This blog post is provided by Emma J. Talis and tells the #StoryBehindthePaper for the paper ‘Variability, Skipped Breeding, and Heavy-tailed Dynamics in an Antarctic Seabird‘, which was recently published in Journal of Animal Ecology. The authors show that the presence of variable rates of skipped breeding in Adélie penguins can challenge our interpretation of abundance fluctuations and obscure relationships with key environmental drivers.

Considered a vital bellwether (indicator) species, the Adélie penguin (Pygoscelis adeliae) has long been used to assess the impacts of both climate change and fishing on the Southern Ocean ecosystem. This species of Antarctic penguin shares life history traits with many long-lived bird species that breed colonially. Here, we investigate the Adélie penguin as a study system for a more general look at the causes and consequences of temporal population fluctuations in long-lived annually breeding species.

Population censuses of colonial seabirds like the Adélie penguin are usually conducted by counting nests or breeding pairs present at a given colony during a breeding season (Fig. 1). Though Adélie penguins are largely site faithful, the population dynamics captured by these time series of breeding abundance are characterized by large interannual fluctuations (Fig. 2). As a result of these interannual fluctuations at individual colonies, it is difficult to deduce the driver of Adélie population changes over time or link these population dynamics to the observed environment.

Like many long-lived species, the fitness of individual Adélie penguins is influenced by many factors; the least understood of these factors is the decision of whether (or not) to show up to the breeding colony to breed in any given year. Skipped breeding is common among many birds, especially among long-lived seabirds, and previous studies have demonstrated the importance of skipped breeding in understanding seabird population dynamics. However, the frequency with which Adélie penguins skip breeding is not well studied and estimates for breeding propensity are limited. Since individuals choosing not to breed will change the observed breeding abundance (but not the true population) in that year, skipped breeding is critical to our understanding of how census data should be used to estimate trends and forecast population responses.

We leveraged the well-established understanding of Adélie life history and demography to investigate how fluctuations in specific demographic rates (including skipped breeding as well as survival and reproduction) manifest in fluctuations in total breeding abundance. Using a combination of observed breeding abundance through time at Adélie penguin colonies and simulations of time series based on those observations, we investigated the causal drivers of these interannual fluctuations and their impact on our use of abundance time series for forecasting and long-term monitoring.

We found that breeding propensity drives observed breeding fluctuations more than any other vital rate, with high variability in skipped breeding decoupling true abundance from observed breeding abundance. Importantly, it is high variability in breeding propensity—not uniformly high rates of skipped breeding—that drives fluctuations in breeding abundance. High variability in breeding propensity, even if skipped breeding is low on average, may explain why fluctuations in abundance have been so difficult to link to environmental conditions, why forecast horizons remain stubbornly short, and why the dynamics of any one breeding population bears little resemblance to those nearby.

We also identified regions of Antarctica in which Adélie time series exhibit such marked interannual fluctuations that a heavy-tailed distribution like the Student’s t, which allows for a higher probability of extreme fluctuations than a Gaussian, is required to represent abundance accurately. These regions of Antarctica also frequently displayed high variability in the availability of zooplankton, which acts as both direct and indirect food sources for Adélie penguins. In simulated time series, the likelihood of these heavy-tailed dynamics also increased considerably with increasingly variable adult survival.

Our results highlight how censuses of breeding populations may appear more variable—and may exhibit more frequent extremes—than those that sample the entire population of a species. Since counts of nests or breeding pairs are common for long-lived bird species like the Adélie penguin and remain a valuable tool for identifying long-term population trends, our findings provide a much-needed caution that such data may be difficult to use for inferring causal drivers.

Read the paper

Read the full paper here: Talis, E.J., Che-Castaldo, C., Şen, B., Krumhardt, K. & Lynch, H.J. (2022). Variability, skipped breeding and heavy-tailed dynamics in an Antarctic seabird.  Journal of Animal Ecology https://doi.org/10.1111/1365-2656.13827

About the author

Emma Talis is an applied mathematics and statistics PhD student at Stony Brook University working in the Lynch Lab for Quantitative Ecology. Her research focuses on modeling animal population dynamics with a particular interest in the
uncertainty and fluctuations in time series of Pygoscelis penguin abundance. Learn more about the Antarctic research happening in the Lynch Lab at https://www.lynchlab.com/ and @TheLynchLab on Twitter.