Gabrielle Alli's notes on Dr. Heather Lynch, a professor of the development and application of statistics and researcher of statistical ecology in the Antarctic.
Dr. Heather Lynch is a professor at Stony Brook University and has been recognized as the Institute for Advanced Computational Sciences Endowed Chair for Ecology and Evolution. Her team, the Lynch Lab, specializes in statistical ecology applied to Antarctic conditions— more specifically, how climate change is affecting seabird populations. Dr. Lynch’s journey to this field consisted of education and research in a variety of scientific subjects.
Dr. Lynch pursued research throughout her high school career, including health physics studies in Fort Belvoir, Virginia. Her interest in chemical engineering as a Princeton undergraduate quickly changed to physics, a field in which she seized the opportunity to do independent research for hours on end. In 2000, she earned her A.B. in physics and was awarded the American Physical Society’s LeRoy Apker Award for the best undergraduate physics thesis in the US. While Dr. Lynch went on to receive her A.M. in physics from Harvard University, her interest in environmental issues led her to switch to biology in the middle of graduate school. This transition was difficult, but allowed Dr. Lynch to harness the abilities she had gained from previous work and apply them in a new way.
Her early research in ecology examines the relationship between insect outbreaks and forest fires, investigating the claim that beetle presence causes flammability in trees. Based on the belief that this does occur, public policy often calls for trees to be cut down to avoid burning. Dr. Lynch examined existing aerial detection surveys of insect outbreaks in Yellowstone National Park¹ and digitized them for large-scale examination. Her analysis focused on the history of mountain pine beetle infestations. Rather than working with a typical regression model to examine this data’s correlation with burn patterns, the Markov chain Monte Carlo (MCMC) was applied. This is a class of algorithms that accounts for the important factor of spatial distribution in this investigation. Dr. Lynch’s study found that there was weak correlation between beetle presence and the likelihood of forest fire. Since then, there has been growing scientific evidence that insect outbreaks do not justify clear-cutting forests.
Compiling visual data, creating a large-scale distribution model, and analyzing its significance in ecological relationships has carried over into the work done in the Lynch Lab. The Antarctic ecosystem can be particularly hard to understand, and therefore requires specific equipment and methods to examine. The Lynch Lab combines field work, imaging techniques, and statistical analysis to approach their vital questions.
Over time, the Lynch Lab’s inquiries have remained the same: How many penguins are there? How have their populations changed? Why did this change occur? Their early work was strongly field-oriented, using physical clickers to track penguin presence one by one. While this was integral to provide a baseline of “known” information, the data and technology available has since become much more efficient. Dr. Lynch also spent considerable time on penguin phenology to find out why penguins breed earlier or later in a given year. While her studies originally led her to believe that Antarctic environmental factors were impacting breeding patterns, further investigation showed that penguins in captivity exhibit the same variability. This same question is now being re-approached with knowledge that the phenomena may be more complex than previously considered.
The Lynch Lab collaborated with Oxford University to examine the impact of human-driven factors on Gentoo penguin breeding, utilizing camera trap systems
¹. The study hones in on the impact of fishing activity, tourism, and precipitation events as a product of climate change. Researchers set up time-lapse imaging of Gentoo penguin colonies throughout a specified daytime interval to assess chick survival. Using Voronoi tessellation, a “neighborhood” around a given nest was established to estimate where chicks belonged. Additional framework was also applied to refine estimations of a chick’s state, alive or dead. Fishing, tourism, and moderate precipitation events were not shown to have major impacts on Gentoo breeding success. However, extreme weather events can have serious implications, resulting in nest flooding, chick hypothermia, and limited mobility and warmth. The frequency of extreme weather has been exacerbated by climate change.
Voronoi tessellations, also referred to as Voronoi diagrams, were defined by mathematician Georgy Voronoy. These “math mosaics” have applications in computer science, chemistry, and art. Imagine a set of points on a plane. Focusing on any one specific point, there is a certain amount of empty space around it. Voronoi tessellations ask, “How much of that space is closer to that point than any other point on the plane?” Polygons, known as cells, are formed around each point based on this condition. It is important to note that mathematical formulas accompany a formal definition of this concept.
In 2018, The Lynch Lab published a major discovery regarding Adélie penguin colony distribution¹. Using on-site ground counts, computer-automated topographical analysis, and satellite imagery, researchers completed a survey of this species in the Danger Islands. Adélie penguin abundance in this region alone is greater than that of the rest of the Antarctic Peninsula combined. In the past, Dr. Lynch and her colleagues manually examined aerial images of pink guano patterns, excrement that penguins nest in, to locate colonies. Researchers were later able to train computers to flag populations that had previously gone undetected. Satellite imagery revealed the size of the Adélie penguin colonies in the Danger Islands, which was verified by on-site ground counts. This discovery affected preservation and protection efforts, resulting in a push to include the Danger Islands as an area of significance in public policy.
Dr. Lynch continues to examine the growing impacts of climate change through her research. She has begun to touch upon the field of environmental sociology, examining the relationship between society and the natural world. There are often disparities between the most environmentally impactful and the most cost-effective conservation strategies¹. This creates a barrier to meaningful change in public policy. Dr. Lynch believes the key to this conflict is improving scientific communication and investigating issues of widespread relevance. As for the public, Dr. Lynch believes that climate change should not only spark worry, but anger. She says, “Researchers are never going to hold corporations and governments accountable.” Science has provided the knowledge that we need to take care of our world. It is up to us, not as individuals, but as a society, to push for the necessary change.
Commentaires