In 1992, Keith Hobson and Harold Welch published a paper in Marine Ecology Progress Series in which they use isotopic analysis to show that the food web in the Barrow Strait–Lancaster Sound area in the Arctic consists of five trophic levels. This was one of the early studies that used isotopes to determine trophic levels. Twenty-four years after the paper was published, I asked Keith Hobson (with inputs from Harold Welch) about the making of this paper and what we have learnt since about the food web in this area.
Citation: Hobson, K. A., & Welch, H. E. (1992). Determination of trophic relationships within a high Arctic marine food web using δ 13 C and δ 15 N analysis. Marine Ecology Progress Series, 9-18.
Date of interview: Questions sent by email on 18th July 206; responses received by email on 14th August 2016
Hari Sridhar: Correct me if I’m wrong, but this seems to be only your second paper on food webs. What got you interested in food webs at this stage in your career? What was the motivation to do this study?
Keith Hobson: Our motivation originally was to study ecological segregation among seabirds in the high Arctic. We realized that stable isotopes could be used for this purpose, but that an understanding of the stable isotope values in many of the food web components would be needed to decipher the seabird isotope data. So, the isotope technique forced me to take a full food web approach from primary production to seabirds. Of course, while involved in this aspect, adding marine mammals was an obvious opportunity and so we ended up studying phytoplankton to polar bears.
HS: In the paper you say “this area has received considerable attention from marine ecologists owing to its projected use as a transportation corridor and the potential development of the region’s hydrocarbon and mineral reserves”. Was this also part of your motivation for this study? Has the region been put to use for transportation and resource extraction, as was expected then?
KH: I have always been driven by a strong conservation ethic and remain deeply concerned about the future of nature on our planet. The Lancaster Sound region of the Canadian high Arctic has always been a focus of potential resource extraction and, now that climate change is happening rapidly, the Northwest Passage ocean route between the Atlantic and Pacific will become a reality with all of its attendant exploration and resource development. So, while major development has not occurred, likely due to low oil prices etc., the eventual development in the region is more likely now than ever.
HS: This paper is “Contribution No 17 from the Resolute Marine Laboratory”. Does this laboratory still exist?
KH: Unfortunately not. It was closed not long after my own work there. Harold Welch was the Director of that facility and he might be able to tell you how many publications came out from this lab. [Harold Welch adds: I have no idea how many papers came out of the Resolute lab but around 25 would be my guess.]
HS: How did the collaboration between you and Professor Harold Welch come about? What were your respective roles in this project? Did you continue to work with Professor Welch after this paper?
KH: Dr. Welch ran the laboratory for the Department of Fisheries and Oceans (DFO) and I was allowed to use the facility as a logistical base for my PhD thesis with the University of Saskatchewan. Dr. Welch was also an expert in marine food webs in the High Arctic. I have continued to work with Dr. Welch over the years through his retirement about 12 years ago. We are good friends and colleagues.
HS: Can you give us a sense of the fieldwork – what was a typical day in field like, where did you stay, how many people were involved in sample collection, did you set up a lab in field etc.?
KH: The laboratory also provided accommodation and was located in the small hamlet of Resolute Bay. From that logistical base, our food web sampling was done from small powerboats locally and a larger research vessel the “Ogak” which is Inuit for “Arctic Cod”. Those sampling trips could be days to weeks. The seabird work was conducted primarily from remote field camps, from tents, and could last for weeks to months. Transport to those camps was by helicopter or Twin Otter. So, much of the work was logistically very challenging and involved lots of concerns about weather and polar bears!
[Harold Welch adds: The Arctic can be an unforgiving place. Keith’s supervisor, Malcolm Ramsay, died tragically in a helicopter crash, along with another close friend and colleague. Subsequently, Marty Bergmann, who learned his stuff as my graduate student, died in another plane crash in Resolute, after becoming head of the Polar Continental Shelf Program (PCSP) that supports Arctic research.
Once, Keith, I and two others were returning in a seven meter boat, from Devon Island about sixty miles away, when the fog rolled in. We navigated through the ice as best we could and finally, just about out of gas, we drove up onto the sea ice to spend the night, without a clue as to where we were. But the fog lifted several meters and I recognized an island where I’d built an iglu, and we made it home on fumes that evening. It was routine, but it made a big impression on Keith.]
HS: If you did this study again today would do it differently, given the development in technology and analytical techniques?
KH: I do not think we could improve much on the field sampling, but of course more money and resources could always lead to more samples. However, the type of analyses conducted at the isotope lab would likely change, and now include also compound specific isotope analyses of amino acids and fatty acids in addition to the bulk tissues (muscle, feathers etc.) we looked at.
HS: How long did the writing of this paper take? Who did most of the writing? During the writing phase, did you and Professor Welch meet often or were discussions mostly over the phone?
KH: I did most of the writing because this work formed a substantial part of my PhD thesis. That likely took about two months, once all the isotope data was secured. However, Dr. Welch provided valuable expert advice and we discussed the various drafts of the paper primarily in person. After my PhD, I spent 6 months with Dr. Welch at the Freshwater Institute in Winnipeg, as a postdoc.
HS: Where did you do the writing – in your field station/ in your office/ in the university/ at home etc.? Do you have a writing routine?
KH: At that time, most of the writing was done at my office in the University of Saskatchewan. I will write anywhere that is quiet and away from distractions.
HS: How were the figures for this paper drawn? Who drew them?
KH: In those days, I did not have access of the wonderful graphics programs available today. So, the graphics are crude by today’s standards. However, I prepared all figures with assistance from a Departmental graphics person (those positions likely do not exist anymore or are rare).
HS: Some questions about the Acknowledgements: what was the Polar Continental Shelf Project? Who was M.A. Ramsay, who obtained a grant for this project? Can you tell us how you knew each of the people who helped you in field, with identification and with illustrations (N. Grant)?
KH: The Polar Continental Shelf Project is a Canadian government funded institute to facilitate research work in the Canadian Arctic. Ramsay was a polar bear biologist who was also my PhD supervisor at the University of Saskatchewan. He has since died in a helicopter accident while working on polar bears. Amamalik and Amarualik were Inuit assistants who lived in Resolute Bay. Bergmann and Siferd worked for Harold Welch as research assistants. Graham was director of the Vancouver Aquarium at the time and helped with field collections. Hop was a fellow PhD student working on Arctic Cod. Martin was the captain of the research vessel Ogak. Curtis and Frank were museum experts who helped identify specimens. Grant was a graphics person who worked in the Biology Department, University of Saskatchewan.
HS: Did this paper have a smooth ride through peer review? Was Marine Ecology Progress Series the first place you submitted this to?
KH: The paper was only submitted to this journal and I recall it was well received and required only one revision before being accepted.
HS: In your paper you use a value of +3.8% for the Delta15N enrichment factor between trophic levels. Has there been improvement in our ability to estimate this value? What is this value now, for this system? Also, is the method used to calculate trophic level
still the same?
KH: Rather remarkably, this value has held up well. Meta-analyses of many food webs now places the value closer to 3.4 per mil, but I still find the best evidence supports our earlier estimate. Then again, this value was generalized for the whole food web and we really did not expect that it would apply precisely to every trophic step. We used this value to model the food web off west Greenland just recently and it continues to make sense. However, the trend today is to use stable nitrogen isotope analyses of individual amino acids to model trophic positions. That work is much more involved, analytically, and, of course, is much more expensive, but will likely provide the most accurate data in the long term.
HS: You say “to expect a consistent isotopic enrichment factor between all trophic levels is inappropriate” Do we have a better understanding, today, of specific isotopic fractionation factors between trophic levels? You also highlight the need for “Controlled laboratory studies..to determine isotopic fractionation values between primary producers and herbivores in this system” – has this happened?
KH: As stated above, it remains unrealistic to suppose there is a universal enrichment factor that could be applied equally to all consumers and we, in fact, know this is not the case. Controlled laboratory studies have continued to some degree but I would like to see more. We now know that these factors can be influenced by the nutritional quality of diets and so researchers must be careful to take this into account when dealing with captive animals raised on homogenous diets. Specifically, I would say that our knowledge of the specific factors influencing isotopic discrimination between primary productivity and herbivores remains poorly known.
HS: Is the model of this food web the same today or has it changed – Is the number of levels different? Is the ordering of species different? Have more species been added with more research? Have species been removed because of local extinctions?
KH: In general terms, the food web remains unchanged with about a five trophic level system separating polar bears from primary productivity. I am unsure if species composition has changed, and I would guess that there have been no follow-up studies conducted in the region since our original work. So, little information would be available to detect any extinctions per se. The key players are the copepods, amphipods, arctic cod, seabirds and marine mammals. With climate change, the food web is poised to change.
Clams from the bottom of Lancaster Sound. The large one is Mya truncata, the primary food of walrus. Walruses detect and suck off the long siphon, leaving the other fifty percent of the clam to die. A walrus can eat about 5000 daily (© Harold Welch)
HS: In your paper you say “We predict that seals taken later in the autumn, after the major onshore movement of Arctic cod, will show tissue isotope values more enriched in Delta15N”. Has support for this been found in later work?
KH: To my knowledge, this has not been investigated. With the closure of the DFO research station, I am afraid there seems to have been little incentive for any follow up work.
HS: Towards the end of your paper, you emphasize the need for isotope work that is across seasons and year-round, that targets various age classes of animals, and that uses several tissue types. Has this happened?
KH: This has happened in other parts of the Arctic. The North Water Polynya program used identical methods to describe the food web off northwest Greenland. That study used multiple tissues and sampled from a Canadian Coast Guard vessel intensively from early summer to late summer. I suspect, but do not know for sure, the Norwegians have pursued this question more off Svalbard, using identical procedures but with more year-round sampling. Logistically, it is still extremely difficult to sample food webs in winter beneath the sea ice.
HS: Are Arctic cod and lower trophic-level invertebrates still critical in this food web?
KH: Most definitely yes, as they are the link between amphipods and copepods and piscivorous marine mammals and seabirds. Having said that, I was very impressed by the direct use of zooplankton by many of these “upper trophic level” groups.
HS: In your study, and other studies around that time, there was a lack of strong enrichment of DeltaC13 beyond the first trophic shift. Do we now have a better understanding of why that is so?
KH: There continues to be much evidence for small trophic enrichment effects for carbon vs nitrogen stable isotopes. This is linked to the kinetics of fractionation and the number and magnitude of the steps involved in the construction of animal tissues from diet. Amination and deamination of proteins and the subsequent voiding of nitrogenous waste strongly fractionates biochemically for nitrogen isotopes. The opportunities for fractionation are simply less and of lower magnitude for carbon isotopes in these reactions. So, it is not so much why the carbon isotope discrimination is low trophically but why the nitrogen isotope discriminations are so high.
HS: Did the paper attract a lot of attention when it was published?
KH: Eventually, but I would say that it really took a few years to be recognized appropriately, as the first isotopic delineation of a full length food web of this magnitude.
HS: At the time this was published, did you anticipate that it would be cited so much? Do you know what this paper has been mostly cited for?
No, I did not anticipate it would stand the test of time so well. I think it is primarily cited as a classic example of how powerful the stable isotope approach can be in modeling food webs. To some degree, it also is cited to illustrate the importance of lower trophic level zooplankton in arctic marine food webs.
HS: What impact has this paper had on your research and your career? Did it open up new lines of investigation?
KH: This paper led to continued links with high latitude marine food web research. I have continued such work in the Gulf of Alaska, the northeast and northwest polynyas of Greenland. In more recent years, my food web research has focused more on freshwater systems and the use of new isotopes such as those of hydrogen. The general notoriety of the work certainly helped me to develop my career using stable isotope methods. That work now has moved on more to tracing migration.
HS: What would you say to a student about to read this paper today? What should he or she takeaway from it? Would you add any caveats?
KH: I think students today must see the paper in context. This was the very first attempt to see if stable isotopes could be used to model whole complex food webs. Before this paper, there really was no straightforward modeling of food webs using stable isotopes. All new and original and aimed at the “big picture” or overview. As I indicate below, technology at the time was also not what it is today and so sample sizes were often limited. Students have a much more convenient means of using stable isotopes in their research today. I would also say that the approach was not without its risks of failure and so it can pay to take chances in planning research questions.
HS: Have you ever read the paper after it was published? When you read it now, what strikes you the most about it?
Of course, I am flooded by good memories of the Arctic, the field sites, the animals and the people I worked with. I think scientifically it is still very sound and impressive. The big difference might be that, today, one would expect larger sample sizes, because the technology used to analyse samples isotopically has changed so much. Samples had to be prepared and run individually (by me) in the early days. Today, the systems are automated and students would simply submit their samples to a lab and sample throughput would be comparatively fast and less expensive.
HS: Is this your favourite paper among all the papers you have published? If yes, why? If no, and if you do have another favourite, which is it and why?
KH: Yes, it is definitely one of my favorites. The other competitor would be the early papers (1, 2) we published on tracking the migration of monarch butterflies using stable hydrogen isotopes. Both research projects were just so different from anything that preceded them!
HS: Do you still work in this field site? How has it changed since the last time you worked there?
KH: No, since the laboratory in Resolute Bay was closed, no further work has been possible. I am sure I would notice a big difference if I returned, as many have commented on the way in which the climate has changed.