Protecting penguins

One of this year’s shortlisted candidates for the prestigious Shackleton Medal is Dr Heather Lynch whose work in Antarctica is untraditional in every way. She has dedicated nearly two decades to mapping the distribution and abundance of Antarctic penguin species with the help of fieldwork, exploration and technological advances. In our interview, she shares how satellite imagery, computer vision and artificial intelligence advances her research and explains why the growing tourism to the Antarctic region might be problematic.

An interview with and photographs by Dr Heather Lynch
Additional photographs by Jean Wimmerlin, James Eades, Tetiana Grypachevska, Dylan Shaw and Rod Long

Oceanographic (OM): Can you tell us a little bit about your background, please?

Heather Lynch: “In college, my background was in experimental physics. When I went to graduate school, I became concerned with environmental issues. That’s why I transferred into the biology program. Initially, I was working on fire-insect outbreak interactions, but when a postdoctoral position opened up to work in the Antarctic, it seemed like a great opportunity to use my quantitative skills and my modelling skills. I went to the Antarctic just a couple of days after finishing my PhD dissertation to start work on a long-term monitoring program for penguin populations. I’ve been working on it ever since. I fell in love with the modelling aspects but also the opportunity to have a real impact on policy.”

OM: How did your fascination with the Antarctic regions develop?

Heather Lynch: “When I took my first trip to the Antarctic, we had a field camp at a place called Petermann Island where we were camping and monitoring the penguins. What I think drew me to it initially is just how little information there was and how much it needed some more sophisticated modelling tools. Fairly early on in my career, I got a chance to go to the Antarctic Treaty Consultative Meeting as part of the Secretariat to see how the policies around the Antarctic were being developed. It was a really great window into how policymakers were making decisions and what kind of data they might need to make good decisions. That started off my journey working at the interface of the policymakers and building relationships. That’s a very difficult work because it’s very time consuming. Those meetings are very slow. It feels like the policies will never come to fruition, it feels like nothing is ever happening, but behind the scenes, over years and years, the relationships that you build with policymakers and scientists, do move things forward in terms of conservation.”

OM: Was the involvement with policymaking something that you had in mind when you started?

Heather Lynch: “Yes, for sure. In fact, I never envisioned going into academic research. I always imagined that I would go into a conservation organization. But I saw that I didn’t have to choose. In some sense, I could take an academic tract and still have a big role in policy. Everything I do, I think of it being policy relevant. We very much want to answer questions that policymakers have and that has allowed us to be quite creative in terms of developing new tools or new skills. I think that we’re still pushing things forward in terms of basic research, but it’s all driven by this larger goal of trying to understand how the analytics are changing so that the people that are designing protected areas have all the information that they need. That’s why a big focus of my lab’s research is what we call ‘open science’, which is looking at the question: how do we not just do the science, but create data sets that are of maximal value and are publicly available? That’s why everything we do goes online immediately, and we try and shorten that gap between discovery stage and dissemination. Often, we will share our results with policymakers even before it’s gone through peer review. We might change the way we think about this after it has been looked at by the scientific community, but I don’t want to wait on results I think are important for a year or two while it goes through the peer review process. We try and get our results to the policymakers as quickly as possible so that they have it in hand when they need it.”


OM: As you mentioned, you do research a bit differently. For example, you don’t rely on support from the US Antarctic Program as is usually the case. Why did you choose to do that? How do you plan the logistics differently than other researchers?

Heather Lynch: “At the beginning of my career in the Antarctic, we were camping. That was a project that was at the time funded by the National Science Foundation. I was subsequently awarded what’s called the Career Award, which is a big grant for young researchers to help them launch their careers. But subsequent to that, because of the way that we work with Antarctic tour operators, we primarily use cruise ships or privately chartered vessels as a means of getting into the field. We’ve been able to work in the Antarctic without that kind of support over the last 12 or 13 years. It’s a very unusual mode of working because we are doing all of our own logistics and planning and permitting. We don’t have help with gear or safety training, so we’re really on our own. While that’s a much harder way of operating, I think we are able to explore a wider range of areas using these alternative modes of transportation than would otherwise be possible. In particular, we’ve been using smaller yachts to get into some very difficult sections of coastline that are far too shallow for larger research vessels and poorly mapped.”

OM: What kind of technological advances are you using for your research?

Heather Lynch: “That’s probably the biggest thing that I’m known for. If I’m known for anything, it’s probably for my use of satellite imagery. That stems from the fact that my PhD work, when I was working on insect outbreaks and forest fires, was done with satellite imagery. My PhD work was focused on how do we use satellites to study ecological phenomena on the ground, so I carried that over into studying penguins. When I figured out that we could see penguin excrement from satellite imagery, it opened all sorts of doors for surveying populations. Combining that later on with computer vision and artificial intelligence has allowed us to try and start automating a lot of the processing of this imagery. Antarctica is a huge continent and we simply can’t manually annotate all this imagery. The satellite imagery work is something that I’ve very much pushed forward and it just happens to have linked up to the concurrent effort in computer vision. Combining those two together puts us in a position where we can start monitoring the entire Antarctic annually, which is unheard of. When I started my career, we knew really only about those colonies that were right next to research stations. There were maybe five or six colonies that we knew something about. And there are about 900 penguin colonies in Antarctica. Now we can study the populations that are too remote to visit and have never been visited by humans. Now we can see them from space and we can tell how their populations are changing over time.”

OM: You briefly mentioned artificial intelligence. How are you making use of that in your research?

Heather Lynch: “We started trying to automate the detection of penguins in satellite imagery and we continue to work on that. It turned out to be a very hard problem. We then discovered that the easier problem was surveying pack ice seals and even Antarctic whales that we can see under the surface of the ocean. We’ve done a lot of work on all three species – penguins, pack ice seals, and on whales. Computer vision is critical when thinking about how to expand this from a pilot study where we prove it’s possible to actually scanning the entire Antarctic. Right now, we’re also working on using satellite imagery and computer vision tools to not only pull out pack ice seals, but also the pack ice environment. We’re trying to determine, at the finest scales, how the sea ice environment is changing in Antarctica, and how the animals are responding to those changes. We’re very lucky that all of this effort to use satellites to study penguins came along at exactly the same time that computer scientists were developing all these new AI tools so that we can marry them together for the benefit of long-term monitoring.”

OM: You’ve already delved into this, but what research are you currently focused on?

Heather Lynch: “Right now we have a whole number of projects going on in the lab. One of the big projects that we’ve been working on for a decade, and I think we’re finally getting close to wrapping it up, is an effort to use Landsat satellite imagery, which is a lower resolution sensor, but it has been in orbit since the late 1970s. With it, we can look at individual pixels on the ground. These are 30 metre by 30 metre pixels, and when we follow that pixel through 50 years of time, we can extract the penguin diet from that satellite imagery. Through it, we can see how penguin diet has changed over 50 years and how these penguin colonies respond to sudden shifts in diet. We now have this really long time series that we’re getting out of the Landsat satellite sensor. We also have other projects applying satellite imagery to seals, as well as Antarctic petrels. We knew very little about Antarctic petrels, one of the most numerous seabirds in the Antarctic, but they’re virtually unknown in terms of their distribution. So we are using satellite imagery to survey for Antarctic petrels across the entire Antarctic coastlines. That’s a big part of our research right now as well.”

OM: What have you found out exactly? How has the penguin diet changed in recent years in the Antarctic?

Heather Lynch: “We did some preliminary work on this about five years ago. It has been a really long effort. What we found was that we saw the regional differences that we expected to find. We found that in some areas of Antarctica, they’re eating a primarily krill-based diet and a primarily fish-based diet in other areas. We didn’t, however, at that time, see an overarching trend towards a fishier diet or a more krill-based diet. I think what’s happening is that in areas where the penguins prefer krill, when the krill availability is low, they’re really starving. I don’t think that they have as much diet flexibility in particular areas and that’s what we’re seeing in the guano. With recent technological advances, we’re now able to look at intraseasonal change which is exciting. It might be that over decadal scales, the diet doesn’t appear to change, but within a year, we might see shocks to the system with temporary deviations where the penguins might be switching from krill to fish. Those are the things we currently look at intra-seasonally because Landsat takes a picture every eight days and so we actually get quite a bit of resolution within a season as well as across seasons.”

OM: When taking data from satellites, how are you certain about what you’re seeing in an image?

Heather Lynch: “The reason we feel quite confident in this is that we, when we were in the field, took down a very expensive hand-held instrument called a spectrometer which can measure guano in the field. We then collected that guano and we analysed it in a lab back at home. What we found was a very striking relationship between the proportion of fish or krill in the diet as measured in the lab and what we were measuring in the field. We were thus able to validate our satellite images. It’s not too surprising that relationship works because penguins basically eat red krill or white fish. So even just visually, there’s a big difference between when they’re eating fish and when they’re eating krill.”

OM: In recent news, scientists warned about the increasing breeding failure of endangered emperor penguins due to the ever-decreasing amount of pack ice in the Antarctic. How are penguins and other seabirds faring in general?

Heather Lynch: “We do some work with emperors, but our main focus is on brush-tailed penguins. What we’re seeing is that we have winners and losers. We have one species that is a more sub-Antarctic species and what climate change looks like to them is that their populations have skyrocketed, they’re spreading throughout the region, and they’ve become really successful because the Antarctic is becoming warmer and it’s becoming less heavily choked with sea ice through winter. Then we have species that appear to be really suffering under climate change. The Adélie penguin, for example, which is a classic Antarctic species. We would sort of expect that when it’s becoming warmer that they would do poorly. The chinstrap penguin is the enigma because we’re not sure why their populations are doing quite so badly, so that’s the species that we’ve been focusing on recently. They feed in these really remote islands and we have dedicated around five or six years to going there and trying to survey these places to understand what’s going on.”

OM: In your eyes, what are the biggest threats to penguins in the Antarctic right now?

Heather Lynch: “The main three that we talk a lot about and worry about are climate change, tourism impacts and fishing. My thinking about these has evolved a little bit. I’ve always thought that climate change was the biggest impact that they’re facing, and I continue to think that. I tend to think that fishing has less of a direct impact than climate change does, although I have colleagues that I respect a lot who think fishing has a bigger impact. The one that I’ve come to see a little bit differently is tourism. We’ve done a lot of work on the impacts of tourism and stress hormones and breeding successes. At first, it didn’t appear as though tourists were having a big impact in the Antarctic, but then COVID happened, and we had an opportunity to conduct a natural experiment to see what would happen when tourists leave the region. What we found when all tourists left was that the penguin populations drastically rearranged themselves in those two years during COVID. That means that we are having more of an impact than we appreciated. In some sense, they’re not stressed by our presence because they are the relaxed subset of the population, while there probably is a substantial portion of the population that is avoiding us and that is stressed by our presence. The Anthropos experiment during the COVID era has us thinking a little bit differently about tourism impacts. I used to be very convinced that it wasn’t an issue and now I’m less convinced. I think, it’s something we need to think more carefully about.”

OM: During COVID, did the penguins suddenly populate and take over the areas where tourists would usually be present?

Heather Lynch: “Yes, penguins are very site faithful. They breed in the same locations every year. It means that we have populations that we think of as genetically closed. We don’t have individuals moving between sites so that we can track an individual site through time and not worry so much about immigration and emigration. What we found during this two-year period when tourists were gone at this really heavily visited site that has the Penguin Post Office – a famous location in the Antarctic – is that several hundred penguins came from a nearby site to the Penguin Post Office. When the tourists returned after COVID, they left again. We’re still working out the finer details of that and trying to figure out how many of those penguins that moved into the Penguin Post Office were able to successfully breed. Initially, we thought that they just started breeding right away and now we’re reanalysing some of the imagery we took with drones during that time. It looks like maybe they weren’t actually nesting. The fact that we had imagery from all of the surrounding sites allowed us to see exactly not just the penguins that showed up at this location but where they had come from and the fact that they returned afterwards. I think that was quite a convincing piece of evidence that these were the shy birds in some sense that showed up when the tourists went away.”

OM: With the growing number of tourists heading to the Antarctic for leisure, how do you think tourism should look like in the region? What needs changing?

Heather Lynch: “It’s quite an issue but I still think that my biggest concern with tourism in the Antarctic is not what’s going on at the actual site, but just the carbon footprint of that entire trip. I think we have to think very seriously about our own emissions when we do ecotourism. That trip has to be eco all the way around. We need to move towards carbon neutral travel and encourage passengers to purchase offsets to try and make their travels not as impactful. In terms of how we operate on the ground, I give the industry a lot of credit for its sensitivity to these issues and in many ways they’re far ahead of the legislation. They legislate themselves before they get legislated by the treaty parties. On another note, the COVID experiment showed that if we’re going to accept the fact that the penguins that are scared of us will avoid us, and the penguins that are habituated to us will stay, then that means that we need to be visiting the same sites year after year. We should avoid the large amount of expedition tourism where we go to never-before-seen islands, and where cruise ships are looking for that new spot that no one else is going to. If we stick to around 20 to 30 sites that are visited regularly, and these sites might see 500 people a day, over time, the penguins that are going to be living there are going to be heavily habituated, and I think that we can coexist in a way in which we’re not having a negative impact other than the behavioural modification associated with habituation. I think there’s a win-win here, but it does mean that we should stick to a set of sites that are the tourist-visited sites and leave the rest of these sites unvisited.”

OM: During your 20 years working in the region, what changes have you personally witnessed?

Heather Lynch: “Oh, absolutely. The penguin populations have changed so dramatically. The numbers of gentoo penguins, which are a sub-Antarctic species, are exploding. Logistically that means that there are some sites now that we can’t monitor as effectively because they’re simply too big. We need too much time while we’re on the ground. Luckily, over the course of my career, we’ve gone from exclusively hand counting penguins to a combination of hand counting and drone imaging. On the other hand, there are areas that used to have chinstrap penguins and they no longer have them there, so a lot of our work is documenting these extra patterns on individual islands. But the rise of the Gentoo penguin, ironically, is what we feel most viscerally. Under climate change, we’re very worried about the declines in the other two species. When we go to the Antarctic and see gentoo penguins everywhere where they haven’t been before, it really makes you feel like you’re at the heart of climate change. It’s weird to think that one would be disappointed to find more penguins, but more gentoo penguins really is a warning sign that the Antarctic is becoming more sub-Antarctic. The ecosystem has really changed. Another sign for that is that we’re finding more king penguins in the Antarctic. The population’s stronghold is usually in South Georgia but we’ve seen king penguins come to Elephant Island in the Antarctic where they attempted to breed. Their chicks have not yet survived the winter, but they are pairing up and mating. And it’s just a matter of time before it gets warm enough until those chicks survive. This is yet another sign that the Antarctic is becoming more of a sub-Antarctic ecosystem.”

OM: What’s the most interesting or most impactful discovery you made?

Heather Lynch: “I think the most impactful one is definitely our discovery of a very large set of Adélie penguin colonies in the Danger Islands which are very hard to access. My colleague and I were looking through a bunch of Landsat imagery and he showed me where the algorithm was saying that there are Adélie colonies. There were all these satellite pixels identified as having guano in the Danger Islands and I was sort of annoyed by this, because I thought our system is clearly making mistakes, and I need to figure out why. When I then looked at at higher resolution imagery, we realized that there were these enormous colonies of Adélie penguins that no one knew were there. These were some of the largest Adélie colonies in the world. And so, along with a number of other partners, we organised an expedition and we found hundreds of thousands of Adélie penguins that were unknown to exist there. Importantly, the Danger Islands, at that time, were not considered to be a high priority for protection. Our discovery of this enormous colony helped protect the Islands further as we immediately passed the information along to the people that needed it. The Antarctic Treaty parties are negotiating these Marine Protected Areas for both sides of the Antarctic Peninsula, but the Danger Islands actually fell between the two proposed MPAs. Now it’s very clear that, if there is going to be a MPA on the western Antarctic Peninsula, the Danger Islands absolutely need to be included in it. The negotiations are ongoing and these things take a long time, but in the meantime, the are has been proposed as an Antarctic Specially Protected Area, which I imagine will go through in the next couple of years which is exciting.

The most recent work we’ve done during COVID is also up there as an impactful discovery in terms of being policy relevant. The very first thing that happened last year when we discovered that the penguins were moving in response to our presence was the institution of a ‘No Wake Zone’ around this colony because we had evidence that what was probably causing a lot of these impacts was boat traffic. It’s nice that it directly translates into a policy that will have an immediate impact.”

OM: What’s next for you?

Heather Lynch: “I’m putting together a stage production with the National Geographic team which will be a three-year national tour called ‘Penguins of Antarctica’. We’ll be touring Canada and the US and I’m really excited to highlight the science and some of the conservation issues and just have a lot of fun with it. Our first show is in October and then we’ll be traveling all over the country, spreading the message about Antarctic conservation.”


Additional photographs by Jean Wimmerlin, James Eades, Tetiana Grypachevska, Dylan Shaw and Rod Long

Current issue

Back issues

Enjoy so much more from Oceanographic Magazine by becoming a subscriber.
A range of subscription options are available.