| Summary: | Population connectivity refers to the exchange of individuals between spatially distinct populations and is an important process governing population dynamics. In species with a larval stage, the combination of processes including reproduction output, timing of spawning, larval development, larval behaviour, settling potential and recruitment rates influence connectivity dynamics. The main objectives of this thesis are 1) to identify spatial and temporal larval connectivity dynamics for L. pertusa, an ecologically important species of deep-water coral, and 2) predict changes to the connectivity dynamics with anticipated loss of habitat due to climate change. I use biophysical modelling and graph theory to simulate larval dispersal for L. pertusa populations over its known range in the Northwest Atlantic Ocean, identify patterns of potential connectivity and quantify which populations are integral to facilitating these connections. I determine that the timing of spawning has limited influence on the strongest connections observed, but that larval development time and larval behaviour can significantly affect the strength of potential connections. Populations in the north of the domain near Nova Scotia show high local retention rates and strong equatorward connections, following the dominant current directions. Populations in the Gulf of Mexico (GOM) also show high levels of local retention and potential migration to the Southeast United States (SEUS), from Florida to Cape Hatteras. Using a cluster analysis on potential connections, I identify 3 dominant subregions of the domain, the GOM, the SEUS, and the Northern Domain from New England to Nova Scotia, which show alignment with available genetic data. Simulating habitat loss due to climate change shows that existing populations in the GOM and SEUS are likely at higher risk of connectivity disruptions that those in the Northern Domain, with more habitat loss and higher isolation of those that remain. There also appears to be a low probability of larval connection ...
|
|---|