Citizen science data reveal possible multi-decadal phenological changes in the arrival time of a migratory tropical seabird species at the breeding ground

Changes in climate and weather variability are having global impacts on the lives of organisms, particularly on upper trophic-level predators such as pelagic seabirds. In the North Atlantic, migratory seabirds are expected to respond to climate variability by adjusting their seasonal events, including the timing of migration and arrival at the breeding site. The timing of these events may be influenced by large-scale atmospheric phenomena such as the North Atlantic Oscillation (NAO) and the Atlantic Multi-decadal Oscillation (AMO) that gauge temporal and spatial variation in environmental conditions across the North Atlantic. In the Bermuda Islands (32 degrees 17' 58? N, 64 degrees 47' 25? W), the White-tailed tropicbird (Phaethon lepturus catsbyii) is one of the few breeding seabird species which returns every year between February and March after migration. According to local belief their return announces the arrival of Spring, and this cultural importance means that the event is noticed. Only recently, early tropicbird returns have been recorded and associated with unusual climate/weather conditions experienced across the region. This raised the question of whether the species is showing some change in its phenology and whether it might be affected by changing climatic conditions. To answer this question, we investigated tropicbird arrival dates at Bermuda using current and historic observation data retrieved from the eBird citizen science platform. After accounting for biases inherent to opportunistic data collection, we selected and used the first annual observation as a proxy for bird arrival time at the Bermuda breeding ground from 1953 to 2023. We found that tropicbird observation dates advanced substantially over the past 70 years suggesting that arrival time has been likely shifting of ca. 20-25 days. However, first observation dates were not related to the variation in annual or winter NAO and AMO Indices, and further investigation is required to understand the underlying causes of these progressively early arrivals. Overall, the study highlights that citizen-science data can be used to unveil hidden phonological patterns when a standardized long-term data collection is missing.