NSFDEB-NERC - Testing effects of resources and competitors at multiple spatial and temporal scales in multiple populations

It is widely recognized that prey populations can be limited not only by direct predation, but also by the costs of avoiding predation ('risk effects'). Logic suggests that risk effects might also exist in competitive interactions. We propose to test whether the avoidance of risk carries energetic costs that translate into effects on survival, reproduction, population dynamics and gene flow in a subordinate competitor, the African wild dog. We will do this by incorporating new methods into our ongoing long-term studies of African wild dog, lion and prey populations in three ecosystems. Specifically, we will couple direct observation of wild dogs to data from animals equipped with GPS collars, high frequency triaxial accelerometers and magnetic field intensity sensors, which, together, will give us very fine-scaled data on movement, dynamic body acceleration, energy expenditure and energy gain for wild dogs hunting in areas with known densities and distributions of lions and prey. Triaxial accelerometers will provide detailed and precise measurements of vectorial dynamic body acceleration (VeDBA), a powerful proxy for energy expenditure at time scales ranging from seconds to days or months. GPS collars will provide inferences on space use and movement from movement models (particularly dynamic Brownian bridge models - dBBMMs) at time scales from hours to years. These models of movement models, fit to trajectories derived from a combination of VeDBA, magnetic field intensity and GPS locations using a process termed 'dead-reckoning' (where animal movement patterns are derived from using vectors on movement data), will test for effects on movement down to the scale of seconds. Direct observation of the same individuals in continuous three-day 'follows' will provide spatiotemporally matched data on encounters with prey, hunts and kills to quantify energy gain at time scales from hours to years, and will provide critical context for the interpretation of other data. By pairing these data with intensive, long-term monitoring of known individuals, we will test relationships with survival, reproduction and population dynamics (using a Bayesian integrated population model), and effects on gene flow using a SNP chip we have developed and validated. With replication across three ecosystems with well-measured variation in the densities of competitors and prey, we will obtain data for a range of ecological conditions that would not be possible with a single site.