Abstract

A decade ago a it was concluded that wandering albatrosses (Diomedea exulans) perform ‘Levy flights’ when searching for prey on the ocean surface—a Levy flight is a special kind of random walk that displays fractal properties. The well-known albatross study motivated substantial theoretical work suggesting that a Levy flight might be an optimal foraging strategy. Analysis of further data has concluded Levy-flight movement patterns by a range of organisms, including deer, bumblebees, reindeer, microzooplankton, grey seals, spider monkeys and fishermen. In this talk I will present: (i) an analysis of newer wandering albatross data; (ii) a re-analysis of the albatross data from the original study; (iii) a re-analysis of the aforementioned deer and bumblebee data sets (using likelihood and Akaike weights). Surprisingly, none of these data sets in fact show evidence for Levy flights, in contrast to what was previously thought. Furthermore, the graphical approach widely adopted to conclude Levy flight movement is problematic. This raises questions as to the strength of the empirical evidence for Levy flights by foraging animals.