Assessing predation on insects in the field is notoriously difficult, as it is challenging to track fates of individuals, prey are often removed without trace, and camera setups are cost prohibitive and scale-limited. Artificial ‘dummy’ caterpillars made from plasticine represent an alternative for determining relative predation intensity. While their use is increasing in ecological studies, plasticine models differ from living caterpillars in several ways. One difference is that many caterpillars exhibit UV reflectance, a potentially important visual cue to insectivorous birds. To test the importance of UV reflectance on predation attempts by birds, we selected a UV reflecting species, the Polyphemus silk moth (Antheraea polyphemus) as our model. We hypothesized that attack rates would be higher for dummies that fluoresce under UV, as this would present additional visual cues. We made 240 plasticine caterpillar dummies, scaled to the size and shape of Polyphemus larvae, and painted 80 with UV-reflective paint in patterns evident on live caterpillars. The dummies were deployed for 6 days across 16 forested sites in northern New York, after which they were collected and assessed for damage from avian predators. Differences in predation between unpainted and UV-painted dummies were compared across sites.
Results/Conclusions
Avian strikes, which leave characteristic beak imprints in the soft plasticine, were recorded on 26 dummies. Predation was lower for UV-painted dummies, 4 bird strikes (5% attack rate) compared to 22 strikes on unpainted dummies (13.75% attack rate). When compared across sites, statistical significance was detected between predation on dummy types (t = -2.61, p = 0.014). No noticeable correlations between site variables (elevation, canopy diversity, aspect) and avian predation rates were detected. Though measurements using plasticine dummies are purported to represent real predation intensity, our results suggest that using plain plasticine dummies with no UV features may lead to inflated estimates of avian predation. The lower attack proportion on UV-painted dummies also opens interesting avenues of further investigation as the patterns may enhance crypsis and thereby lower detection or could present warning cues to avian eyes.