Genetic assessment of population structure and the origin of recruits in brooding corals: understanding population connectivity on the Great Barrier Reef on various timescales

Ecological connectivity (i.e., population connectivity over ecological rather than evolutionary timescales) in the marine environment, including coral reefs, is poorly understood as few studies have specifically estimated contemporary dispersal. This limits our ability to evaluate the design and potential benefits of novel conservation and resource management strategies. Most population genetic studies on reef corals have estimated population connectivity over evolutionary time scales (i.e., over thousands of generations) using conventional statistical methods (F-statistics), as statistical methods to estimate connectivity on ecological time scales from genetic data have only recently been developed. Further to advances in statistical methods, a novel and promising approach to understanding ecological connectivity is to genetically characterise newly settled recruits and identify potential source populations by genetic assignments. This approach also overcomes the confounding effects of analysing multiple cohorts simultaneously, and addresses the potential temporal stochasticity of recruitment, which is largely unquantified for reef corals. Taking advantage of methodological improvements, I assess population connectivity on ecological as well as evolutionary timescales of the model brooding coral species Pocillopora damicornis from microsatellite genotype data, and I compare estimates of ecological connectivity with that of the closely related coral Seriatopora hystrix. First, I present a novel genetic technique to reliably identify species and recently described lineages within the pocilloporid family, which enables population genetic analyses to be focused on genetically (and presumably biologically) meaningful units of observation. I then infer recent and historic larval exchange between populations of the two most common lineages of P. damicornis (Type α and β) from population genetic patterns of adult coral colonies from 28 sites on the Great Barrier Reef (GBR). And finally I estimate larval dispersal and retention over contemporary timescales by genetically assigning newly settled recruits of P. damicornis Type α and β, as well as of S. hystrix, to potential source populations. I found both high genetic similarities over large spatial scales and high genetic differentiation at local spatial scales among populations of P. damicornis Type α and β. I explain the high genetic similarity among populations separated by hundreds of kilometres by the potential of brooded larvae to have long pelagic larval duration (PLD) combined with strong longshore currents along the GBR. Further, I hypothesise that genetic differentiation among populations over small spatial scales is a signature of more recent single recruitment events from genetically distinct sources combined with the spatially and temporally stochastic nature of recruitment. I found no evidence for predominantly clonal reproduction within populations, despite the large energy allocation of this species to asexually produced brooded larvae, further undermining the paradigm that brooded larvae settle close to parent colonies shortly after release. Genetic assignment of recruits to potential source populations confirmed that larval dispersal and retention is site-specific for both lineages of P. damicornis studied, as well as for S. hystrix, in accordance with the capacity of brooded coral larvae to both settle immediately upon release, and stay in the water column for extended periods of time, and the spatial and temporal differences in current regimes between locales. Strongly self-recruiting populations of P. damicornis Type α were found at exposed sites in the Palm Islands, while predominantly self-recruiting populations of P. damicornis Type β and S. hystrix were located on sheltered sites in the Palm Islands, and at the NE exposed site of Lizard Island; other populations were found to be largely dependent on exogenous larval replenishment. Differences in habitat preference and dispersal characteristics between P. damicornis Type α and β support the recent delineation of lineages within the morphospecies, and demonstrate the timeliness of reevaluating the biology of several coral species using units of observation identified by techniques of molecular taxonomy. My results suggest that as a consequence of their complex reproductive strategies and dispersal capacities, pocilloporid species studied here likely form resilient metapopulations on the GBR. Because different habitat types sustain reproductive hotspots of these closely related species, it is reasonable to suggest that whole reefs are included in MPAs, rather than parts of reefs. Genetic connectivity estimates from my study, coupled with realistic high-resolution water circulation models may provide guidelines for suitable spacing of protected areas in inshore, midshelf and outer reef environments of the GBR, that ensure ecologically and demographically relevant connectivity between populations of brooding coral species