Use of hidden Markov capture-recapture models to estimate abundance in the presence of uncertainty: Application to the estimation of prevalence of hybrids in animal populations.

摘要: Estimating the relative abundance (prevalence) of different population segments is a key step in addressing fundamental research questions ecology, evolution, and conservation. The raw percentage individuals sample (naive prevalence) generally used for this purpose, but it likely to be subject two main sources bias. First, detectability ignored; second, classification errors may occur due some inherent limits diagnostic methods. We developed hidden Markov (also known as multievent) capture-recapture model estimate prevalence free-ranging populations accounting imperfect uncertainty individual's classification. carried out simulation study compare naive model-based estimates assess performance our under sampling scenarios. then illustrate method with real-world case estimating wolf (Canis lupus) dog lupus familiaris) hybrids northern Italy. showed that could estimated while both uncertainty. Model-based consistently had better than presence differential assignment probability was unbiased scenarios high detectability. also ignoring would lead underestimating hybrids. Our results underline importance approach obtain segments. can adapted any taxa, absolute variety cases involving detection (e.g., sex ratio, proportion breeders, infected individuals).