I’m Murat, a second year student in the Master’s programme in Life Science Informatics at the University of Helsinki, specializing in Biomathematics and Biostatistics. Last summer I embarked on a 4-month internship with the Environmental and Ecological Statistics group that was funded by HiLIFE, where I was able to apply my quantitative skills to a field I have been passionate about since I was a child: wildlife conservation. And what a journey it has been! Although my internship officially ended in September, I have kept working on the project until now.
My mission was to develop a Bayesian state-space model to help monitor Baltic ringed seals, a sub-species of ringed seals that was once on the verge of extinction due to unsustainable hunting practices and large-scale reproductive failure caused by pollution. Thankfully, the population has been recovering during the last few decades, at least in the northern region of the Baltic Sea¹. However, highly dynamic conditions brought on by changing sea ice patterns, decreasing pollution levels and the recent re-introduction of seal hunting have made reliable population monitoring increasingly difficult, and assessments of population status have not been possible for over a decade. The inability to estimate population size and growth rate has been a major obstacle in determining sustainable management practices.
During the summer and fall, ringed seals spend most of their time feeding out at sea, trying to accumulate fat reserves for the winter (foraging period). It is common for young seals to accidentally become entangled in fishing gear during this time. As the Baltic sea begins to freeze over, seals construct snow lairs on the sea ice for protection from harsh weather and predators (subnivean period). Adult females give birth to a single pup inside the snow lair. Once the snow and ice begin to melt, seals move out of their lairs and spend most of their time molting on the ice under the sun (molting period). This is the only time when ringed seals are easily visible, and aerial surveys are conducted each year in the spring to count the number of seals basking on ice. [2] (The illustration was created with the aid of DALL-E 3.)My challenge was to create and parametrize a model that could accommodate all of these changing factors as well as the uncertainties associated with them, something conventional monitoring methods fell short of. Instead of analyzing different data sources separately, I integrated them into a single, unified model of ringed seal population dynamics – a method that is often called integrated population modeling³. By analyzing all available data holistically, integrated population models can exploit synergies between different data sources, making it possible to parametrize detailed and mechanistic population models.
The results we obtained with this approach were as exciting as they were important. For example, we discovered that the reproductive rates of seals might have fully recovered from the effects of past pollution, and the population may have increased from less than 5,000 to nearly 30,000 individuals! We also discovered that the recent re-introduction of seal hunting has had a notable impact on population growth, though not critical enough to ring any alarm bells for the seal population—yet. Another interesting finding was that seals may be behaving differently during years with low sea ice cover, hauling-out on ice in larger numbers during their annual molt in the spring. Since hauled-out ringed seals are counted each year during this time, population counts after warm winters may be significantly higher! This means that as sea ice patterns are altered due to climate change, reliably estimating the population size could become increasingly more difficult.
A photo of my research group on a beautiful summer day, after an all-day boat party with swimming, sauna and other fun activities.
During the time I was working on this project, every day brought new and diverse challenges – a rich blend of mathematics, statistics, and ecology. Collaboration with researchers from other fields was not just helpful, but essential, making my days anything but repetitive. A year of hard work has finally culminated in a manuscript that is ready to be submitted for publication at a scientific journal! Reflecting back at the journey, it is amazing to see how much I have learned. I started my internship with many doubts – unsure whether I would enjoy doing research, and whether I had what it takes. I am now seriously considering a PhD, confident that a career in research is right for me. As I near the end of this project, I am filled with gratitude for the opportunity to have played a role in preserving these amazing animals, and I eagerly anticipate my next endeavor in the world of research.
References:
¹Sundqvist, L., Harkonen, T., Svensson, C. J., and Harding, K. C. (2012). Linking climate trends to population dynamics in the Baltic ringed seal: Impacts of historical and future winter temperatures. Ambio, 41:865–872.
²Kelly, B. P., Bengtson, J. L., Boveng, P. L., Cameron, M. F., Dahle, S. P., Jansen, J. K., Logerwell E.A., Overland J.E., Sabine C.L., Waring G.T., & Wilder, J. M. (2010). Status review of the ringed seal (Phoca hispida).
³Schaub, M. and Abadi, F. (2011). Integrated population models: a novel analysis framework for deeper insights into population dynamics. Journal of Ornithology, 152:227–237.
Have you ever seen a seal in Helsinki? Count yourself lucky if you have (I, for one, have yet to see one). However, there was a time when ringed and grey seals thrived throughout the Baltic Sea. They would gracefully search for fish along the coast and take leisurely breaks on beaches and sea ice. Sadly, today they have become a rare sight, with ringed seals in particular confined to a few scattered locations across the Baltic.
Now, here’s a little about me: I’m no biologist. In fact, my last biology course was almost a decade ago. However, those who knew me as a child can vouch for my infamous obsession with wildlife. After obtaining my Bachelor’s degree in engineering, I spent several years working in financial services. But now, after a decade-long detour, I find myself returning full circle to my childhood passion, albeit with a twist. Recently, I completed my first year in the Master’s Programme in Life Science Informatics (LSI) at the University of Helsinki, with a specialization in mathematical ecology.
Just a few weeks ago, I embarked on an exciting 4-month internship funded by HiLIFE. During this internship, I will be working alongside the Environmental and Ecological Statistics Group at the University of Helsinki and the Natural Resources Institute Finland (Luke). Together, we aim to develop mathematical and statistical models that can predict future population sizes of Baltic ringed seals, and if things go well, perhaps grey seals too. We are particularly interested in the effects of hunting, fishing and climate change.
The history of Baltic ringed seals stretches back over 10,000 years. As the glaciers receded during the last ice age, these remarkable creatures migrated into the Baltic region. Over time, the changing climate forced them further north, eventually leading to the isolation of the Baltic population from their Arctic brothers and sisters. Today, these seals are adapted to the unique conditions of the Baltic Sea, and are considered a distinct subspecies.
Unfortunately, the journey for Baltic ringed seals has not been without challenges. Humans have been hunting these seals since their arrival in the Baltic, initially employing traps, harpoons and nets, and later transitioning to firearms. In the early 1900s, conflicts with fisheries prompted the implementation of bounties on ringed seals in the Baltic states. These hunting practices, coupled with chemical contamination in the Baltic, posed severe threats to the population.
The consequences of these detrimental factors became evident as the population plummeted from an estimated 200,000 individuals in 1900 to just a few thousand by the 1970s. The chemical contamination led to widespread female sterility, exacerbating the decline. Today, the Baltic ringed seal population is fragmented into four sub-populations in the Bothnian Bay, the Gulf of Riga, the Gulf of Finland, and the Archipelago Sea.
Despite some progress in mitigating the threats faced by Baltic ringed seals, uncertainties loom on the horizon. The ban on seal hunting and the discontinuation of using harmful substances such as DDTs and PCBs in the late 20th century have contributed to the recovery of ringed seal populations, particularly in the Bothnian Bay. However, the status of the remaining sub-populations remains uncertain.
In recent years, rising seal populations in the Bothnian Bay have led to conflicts with coastal fisheries. To address these conflicts, hunting has been reinstated as a management strategy. Balancing the needs of both seals and fisheries is a complex challenge that requires careful consideration and effective management decisions.
In addition to hunting and chemical contamination, Baltic ringed seals are threatened by entanglement in fishing nets. The unintentional capture of seals in fishing gear poses a potentially serious danger to their survival and calls for the development of sustainable fishing practices that minimize bycatch.
Another pressing concern that looms over Baltic ringed seals is climate change. These seals heavily rely on sea ice for their survival. They construct lairs on thick and stable sea ice, where they overwinter, give birth, and raise their pups. Sea ice also serves as a crucial resting and molting platform for them. However, as climate change accelerates, the loss of sea ice becomes an imminent threat to their habitat.
Illustration of a ringed seal pupping lair. The lair provides shelter to newborn pups from harsh winter weather, predators and even pathogens. Loss of sea ice and reduced snowfall due to climate change are likely to have significant negative effects on pup survival. Image credit: Robert Barnes, UNEP/GRID-Arendal.
The extent to which the loss of sea ice will impact seal populations remains uncertain. Predicting the future dynamics of seal populations in the face of climate change requires sophisticated mathematical and statistical models that can account for various ecological variables and complex interactions.
The Role of Mathematics and Statistics
In the realm of wildlife conservation and population dynamics, the use of mathematical and statistical models plays a crucial role. These models enable researchers to predict and understand the consequences of various management decisions, thereby aiding decision-making processes.
During my HiLIFE funded research project with the Environmental and Ecological Statistics Group, my goal is to construct a Bayesian State Space Model (SSM), which is a type of hidden process model. As the name suggests, hidden process models aim to infer processes that cannot be directly observed. In the case of Baltic ringed seals, our knowledge of the population is based on hunting reports, interviews with fisherman, and annual surveys that count the seals hauled out on ice. However, the true underlying process, encompassing the births, lives and deaths of seals, remains hidden from our direct observation.
Bayesian SSMs provide a powerful tool to unravel the hidden dynamics of seal populations. By combining available data with probabilistic modeling techniques, we can make informed inferences about what is happening “behind the curtain”. These models enable us to estimate demographic rates, assess population trends, predict the effects of management decisions, and gain deeper insights into the complex dynamics of Baltic ringed seals.
A candid photo of my deskmates.
My First Weeks on the Job
During my initial days, I dedicated a significant amount of time reading up on ringed seal biology and learning about the application of state-space models (SSMs) in wildlife population dynamics. Once I gained a reasonable understanding of the ringed seal lifecycle, I constructed a simple age-structured model of their population dynamics. Through the use of Bayesian techniques, I inferred key vital rates, such as age-dependent fertility and mortality rates, achieving a good fit to the available data.
Over the next few months, my objective is to gradually enhance the complexity of this model, striving to develop an SSM that incorporates the intricate mechanistic details of ringed seal biology, as well as the effects of hunting, fishing and the loss of sea ice due to climate change. Among the major challenges ahead is the modeling and inference of density-dependent processes. No population can grow perpetually. Whether it’s the availability of food, space, or the presence of predators, there will inevitably be limiting factors. Unraveling these factors for ringed seals poses a significant challenge, especially since the population is currently rebounding from historically low numbers. However, understanding these limitations is crucial if we are to make meaningful predictions about the future of ringed seals.
It is precisely these kinds of challenges that make research truly exhilarating, and I consider myself fortunate to be confronted with them!
University of Helsinki
