, and another colleague and NPS employee, Keith Rutz and a colleague of mine Chris Starbird, the three of us went to Santa Rosa Island to basically do a sort of reconnaissance mission; a short trapping mission to see what we might find there. And when we went there we only caught nine foxes over the course of about 80 trap nights. And normally, if the fox population had been very abundant, over 80 trap nights, we would have caught anywhere from about 20 – 40 animals. And we had put the transects in a variety of different places around the island over the course of a few days. We put 10 traps here and 10 traps there and we moved them around and then we looked for sign and we found hardly any sign and so we convinced the park that the fox population on Santa Rosa Island had probably declined as well. When the Park Service finally started the captive breeding program on Santa Rosa Island, they only caught 15 individuals to start it. So that’s where that information was relevant.

HS: Did the paper attract a lot of attention when published?

GR: Oh yeah it did. It was recognised as a really interesting study with respect to how exotic species can cause native species endangerment, and one that uncovered a mechanism that was contributing to a decline of an endemic species. I got a lot of calls and there was some press and I have a folder full of things about where it was showcased in different articles. Then I had a number of requests for speaking engagements, especially at universities. So yes, it created quite a buzz. It was kind of neat you know. I mean it was sad for the foxes, but it was a neat piece of science and it jump-started my career.

HS: When I emailed you requesting an interview you said your paper was “more infamous than famous”. Why was that?

GR: [Laughs] Well, I think because at the time maybe I wasn’t as politically astute as maybe I am now, although some would say I haven’t changed (!), and it resulted eventually in me kind of being blackballed by the National Park Service. And I can send you a couple of other papers that sort of outline the scenario that unfolded with this study. They were written by Josh and I, and we published them in Endangered Species Update (Part 1Part 2). And maybe after I send you those if you want to have another discussion we can, but basically I really upset the park service with the things that I did to try to save the foxes.

HS: What kind of impact did this paper have on your career and the future course of your research?

GR: It got me my job, for sure. I hope the quality of the science was important, but certainly, I think, the publicity and whatnot helped me get my job. At the time in 2001, when I had the interview here at New Mexico State University, I presented that work, and I think that helped solidify the offer for my position.

In answer to the second part of your question, well, no, I wouldn’t say that. Certainly, I was interested in apparent competition and the effects of predation, but once I got here to New Mexico State, I started doing other things that were related to population demography of south-western taxa. So it really didn’t, although it probably did inform me to some degree on a recent paper that we published in Ecology on intraguild predation.

HS: You haven’t worked on this particular system after this paper?

GR: Essentially, well I did fieldwork until 2001, and we have published some other papers since then that were related to that. I don’t know if you are familiar with a paper by Vicky Bakker that was in Ecological Monographs. That was a real nice piece of work, I was really happy to be involved in that. So there were a few things that came after, but after my fieldwork sort of stopped in 2001, anything that I published afterward utilized data that I had collected in the past. I am not actively working with island foxes now.

HS: Did this paper have an impact in terms of how people think about introduced species and apparent competition. Have people gone out and looked for more examples?

GR: Yes, I think it did or perhaps I am just noticing studies that are similar. My paper has been fairly widely cited. You know Dr. Robert Holt was the original individual that came up with the theoretical underpinnings for apparent competition. I think that because my study dealt with some charismatic species, it helped to elevate the awareness of this mechanism, especially with respect to indirect impacts wrought by introduced species. But certainly, the idea of apparent competition had been around for quite sometime. And, you know, Franck’s work sort of recoined it – he used a different term for apparent competition in his paper – hyperpredation. I think Franck’s work originally tied the mechanism to introduced species, that’s sort of what alerted me to the potential interaction, because when I read Francks’ work on how rabbits could influence cats, which could then negatively influence seabirds, that sort of was a corollary for me with regard to what was going on with pigs, eagles and foxes. I think our paper may have at least made other scientists and managers a little bit more aware of the potential role of that process, especially with respect to introduced species.

HS: Do you have a sense of what this paper mostly gets cited for?

GR: In general, as an example of how introduced species can influence native species, in more of an indirect way rather than a direct way.

HS: In general, would you say the citations are appropriate, i.e. for things that are in the paper?

GR: Yes, with respect to this whole idea of how apparent competition is a process that’s important for us to explore in the decline of native species.

HS: Did the study lead to conservation action, e.g. the translocations?

GR: Oh yes, the study itself formed the biological underpinnings for the recovery actions implemented by both the National Park Service and The Nature Conservancy to start to restore the islands. As a matter of fact, this year, the island fox was delisted, at least three of the four subspecies were delisted – the three northern island subspecies where apparent competition had occurred – were removed from the Endangered Species Act (ESA). Also, one of the subspecies on Santa Catalina Island, which had declined as a consequence of a canine distemper outbreak, is going to be downgraded from Endangered to Threatened. They are showcasing this as an ESA success story, because it’s the fastest recovery of an endangered mammal in the history of the ESA.

HS: In the model you built you made some assumptions about the system? Have those assumptions been shown to be true by future work? Is there research work happening on this system by other people?

GR: Well, the other main piece of work that came out was the work by Dr. Vicky Bakker, which we did collaboratively. That analysis was done primarily by Vicky and Dr. Dan Doak and published in Ecological Monographs. I don’t know if you have a copy of that; I can send it to you. But that was a really in-depth study, and there’s a bunch of natural history in there as well, which I had contributed to. That paper basically confirmed what we did in the PNAS and Animal Conservation papers, but extended our earlier work via a population viability analysis (PVA) that identified population drivers, predicted extinction risk, and also suggested monitoring strategies. And all of this, in part, contributed to the process that the Park Service, and then eventually the United States Fish and Wildlife Service (USFWS) in conjunction with The Nature Conservancy, put together to restore the foxes. So we ended up.. I guess I should say they ended up removing pigs, translocating golden eagles, reintroducing bald eagles, captive breeding and releasing island foxes, and then establishing a monitoring programme whereby they are monitoring fox populations and also looking at causes of mortality to ensure that golden eagles don’t come back and nest on the islands again.

HS: Is the island now free of golden eagles and pigs?

GR: Yes. Well, there still maybe eagles that come out and visit the islands periodically.  They always used to. It wasn’t abnormal to see golden eagles on the island, especially during the non-breeding season. I guess the pigs were removed by about 2006. And then the bald eagles were reintroduced, and the bald eagle population is doing very well. They are breeding there and because they are territorial they act as a deterrent to the golden eagle. They’ll tend to keep other raptors out of their territories. So all that together was something that probably facilitated the recovery.


From L to R: Ron Jackman, Gary Roemer, Brian Latta, Primo (the first Golden Eagle captured on the island) and Chris Kuntzsch (© Sam Spaulding)

HS: In the paper you say that disease is probably not a very important factor. Is that still your view? I am asking because you mentioned canine distemper a short while ago.

GR: Well disease was important on Santa Catalina Island. But on the northern Channel Islands – that would be Santa Cruz, Santa Rosa and San Miguel – hyperpredation was responsible for the declines. On Catalina Island, which is a fair amount south, there had been an introduction of canine distemper virus, which wiped out about 3/4ths of the population. And the folks who were working there recognised what was going on, implemented a vaccination programme, a captive breeding programme, and a translocation and release programme, and recovered the fox population. So the idea that diseases can impact island fox populations is definitely real, and perhaps even more important given the fact that island foxes tend to be less genetically variable than mainland populations.

HS: This is a minor point – you say “Foxes and skunks were live-captured on Santa Cruz Island on two grids (13 km apart) from 1993 to 1999 with a hiatus in 1994”. Can you tell us why there was a break in 1994?

GR: On Santa Cruz specifically?

HR: Yes.

GR: Well, the hiatus wasn’t throughout the entire year. I am trying to remember what exactly happened there. I think what happened was that I was focusing not on the telemetry aspects but on the demography aspects, and I was unable to run the grids that particular year. I have to go back and look again to recall, but I think that’s what happened. Let me pull up the paper, but I’m pretty sure that’s where the data glitch was. Let’s see. It’s a long time since I looked at this paper. Yes, so if you look at Figure 1, you can see I don’t have any demographic data for 1994. That data is from the grids that we ran – these were mark-recapture grids – so from about mid-1993 to the end of 1995, I was doing telemetry and running these demographic grids during the summer. In 1994 I elected to not run the grids and to focus my efforts on the telemetry. I was still there doing research, but I thought that evaluating the mortality and keeping a closer eye on the foxes was more important, because the grid trapping is a considerable effort and takes at least a month’s worth of time to run.

HS: Did the telemetry data go into the Animal Conservation paper?

GR: Yes, it’s in there. In that paper I did some matrix population modelling, and the survivorship values come from the telemetry data. I also showed changes in survival across islands based on telemetry data. Then I have another paper – I don’t know if you saw this one – in Journal of Zoology on the fox’s behavioural ecology, on what happens to their social organisation as individuals are killed by eagles.

HS: You just mentioned that it was a long time since you read this paper. When was the last time you read this?

GR: Oh geez I don’t know! It was probably a decade ago. I haven’t really looked at the paper in a while. I’m doing other things now, and I haven’t sort of used this approach per se since then. Although, I have looked at the Animal Conservation paper, in the context of some matrix population modelling stuff that I was doing, to look at what I did there and how I can do things differently.

HS: I would like to go over the list of people you acknowledged, to find out a little more about their roles.

GR: Sure. Let me look at it. Okay, Lyndal Laughrin. I believe he is still the reserve manager. The University of California has a number of reserves that are located throughout California. I think there are somewhere around 33-35 of them. And one of them is on Santa Cruz Island. These are places where both researchers and educators can go. Mostly people do field trips. They go to the island for a weekend or 10 days, something like that. But also, long-term research emanates out of these reserves. These reserves had vehicles and a place to stay, so I would stay at my remote field site for 20 days every month. I lived on the island for 2.5 years and Lyndal supported me in many ways. I can’t count the ways he helped me. Actually, Lyndal did his PhD on the foxes as well.

HS: You have already told me how Coonan helped. What about G. Koch and B. Hungate?

GR: Those were two colleagues and mentors of Josh who helped us with the isotope analysis. In particular, with making sure we were interpreting things correctly. Both of them were previously from Northern Arizona University, where Josh finished his undergraduate degree.

HS: D. Garcelon, who provided skunk serum samples?

GR: He was a colleague who I had worked with on island foxes for many years. He had come out to Santa Cruz Island and sampled skunks out there, and I had helped him trap and draw blood from skunks for a genetic analysis. Subsequently, the serum samples were used in the isotope analysis.

HS: T. Gorton, who assisted with the figures?

GR: Yes, that was another colleague of Josh’s, who gave us some ideas on how we could improve our figures, etc. Once we had put the paper together, we had discussions with individuals before the paper was finalised, and T. Gorton gave suggestions on how we could improve the figures.

HS: And then there are a number of people who commented on the paper.

GR: Yes, we sent the paper out before the submission to Science, and then we sent the paper out before the PNAS submission. We tried to get as much feedback on the paper, because we definitely wanted to try and get it published in PNAS. All those individuals made valuable comments to improve the quality of the paper.

HS: You acknowledge numerous sources of funding. Were all of these given specifically for this work?

GR: Well it was in conjunction with my dissertation research originally. Everything supported my dissertation research. So I wouldn’t say that funding was specifically to study this interaction. Rather, the funding was for doing the ecological study of the island foxes and then this research emanated from that.

HS: What would you say to a student who is about to read this paper today? What should he or she take away from it?

GR: Well, first, that sometimes when you go into a study, other things may surface that you should pay attention to. In my case, I think because of my previous experience, I was able to recognise what was going on, and then because of the observations that I made while I was on the island, it sort of made me figure out what the mechanism was. I’ll just digress a bit to give you an example. We knew that eagles were killing foxes, but there was one day when I was driving into my study area, and I saw a golden eagle take off from a hillside and I noticed it had a full crop.  It had just fed. So I stopped the truck and I started scanning the hillside to see if I could see any remains. I spotted some blood and what looked like a small carcass. So I grabbed my … I didn’t grab my camera , this is where I made an error – but, anyway, I hoofed out there  – it was quite a way, it was several 100 yards  – and when I got there, I saw a piglet. The golden eagle had killed and fed on a piglet. I think that was my ‘aha’ moment. I realised that while the eagles were killing foxes, that wasn’t the only thing that was supporting the eagle population. The piglets are really important for supporting the eagle population. So, by way of advice to students, I would say that you should always be observing nature and shouldn’t be afraid to make inferences from your observations, or follow up on those inferences, even if those weren’t things you originally planned to do.

Then the other thing I think that’s really important is to try and ground your research in ecological theory. My background is real big in natural history, and wildlife biology in general, but I found that the most interesting way to advance science, or contribute to science, is to look at how theory relates to the work that you do. Subsequent to this study, I have also tried to set up studies with specific ecological theory in mind. I am an empiricist, not a theorist, but I try to collect data that might address certain theories. I always have theory in mind.

I guess the third thing would be: serendipity happens. Don’t ignore it, but, instead, take advantage of it.

Finally: collaborate. You know, some people – I was certainly guilty of this when I was young as well – get proprietary with their work and they want to get fully credited for the work they do. But the work can be much richer and much better when you collaborate with folks, and then, as a consequence of that collaboration, you end up learning more and you end up becoming a better biologist, a better scientist.

So those are probably the four things I would ask students to be aware of as they proceed along in their career.

HS: Would you count this as one of your favourites, among all the papers you have written?

GR: Yes, it is. More recently, another paper I really like is the one in Ecology on intra-guild predation. I thought my graduate student, myself and one of my colleagues did a really nice job there. There are a few other papers too. I really like my Animal Conservation paper and also the paper on the behavioural ecology of the foxes, which was published in the Journal of Zoology. I think when I look back upon my scientific career now – I still hope I have a little more to go  – I have to say that my dissertation experience was one of the most quality research experiences that I have ever had. I was living on Santa Cruz Island for 2.5 years and I was completely focused on doing field work and science.  It was just a fantastic experience for me. I really look back on that time fondly, and it certainly enabled me to have the career that I have today.