Abstract
Understanding the relationships between organisms and their environment is crucial to determining important areas for conservation and monitoring. In rapidly changing oceans, one approach to quantify these relationships is to identify species assemblages. This study used a nine year dataset of seabird observations sampled approximately every two months during a cross-shelf transect to describe assemblages at the Subtropical Frontal Zone, in southeast Aotearoa/New Zealand. During 36 voyages, 47 species and a minimum of 69,025 individual birds were recorded. We used multivariate, model-based ordinations to identify assemblages against spatial (distance from the coast), temporal (season) and environmental (water mass) predictors. The multivariate models suggest that the distance from the coast and seasons explain most of the observed variability. Gulls and shags influenced a coastal assemblage (<25 km from the coast), and most albatrosses and petrels were only recorded offshore (>35 km). Seasons strongly influenced the assemblages, with 31 of the 39 analysed species classified as migratory or dispersive. Over the nine year dataset, the probability of occurrence of nearly 40% of the analysed species changed, indicating possible changes in the assemblage structure and species ranges. This study shows the importance of accounting for seasonality when describing assemblages in regions supporting high proportions of migratory and/or wide-ranging species. The observed changes in the probability of occurrence of several species may be the first evidence for the effects of oceanographic changes recently described for the southwest Pacific Ocean due to above-average warming caused by climate change.
• We described seabird assemblages at the Subtropical Frontal Zone off New Zealand.
• At coarse-scale, seabird assemblages did not correlate with water masses.
• Seabird assemblages were mostly explained by season and distance from coast.
• High number of migratory and wide-ranging dispersive species were recorded.•38% of the analysed species changed their probability of occurrence over the years.