Factors Affecting Reed Warbler Risk of Brood Parasitism by the Common Cuckoo

Abstract In this study we conducted a multiple logistic regression analysis of factors hypothesized to influence the risk of Reed Warbler (Acrocephalus scirpaceus) brood parasitism by the Common Cuckoo (Cuculus canorus) on study sites in the south-eastern part of the Czech Republic. We collected data from Common Cuckoo nesting sites surrounding two fishponds. Our logistic regression models were based on the dichotomous dependent variable, parasitism of the Reed Warbler nest, and seven independent variables. Our first model used all data available across sites and years and resulted in a final model in which the only significant contributor was the independent variable “cuckoo view,” the view of host nests from the cuckoo's vantage point in a tree. A second model was developed using data limited to sites and years with the largest sample sizes and expected to yield the most reliable results. That model resulted in three significant contributors: site, cuckoo view, and neighborhood view. In both data sets, the odds of nest parasitism were shown to increase as the view of the host nest became more direct. However, a direct view of the focal nest raised the risk of parasitism to a much greater degree than did a direct view of the neighborhood of nests. Our results provide support for a nest-exposure hypothesis of brood parasitism risk. Although our models have identified nest exposure to be the best predictor of nest parasitism in this system, work remains to unravel the potentially complex relationship among Common Cuckoos, habitat structure, and Reed Warbler hosts.