Abstract
Islands of the vast Southern Ocean host abundant endemic wildlife populations representing important breeding grounds for many seabirds. With contrasting geological, glacial and human-impact histories, these islands represent strong systems for inferring evolutionary processes. Although penguins (Sphenisciformes) spend much of their lives at sea, most taxa require ice-free terrain for breeding, inhabiting every major landmass and archipelago in the Southern Ocean. While penguins are distributed widely across sub-Antarctic and Antarctic coastlines, nearly a third of all taxa are endemic to geologically young islands, especially in the New Zealand (NZ) region. Penguins therefore represent an ideal group of seabirds with which to study the biogeographic and evolutionary effects of island history. This thesis extends classic approaches in island biology research to the Southern Ocean islands, using penguins as the focal system. The various analyses include evidence from near-complete mitochondrial genomes (mitogenomes), thousands of genome-wide single nucleotide polymorphisms (SNPs), together with mitochondrial cytochrome oxidase subunit 1 (COI) and control region (CR) sequences, generated to examine spatial and temporal evolutionary patterns across all extant and recently extinct taxa. First, this thesis generates a broad framework for understanding penguin evolution by providing the first time-calibrated phylogeny to encompass all modern taxa. Divergence-time estimates demonstrate that many island-endemic taxa diverged following the geological emergence of their natal islands. This study also provides the first molecular demographic analyses to encompass all sub-Antarctic and Antarctic penguin taxa. Genome-wide SNPs yielded signatures of concerted demographic expansions following the Last Glacial Maximum (LGM), suggesting taxa inhabiting islands south of the LGM sea-ice limit underwent rapid post-glacial expansions associated with rapid climate change. This result is further supported by very low population structure across entire Southern Ocean distributions. This thesis provides the first multi-species genetic assessment of the impacts of Polynesian and European arrival on NZ and Chatham Island penguin assemblages. Phylogenetic analyses of modern and ancient mitogenomes, together with COI and CR sequences, revealed the presence of two recently extinct penguin taxa from the Chatham Islands: Eudyptes warhami and Megadyptes antipodes richdalei, that are formally described here using genetic and morphological comparisons. Recent extinction of these island-endemic lineages was likely a direct result of human pressure. By contrast, phylogenetic and demographic analyses of COI and CR sequences revealed limited evidence for demographic reductions in NZ Eudyptes penguins, with ecological preferences possibly buffering those taxa from human impacts. This thesis also provides new systematic insights for ‘extinct’ prehistoric penguins elsewhere, demonstrating that the so-called Hunter Island penguin is actually an artificial assemblage of three extant taxa. Finally, this study concludes that rockhopper penguins comprise three species (Eudyptes moseleyi, E. filholi and E. chrysocome), that extant and extinct Megadyptes penguins comprise three subspecies (M. antipodes antipodes, M. a. waitaha and M. a. richdalei) and that macaroni (Eudyptes chrysolophus chrysolophus) and royal (E. c. schlegeli) penguins are probably incipient species. Overall, this study provides a global assessment of recent penguin biogeography, evolution and extinction, providing a comprehensive basis for ongoing management and conservation of penguin biodiversity.