Author: Andrea Barcelo Celis
Barcelo Celis, Andrea, 2021 Seascape genomics of Australian and New Zealand common dolphins (Delphinus delphis), Flinders University, College of Science and Engineering
Terms of Use: This electronic version is (or will be) made publicly available by Flinders University in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. You may use this material for uses permitted under the Copyright Act 1968. If you are the owner of any included third party copyright material and/or you believe that any material has been made available without permission of the copyright owner please contact copyright@flinders.edu.au with the details.
The common dolphin (Delphius delphis) is a wide-ranging near top marine predator that inhabits a variety of environments in temperate and subtropical waters around the world. The movements of common dolphins are generally associated to those of their prey, which are mostly small-pelagic fishes targeted by commercial fisheries. This occasionally leads to common dolphin-fishery interactions, including mortalities due to by-catch. Although common dolphins are relatively abundant in Australasia, limited information is available for this species about population differentiation, connectivity and adaptation, making it difficult to develop conservation and management plans. Coastal waters in Australasia are characterised by highly heterogeneous and dynamic environments that are expected to influence eco-evolutionary processes in marine populations, including in predatory species with vast distributions. This thesis utilised genome-wide markers, population genomics and seascape genomics to elucidate population differentiation, connectivity, and environmental drivers of adaptive divergence in Australasian common dolphins.
The first aim of this work was to investigate neutral genomic diversity and clarify population structure and connectivity of common dolphins at different spatial scales across Australasia. A double digest restriction-site associated DNA sequencing (ddRADseq) method was used to obtain over 14,700 filtered and putatively neutral Single Nucleotide Polymorphisms (SNPs) for 478 Australasian common dolphins. To date, this represents one of the largest population genomic datasets reported for a small cetacean. This section included samples from both strandings and free-living dolphins from across the species range in Australasia: southern and eastern Australia, and the west and east coasts of New Zealand. In addition, population genetic analyses were carried out using novel and published mitochondrial DNA (mtDNA) control region sequences (173 haplotypes), which suggested differentiation between Pacific and Indian Ocean common dolphins. Analyses of the more resolving genome-wide SNP dataset pointed to a hierarchical metapopulation structure, with three main regional common dolphin populations: southern Australia, eastern Australia, and Tasmania/New Zealand combined. Additional sub-substructure was disclosed within regional populations, with a signal of isolation by distance along coastlines. Nonetheless, connectivity was still observed over thousands of kilometres and across multiple jurisdictions, with considerable genetic exchange across the Tasman Sea. These findings indicate that state and international collaborations are needed for the management of common dolphin populations and fishery-dolphin interactions in the region.
The second aim of the thesis was to implement a seascape genomics framework to identify signals of selection due to environmental heterogeneity and putatively adaptive genomic variation in common dolphins from across southern Australia. This section focused on Indian Ocean common dolphins because this heterogeneous region sustains one of the main regional populations of the species in Australasia and is well represented in our sampling. For this part of the study, only samples from free-ranging individuals for which an associated geolocation was available were used. From an initial dataset of 17,327 filtered SNPs, a putatively adaptive dataset of 806 SNPs was identified based on a multivariate Genotype Environment-Association (GEA) analysis and a Bayesian method. The results of multiple statistical tests point to five adaptively divergent populations of common dolphins in southern Australia. The GEA analysis suggested that adaptive diversity in these populations is mainly influenced by variation in four environmental variables: current velocity, sea surface temperature, primary productivity, and salinity. In turn, these variables appear to be broadly governed by differences in ocean circulation, and by the presence of upwellings and semi-enclose coastal habitats. Both coding and non-coding regions in the candidate adaptive loci appear to have responded to environmental selection. The results of this study of the southern Australian segment of the metapopulation, highlight the impact of environmental heterogeneity on common dolphin connectivity and adaptive divergence.
The third section of the thesis includes free-ranging samples from both southern Australia (Indian Ocean) and the Australian east coast (Pacific Ocean) to carry out a seascape genomics study of adaptation across the Australian-wide common dolphin metapopulation. A combination of analytical tests of selection produced two putatively adaptive SNP datasets. At the broad-scale, over 1,000 candidate adaptive SNPs were identified for the Australian-wide metapopulation. At a fine-scale, 200 candidate adaptive SNPs were found for the eastern Australia segment of the metapopulation. This two-scale strategy allowed determining associations between adaptive genomic variation and environmental variables, after accounting for the effects of spatial distance within the two Australian coasts. At the metapopulation scale, two regional populations were identified, consistent with the previous results based on putatively neutral markers. Genomic variation in these populations appeared strongly associated with variation in sea surface temperature, current velocity, salinity, bathymetry and primary productivity. Lower adaptive divergence was detected along the east coast population segment, which appeared associated mainly to variation in primary productivity. It appears that both broad- and fine-scale adaptive divergence in Australian common dolphins is influenced by three oceanographic and coastal features: (i) ocean circulation patterns at range edges, (ii) areas of eddies and upwellings, and (iii) semi-enclosed coastal habitats. We also detected preliminary evidence for repetitive selection in some regions of the genome, which is speculated in the context of parallel evolution of dolphins on separate coastlines. These results provide environmentally-influenced perspectives for conservation management of multiple segments of the Australian common dolphin metapopulation.
By analytically integrating neutral and adaptive genomic variation and information from key ecological predictors, this thesis generated novel baseline information for the conservation and management of Australasian common dolphins in the context of a rapidly changing and heterogeneous marine environment. The outcomes of this work also call for new collaborative efforts across state and international jurisdictions to ensure that management goals for the species, including those related to maximum by-catch levels in fisheries, are achieved. It also provides an initial first step towards our understanding of adaptive resilience of local and regional populations of a small cetacean, to naturally- and anthropogenically-driven environmental changes.
Keywords: seascape, genomics, SNPs, delphinds, gene flow, adaptation, metapopulation, conservation genomics, common dolphins, Australasia, Australia, New Zealand
Subject: Biological Sciences thesis
Thesis type: Doctor of Philosophy
Completed: 2021
School: College of Science and Engineering
Supervisor: A/Prof. Luciana Möller