Abstract
Wind power is one of the most promising energy sources found in nature. Despite being considered a clean energy source, the existence of potential environmental impacts, namely, on flying vertebrates, is broadly recognized. In monitoring studies, estimation of avian (or bats) mortality caused by collision has particular interest and must take into account carcass removal by scavengers. For this purpose, scavenger removal trials are conducted at wind turbine sites. Data from scavenger removal trials refer to time until removal of the carcass and are “classical” examples of survival times.Parametric survival models based on the exponential, Weibull, log-logistic, and lognormal distributions are among the most repeatedly used throughout literature. In this study we aim to discriminate between these four competing parametric models to analyze removal data from trials conducted in ten Portuguese wind farms. Plotting procedures and model selection strategies are used and discussed.