HABITAT SELECTION AND THE VALUE OF INFORMATION IN HETEROGENOUS LANDSCAPES

30 September 2018

Schmidt, Kenneth; Massol, Francois

Despite the wide usage of the term information in evolutionary ecology, there is no general treatise between fitness (i.e., density-dependent population growth) and selection of the environment sensu lato. Here we (1) initiate the building of a quantitative framework with which to examine the relationship between information use in spatially heterogeneous landscapes and density-dependent population growth, and (2) illustrate its utility by applying the framework to an existing model of breeding habitat selection. We begin by linking information, as a process of narrowing choice, to population growth/fitness. Second, we define a measure of a population’s penalty of ignorance based on the Kullback-Leibler index that combines the contributions of resource selection (i.e., biased use of breeding sites) and density-dependent depletion. Third, we quantify the extent to which environmental heterogeneity (i.e., mean and variance within a landscape) constrains sustainable population growth of unbiased agents. We call this the heterogeneity-based fitness deficit, and combine this with population simulations to quantify the independent contribution of information-use strategies to the total population growth rate. We further capitalize on this example to highlight the interactive effects of information between ecological scales when fear affects individual fitness through phenotypic plasticity. Informed breeding habitat selection moderates the demographic cost of fear commensurate with density-dependent information use. Thus, future work should attempt to differentiate between phenotypic plasticity (i.e., acute fear) and demographic responses (i.e., chronic changes in population size). We conclude with a broader discussion of information in alternative contexts, and explore some evolutionary considerations for information use. We note how competition among individuals may constrain the information state among individuals, and the implications of this constraint under environmental change.

Doi
10.1111/oik.05722