Dispersal plays a central role in population and community dynamics and is a deceptively complex process. Dispersal is the product of multiple interacting factors that are expressed across distinct behavioral stages (emigration, transience, and immigration). We examine the mechanisms underlying the decision to continue moving in lieu of breeding, and we do so within the framework of statistical decision theory. We examine the pre-breeding dispersal movements of bobolinks (Dolichonyx orzyivorus) and how search duration changes according to habitat configurations.
Results/Conclusions
Our results indicate that dispersal duration was short across a wide range of habitat configurations, but that the duration of dispersal increased with an increase in the amount of unsuitable habitat found within the landscape. Due to the fact that bobolinks are long-distance migrants and their breeding season is relatively short (6-8 weeks) finding and prospecting for suitable habitat creates an additional challenge. Bobolinks must be able to arrive and find a breeding territory relatively quickly, and as predicted by our model, long periods of searching results in costs that are too high to be sustainable. From a management standpoint it is thus clear that we must be able to maintain habitat that does not present costly search scenarios in terms of length of search. To our knowledge statistical decision theory has never been used to analyze changes in dispersal behavior. We suggest that it provides a coherent framework with which to evaluate how various factors interact to produce a complex ecological process, such as dispersal.