A0942
Title: Extreme-value modelling of migratory bird arrival dates: Insights from citizen science data
Authors: Jonathan Koh - ETH Zurich (Switzerland) [presenting]
Abstract: Citizen science mobilizes many observers and gathers huge datasets, but often without strict sampling protocols, resulting in observation biases due to heterogeneous sampling effort, which can lead to biased predictions. A spatio-temporal Bayesian hierarchical model is developed for bias-corrected estimation of arrival dates of the first migratory bird individuals at their breeding sites. Higher sampling effort could be correlated with earlier observed dates. Data fusion of two citizen-science datasets is implemented with fundamentally different protocols (Breeding Bird Survey, eBird), and posterior distributions of the latent process are obtained, which contain four spatial components endowed with Gaussian process priors: Species niche, sampling effort, position, and scale parameters of annual first arrival date. The data layer consists of four response variables: Counts of observed eBird locations (Poisson), presence or absence at observed eBird locations (Binomial), BBS occurrence counts (Poisson), and first arrival dates (generalized extreme-value). The aim is to devise a Markov chain Monte Carlo scheme and check by simulation that the latent process components are identifiable. The model is applied to several migratory bird species in the northeastern US for 2001-2021, and it is found that the sampling effort significantly modulates the observed first arrival dates. This relationship is exploited to effectively bias-correct predictions of the true first arrivals.
