Marie Auger-Methe

Indigenous Knowledge (IK) holds information on the relationships between animals and theirenvironment, among many other things. Although the depth of ecological information embodiedwithin IK is often recognized, it is rarely included in species-habitat models as a sole data source or combined with scientific data. In partnership with IK holders, I have developed methods to include IK in statistical approaches to model species-habitat relationships. First, I documented IK focused on species-habitat relationships of ringed seals (natchiq in Iñupiaq; Pusa hispida), bearded seals (ugruk; Erignathus barbatus), and spotted seals (qasigiaq; Phoca largha), in the waters near three Arctic communities: Utqiaġivk, Tikiġaq, and Kotzebue, Alaska. Results showed that all three species use currents during foraging activity, which is not a behaviour captured by previous satellite telemetry studies, but have differing associations with sea ice and thus potentially different responses to climate change. Regional differences in seal behaviour and habitat between the IK from each community were also apparent, highlighting changes along species migration routes. I then developed methods for species-habitat models that rely on IK as the sole data source. The method provides an approach to interpret different types of IK as probability of species presence associated with different habitat types, including dynamic habitat covariates, and combines them in a single model. I applied the method to ringed seals, providing an approach where IK is presented in a way that can be easily considered and included in current species and ecosystem management frameworks. Next, I developed a method to include IK as informed priors and habitat covariates in Bayesian habitat selection functions applied to animal movement data. I show that the inclusion of IK in habitat selection functions can identify important areas for the species that would not have been predicted using scientific data alone, due to the lack of data at appropriate scales. Overall, my work has provided new methods to include IK in species-habitat modelling, highlighting the depth of ecological information within IK. These methods provide approaches to better our understanding of species-habitat relationships that can fully consider IK in species management and conservation.

View record

Habitats are heterogeneous landscapes that vary in resource type, abundance, and availability. Animals are expected to select habitat that maximizes energy acquisition to, in turn, increase fitness. As such, food availability influences animal distribution, and the movement of an animal may reveal information on their foraging patterns and food availability. Additionally, predator avoidance affects animal movement, distribution, and behavior. Understanding these dynamics is increasingly important for species that face anthropogenic-caused ecosystem change. I used satellite-telemetered ringed seals (Pusa hispida) in Hudson Bay as a study species to assess the relative influence of bottom-up and top-down pressures. First, I reviewed common statistical methods for using movement data to understand an animal’s relationship with its habitat and created a practical guide for ecologists. Next, using a dynamic bioclimate envelope model, I modeled the changes in the prey base of ringed seals from 1950 to 2100 and demonstrated a climate-driven regime shift in prey, with a 50% decline in the abundance of the well-distributed, ice-adapted and energy-rich Arctic cod (Boreogadus saida), and an increase in the smaller temperate-associated fish in southern and coastal areas. However, I found that all species declined in mean body size, despite an overall 29% increase in total prey biomass. Then, by linking the modeled prey data from 2006 to 2012 to ringed seal movement, I found that seal movement and foraging behavior relative to prey did not match commonly made assumptions, where I found that seals were foraging less in high prey areas. Finally, I considered daily estimates of ringed seal movement relative to polar bear habitat use and found that there were important interactions between prey diversity and polar bear habitat use, where seal movement, habitat selection, and foraging behavior were influenced by the dynamics between top-down and bottom-up pressures. I found that omitting either prey or predators in the seal models biased the estimates of an area’s importance. My thesis provides insight on the trophic interactions and pressures for an Arctic mesopredator in a changing climate and highlights the importance of considering interspecific interactions when modeling movement.

View record

Animal behaviour may represent an early response to variation in habitat suitability. Identifying factors that promote behaviours may be particularly important in areas undergoing environmental change. Recent advances in remote tracking, satellite imagery, and associated methodologies have enabled behavioural research in animals occupying remote environments where direct observation is impractical. Moving habitats (e.g., drifting sea ice) elicit complex behaviours, affect the apparent movement of animals, and are associated with high observation error. In this thesis, I investigated the foraging ecology of polar bears (Ursus maritimus) during the winter. First, I investigated the accuracy of a commonly used model for sea ice motion using dropped GPS collars. I showed that these satellite-based models underestimate the drift speed and have large errors estimating its direction at low speeds. Second, I developed models for remote-tracking data to study behaviours with orientation bias (e.g., relative to wind). Using a popular class of statistical models, hidden Markov models, I developed movement models that allow for error-prone environmental data. I showed that my method effectively recovered behaviour and outperformed other methods when faced with coarse environmental data. Last, I developed a model to correct for sea ice drift and investigated the effect of diurnal, seasonal, and environmental covariates on polar bear behaviour. I identified a peak in diurnal activity later in the day compared to other populations, as well as an increase in activity as the season progressed, which may be indicative of an increase in active foraging. I also identified spatial patterns of distribution with respect to season, ice concentration, and bear age that may reflect high habitat quality in western Hudson Bay and the potential presence of competitive exclusion. My thesis provides a novel assessment of the error present in remotely-sensed sea ice drift models and data that can be used to improve them in the future. In addition, my thesis presents models that can be applied to investigate important, and previously difficult to model, behaviours with orientation bias across taxa. Finally, my thesis expands on our understanding of polar bear foraging ecology with novel insights on its association with the environment.

View record

Spatio-temporal statistical methods are widely used to model natural phenomena across both space and time. Example phenomena include the concentrations of airborne pollutants and the distributions of endangered species. A spatio-temporal process is said to have been preferentially sampled when the locations and/or times chosen to observe it depend stochastically on the values of the process at the chosen locations and/or times. When standard statistical methodologies are used, predictions of a preferentially sampled spatio-temporal process into unsampled regions and times may be severely biased. Preferential sampling within spatio-temporal data may be the rule rather than the exception in practice. The work demonstrated in this dissertation addresses the issue of preferential sampling. We develop the first general framework for modelling preferential sampling in spatio-temporal data and apply it to historical UK black smoke measurements. We demonstrate that existing estimates of population-level black smoke exposures may be highly inaccurate due to preferential sampling. By leveraging the information contained in the chosen sampling locations, we can adjust estimates of black smoke exposure to the presence of preferential sampling. Next, we develop a fast, intuitive, powerful, and general test for preferential sampling. A user-friendly R-package we wrote performs the test. We demonstrate its utility in both a thorough simulation study and by successfully replicating previously-published results on preferential sampling. Finally, we adapt our ideas on preferential sampling to the setting of spatio-temporal point patterns. By considering the observed point pattern as a spatio-temporal thinned, marked log-Gaussian Cox process, we show that preferential sampling can be directly accounted for within the model. Under certain assumptions, the true distribution of locations can then be attained. Using these ideas, we develop a framework for combining multiple data sources to estimate the spatio-temporal distribution of an animal. We then apply our framework to estimate effort-corrected space-use of an endangered ecotype of killer whales. Ultimately, we hope that investigations into preferential sampling will become an essential component within spatio-temporal analyses, akin to model diagnostics. The methods developed in this dissertation are widely applicable, allowing researchers to routinely perform such investigations.

View record