Mechanisms of genetic differentiation among seabird populations

Connectivity among populations is considered beneficial for population persistence, reducing levels of inbreeding, increasing genetic diversity, and potentially providing demographic and fitness benefits. Consequently, predicting gene flow between populations based on factors such as wind-dispersed seeds in plants, or presence of pelagic larvae in fishes, is highly desirable for identifying conservation priorities and maintaining viability of species. Seabirds provide useful model systems for studying the factors influencing gene flow given their discrete breeding distributions. As little is known about the non-physical (biotic) factors affecting movement among oceanic seabird colonies, my research addresses this crucial knowledge gap in the context of maximizing persistence and resilience of seabird species. This new knowledge can be extended to other organisms in further studies. The aims of my Ph.D. project are to provide practical case studies to investigate contemporary mechanisms of genetic differentiation among seabird colonies. To explore how these and other factors explain seabird population genetic differentiation, I combined my results with those from over 71 studies reporting population genetic data for seabird species and performed a meta-analysis. In my first case study, I tested for a relationship between differences in non-breeding distributions and genetic structuring of flesh-footed shearwaters Ardenna carneipes, a migratory species nesting at Lord Howe Island, New Zealand, southwestern Australia and Saint-Paul Island in the Indian Ocean. Telemetry studies suggest that eastern and western colonies migrate to different non-breeding grounds (North Pacific Ocean and northern Indian Ocean, respectively), and segregation based on migratory patterns has been hypothesised to contribute to population genetic divergence. My results showed strong genetic differentiation between Pacific colonies relative to those to the west. However, molecular analyses of fisheries’ bycatch individuals sampled in the North Pacific Ocean indicated that individuals from both eastern and western colonies were migrating through this area. As the apparent segregation of the non-breeding distribution based on telemetry was not corroborated by my genetic analyses, I concluded that this factor was not a contributor to the population genetic structure observed among colonies. In a case study of a second species, I tested whether genetic isolation exists among colonies that exhibit other phenological or circadian differences—e.g. diurnal versus nocturnal colony attendance. The providence petrel Pterodroma solandri, is an oceanic seabird restricted to two breeding colonies off eastern Australia: Lord Howe Island, and a recently discovered colony on Phillip Island (adjacent to Norfolk Island). Historically, the providence petrel also nested on Norfolk Island, comprising ~ 1 million breeding pairs, before becoming locally extinct by the late 18th century. The reasons for extinction include exploitation by European settlers subsequent to 1788 and the introduction of mammalian predators. The two extant colonies show different times of return to nesting sites (diurnal versus nocturnal), which may represent local adaptation that could inhibit dispersal between populations. I used genetic data to investigate connectivity between these colonies. My results showed genetic homogeneity of colonies, indicating that the small population on Phillip Island represents a recent colonization from the Lord Howe population rather than a relic population from the geographically closer but now extinct Norfolk population. Hence, it is likely that prospectors from Lord Howe Island or their descendants have switched their behaviour on Phillip Island. In a separate study, I analysed subfossils of providence petrel from the extinct (Norfolk Island) population to assess whether population extinction occurred in the presence of genetic connectivity, which is essential to assess the limits of connectivity to attenuate processes that have driven extinctions. The majority of subfossil Norfolk Island individuals exhibited the most common mitochondrial haplotype from Lord Howe Island, consistent with high genetic connectivity. This study provides an insight into how rapidly even very large seabird populations can be decimated by humans despite genetic connectivity with unperturbed populations, which has significant conservation implications for predicting the resilience of other species. I then incorporated my case studies into a multi-species dataset of genetic variation among seabird colonies. I evaluated a candidate set of generalized linear models (GLMs) to identify contributors to population genetic differentiation for these 73 seabird species. Historical fragmentation was the best predictor of genetic differentiation within seabird species and was supported by variation in phenotypic traits, whereas non-physical barriers such as differences in non-breeding movement patterns among colonies did not appear a significant predictor of genetic structure. These results show that signatures of historical events still dominate as contributors to contemporary genetic structuring among seabird colonies even if they are not enduring, provided that they are subsequently reinforced by factors such as constraints on foraging imposed by spatial heterogeneity of ocean productivity