Abstract:
Which marine populations are “closed”, which are “open” and what are the major forces driving their geographic patterns of connectivity? Understanding connectivity among populations remains a fundamental challenge for marine science. New Zealand's marine ecosystems and species are highly diverse. Its variable and extensive coastal landscape provides a valuable scenario for studying the processes that drive larval dispersal patterns and population genetic structure. This study aimed to address this issue by taking a community perspective of connectivity, rather than simply a species-by-species view. By simultaneously examining the genetic population structure of a diverse range of coastal species, we can ascertain if there are common locations where connectivity is either restricted or promoted. It is also likely we can only see the broad picture of connectivity around NZ by examining many species simultaneously. Based on the common patterns of connectivity, this study then aimed to determine which of a wide range of factors (including environmental, geographic, historical and species-specific) best explained those patterns. A detailed phylogeographic comparative study was first carried out using two endemic coastal invertebrates with contrasting life history traits, the cat’s eye snail Lunella smaragdus, with short-lived pelagic larvae, and the half-crab Petrolisthes elongatus, with pelagic larvae that live up to four weeks (Chapter 2). Two different genetic markers were used, mitochondrial and nuclear, to study the population structure of these two organisms throughout their NZ-wide distribution. The half-crab revealed much stronger population structure than the snail, contrary to expectations based on their pelagic larval duration (PLD). This appeared to be due to a more recent demographic expansion in the snail. Overall, the species exhibited some common barriers to dispersal, supporting some previously described boundaries to gene flow. Predictions of patterns of gene flow within species are often unreliable, and this may be due largely to differences in recent demographic history. Seascape genetic analyses were conducted in order to highlight the processes behind the patterns of spatial genetic diversity found in L. smaragdus and P. elongatus (Chapter 3). These analyses considered the influence of a wide range of factors, not only environmental variables but also hypotheses associated with historical and oceanographic processes. Several seascape factors had common influence on the genetic patterns of these two co-distributed species. The most striking result was the common influence of an historical southward range expansion in both species. In addition, measures related to sea surface temperature fluctuations were also found to be key variables shaping the genetic patterns in L.smaragdus and P.elongatus. Interestingly, there was evidence that the role of localized environmental factors was more important in the species with smaller PLD, whereas the broad-scale environmental variables better explained the genetic variation in the crustacean. A complex mix of historical, geographic and environmental processes provided a better understanding of the drivers behind the spatial genetic structure of L. smaragdus and P. elongatus. It is clear that to fully understand the community-level patterns of connectivity and the evolutionary processes shaping them, it is necessary to quantitatively compare genetic patterns among multiple species. An existing approach (based on Pelc et al. 2009) was first used to identify the common dispersal breaks among a wide range of coastal endemic species (Chapter 4). It was found that the main significant breaks, around Cook Strait and northern North Island, were not different among species with different dispersal capabilities. After evaluating the limitations of the existing methodologies, a new approach was then developed and tested with NZ marine species (Chapter 5). The proposed methodology was based on first generating curves capturing the spatial genetic patterns of each species using common diversity parameters (called “genogeographic” curves), followed by clustering of the species-specific curves according to their similarities (“genogeographic species clustering”). The results showed that there were several species groups with considerable similarity in their spatial genetic patterns. Most interestingly, several factors such as Spawning Time and Taxon were important in explaining the similarities among species, and deserve further investigation. This thesis provided considerable quantitative information about patterns of connectivity in the New Zealand coastal marine community, and the possible processes driving these patterns. Further development, including the use of a wider range of species and multidisciplinary tools will improve their application to conservation management and the effective design of MPA networks.