Search tactics of a pause-travel predator: adaptive adjustments of perching times and move distances by hawk owls (Surnia ulula)

Abstract

After having erected artificial perches of different heights in a clear-cut, I studied the effect of perching height on the allocation of search and pursuit effort of 9 individuals of the hawk owl. Hawk owls are diurnal and locate their small mammal prey visually. Horizontal attack and move distances increased with perching height in most individual cases, as well as across cases. Prey detection and giving-up times increased with perching height in half of the individual cases, but not across cases. Attack and move distances increased with perching time in only a few individual cases, and not across cases. Perching height explained more of the variation in attack and move distances than did perching time. Move distance depended on both preceding and subsequent perching heights. When giving up a perch the owls on average moved slightly farther than the distance predicted if travel distance as well as overlap between successive search areas are to be minimized. When the ground was snow-free, distances of successful attacks tended to be shorter than those of unsuccessful attacks. Moreover, perches from which prey were captured, but not perches from which unsuccessful attacks were launched, tended to be lower than perches which were given up. This may suggest that the predator-prey distance is more critical to the probability of capture than to the probability of detection. Longer horizontal attack distances from higher perches suggest that the size of the area effectively searched around each perch increases with the height of the perch. Therefore, longer giving-up times on higher perches indicate allocation of more search time to larger search areas, and longer moves after giving up or before landing on higher perches indicate avoidance of overlap with the larger search areas of these perches. These adjustments of giving-up time and move distance allow hawk owls to forage efficiently.