Robustness of simple avian population trend models for semi-structured citizen science data is species-dependent

Abstract

Accurate and robust population monitoring is essential to effective biodiversity conservation. Citizen scientists are collecting opportunistic biodiversity records on unprecedented temporal and spatial scales, vastly outnumbering the records achievable from structured surveys. Opportunistic records may exhibit spatio-temporal biases and/or large heterogeneity in observer effort and skill, but the quantity-quality trade-offs between surveys and less structured schemes remain poorly understood.Recent work has advocated the use of simple trend models for opportunistic biodiversity records. We examine the robustness of population trends of common United Kingdom birds derived from two citizen-science schemes; BirdTrack, an opportunistic recording scheme, and the structured Breeding Bird Survey (BBS). We derived reporting rate trends from BirdTrack lists using simple statistical models which accounted for list-level effort covariates but not for preferential sampling, and compared them to abundance and occurrence trends derived from BBS survey data.For 90 out of 141 species, interannual changes in reporting rates were positively correlated with trends from structured surveys. Correlations were higher for widespread species and those exhibiting marked population change. We found less agreement among trends for rarer species and those with small or uncertain population trajectories. The magnitude of long-term changes in reporting rates was generally smaller than the magnitude of occupancy or abundance changes, but this relationship exhibited wide scatter, complicating the interpretability of reporting rate trends. Our findings suggest that simple statistical models for estimating population trends from opportunistic complete lists are robust only for widespread and common species, even in a scheme with many observers and extensive coverage.