Abstract:

Dynamic energy budget individual-based models (DEB-IBMs) provide a well-tested framework for modelling the acquisition and use of energy throughout the life cycle of organisms. These models are often developed using species-specific data to link behaviour and physiology and predict how individuals and populations perform in changing environments. Novel bio-logging technology, applied to free-ranging animals, provide detailed information about individual behavioural variations and responses to the environment. However, so far, no DEB-IBM has attempted to link and parameterise energy dynamics with the fine-scale activity data obtained by bio-logging to answer questions related to life history events in wild populations. The aim of this study is, therefore, to develop and test a DEB-IBM framework which integrates individuals behaviour obtained from high-resolution activity data. Our approach uses data collected with GPS and accelerometer tags attached to two muskox (Ovibos moschatus). The individuals had different fates: one survived the Arctic winter, while the other died. Applying hidden Markov models to accelerometer data, we quantified the variation in behavioural activity budgets (in stationary, feeding and transit) in relation to environmental conditions. We subsequently test how differences in time allocation affect fitness outcomes such as body mass, energy reserves and survival. The proportion of time spent feeding by each animal is used within the DEB-IBM framework to scale the functional response for food acquisition. Simulations using the seasonal feeding budget of the deceased individual lead to drastically reduced energy reserves and body weight during the Arctic winter, and to elevated probabilities of mortality by the end of winter. In contrast, simulations using the seasonal feeding pattern of the surviving muskox, result in much less pronounced losses in reserves and body weight, leading to negligible over-winter mortality. The framework we present here uses standardized modelling techniques, is applicable across species and can easily be extended to include other fitness measures or environmental stressors. Moreover, our framework provides the foundation needed for future studies to link individual movements, energetic processes, life history events and population dynamics in wildlife, which would open up possibilities to test how changing (future) environmental conditions influence animal decision-making and ultimately population level consequences.

https://doi.org/10.1016/j.ecolmodel.2020.109250