The method of multiple hypotheses and the decline of Steller sea lions in western Alaska

Abstract

In recent years, enormous effort has been expended to explain the cause of the precipitous decline of the western population of Steller sea lions ( Eumatopias jubatus ) since the late 1970s; however, despite these efforts and the proposal of a wide variety of hypotheses, the decline has proven to be very difficult to explain. The authors of a recent comprehensive review of the problem emphasized repeatedly that the system is in dire need of a modelling approach that takes advantage of the data available at small spatial scales (at the level of the rookery). We view this as an opportunity for ecological detection, a process in which multiple hypotheses simultaneously compete and their success is arbitrated by the relevant data. We describe ten hypotheses for which there are sufficient data to allow investigation, a method that allows one to link various sources of data to the hypotheses and the conclusions from this approach. The decline of the western Alaska population of Steller sea lions has proven to be very difficult to explain, in part because most aspects of the population and the environmental variables proposed to explain its decline involve a combination of high spatial and temporal variability, and limited data. Consequently, most researchers pooled data across rookeries or across time, obscuring spatial and/or temporal patterns (Fig. 19.1). Some of these previous studies are described below.