Search strategies for landscape-level interpatch movements

Ecologists need a better understanding of how animals make decisions about moving across landscapes. To this end, we developed computer simulations th at contrast the effectiveness of various search strategies at finding habit at patches in idealized landscapes (uniform, random, or clumped patches), w here searchers have different energy reserves and face different mortality risks. Nearly straight correlated random walks always produced better dispe rsal success than relatively uncorrelated random walks. However, increasing patch density decreased the degree of correlation that maximized dispersal success. Only under high mortality and low energy reserves in a uniform la ndscape did absolutely straight-line search perform better than any random walk. With low mortality risks and high energy reserves, exhaustive systema tic search was superior to the best correlated random walk; an increase in the perceptual range of the searcher (i.e., patch detectability) also favor ed exhaustive search over relatively straight random walks. For all conditi ons examined, the "average distance rule," a hybrid search rule incorporati ng both straight-line and systematic search, was best. Overall, however, ou r results suggest that a simple and effective search rule for many landscap e-explicit models would involve straight or nearly straight movements.