From Understory to Canopy: In situ Behavior of Neotropical Forest Katydids in Response to Bat Echolocation Calls

Abstract

Predator-prey interactions take place in complex environments, and research on the sensory ecology of predator-detection relies on understanding when, where, and how prey experience and respond to predator cues. Bats are significant nocturnal predators, and insects have evolved diverse strategies for avoiding predation by bats. While it is well-known that insects exhibit anti-bat strategies, from avoidance flight to reduced acoustic signaling, the specific conditions that elicit some of these behaviors are less well-known. To illuminate how insects respond to bats in nature, we studied how calling behavior changed when katydids experienced echolocation calls in a Neotropical forest. The diverse Neotropical bat community includes species that eavesdrop on prey sounds, such as the songs produced by male katydids. Previous research has shown that some katydid species respond to echolocation calls by reducing acoustic signaling. To capture the interactions of bats and katydids, we placed acoustic monitors at 8, 16, and 24 meters in 10 locations in the forest on Barro Colorado Island, Panama and recorded continuously for 24 hours at each location. We randomly selected 250 recordings with echolocation calls and compared the acoustic spectrum of the forest before a bat arrived, when a bat was present, and after the bat was no longer detectable. We tested whether the response to bat calls changes with height, the family of bat producing the calls, the duration of the echolocation sequence, call amplitude, and call peak frequency. Bats appeared on ~50% of nighttime recordings, but echolocation calls that could have been produced by eavesdropping bats were rare (<4% of calls). Insect response to bats was nuanced and context-dependent. Despite the rarity of truly dangerous predator cues, echolocation decreased insect sound at several frequencies and heights. Insect response was not uniform, and in many cases echolocation calls had little effect on insect activity, perhaps reflecting the fact that echolocation calls were an inconsistent cue for the presence of eavesdropping bats. These nuanced responses raise interesting questions about predator detection in noise and provide valuable context for laboratory investigations on the sensory ecology of how individual prey species respond to predator cues.