State-dependent behavioural theory for assessing the fitness consequences of anthropogenic disturbance on capital and income breeders

Anthropogenic disturbance is of increasing concern for wildlife populations, necessitating the development of models that link behavioural changes at the individual level with biologically meaningful changes at the population level. We developed a general framework for estimating the fitness consequences of disturbance that affects foraging behaviour using state-dependent behavioural theory implemented by Stochastic Dynamic Programming (SDP). We illustrate this framework using generalized examples of pinnipeds, a group of marine carnivores that include both capital- and income-breeding species. We examined how disturbance affected pup recruitment separately for each reproductive strategy, and the impact of foraging decisions and parameter values on model results. The effect of disturbance on pup recruitment was greater for income than capital breeders for all disturbance scenarios. Disturbance had negligible effects on pup recruitment when it occurred within less frequented foraging patches, but moderate to large effects when it occurred within an important foraging patch. Model results were sensitive to values of patch productivity (the energy gained from successful foraging), the probability of disturbance and individual behavioural choices in the face of disturbance. State-dependent behavioural theory implemented by SDP is a powerful tool for investigating when behavioural changes in response to disturbance may be meaningful at the population level. This approach allows us to incorporate many factors that are known to influence the behavioural and physiological responses of animals to anthropogenic disturbance, and places disturbance within the context of a temporally and spatially variable environment. The general framework we have developed can be used to estimate the consequences of anthropogenic disturbance across a broad range of species.