Understanding the pattern of connectivity among populations is crucial for the development of realistic and spatially explicit population models in marine systems. Here we analysed variation at eight microsatellite loci to assess the genetic structure and to infer patterns of larval dispersal for a brooding coral, Seriatopora hystrix, at an isolated system of reefs in northern Western Australia. Spatial autocorrelation analyses show that populations are locally subdivided, and that the majority of larvae recruit to within 100 m of their natal colony. Further, a combination of F- and R- statistics showed significant differentiation at larger spatial scales (2-60 km) between sites, and this pattern was clearly not associated with distance. However, Bayesian analysis demonstrated that recruitment has been supplemented by less frequent but recent input of larvae from outside the local area; 2-6% of colonies were excluded from the site at which they were sampled. Individual assignments of these migrants to the most likely populations suggest that the majority of migrants were produced at the only site that was not decimated by a recent and catastrophic coral bleaching event. Furthermore, the only site that recovered to prebleaching levels received most of these immigrants. We conclude that the genetic structure of this brooding coral reflects its highly opportunistic life history, in which prolific, philopatric recruitment is occasionally supplemented by exogenously produced larvae.

Approximately one quarter of zooxanthellate coral species have a depth distribution from shallow waters (<30 m) down to mesophotic depths of 30-60 m. The deeper populations of such species are less likely to be affected by certain environmental perturbations, including high temperature/high irradiance causing coral bleaching. This has led to the hypothesis that deep populations may serve as refuges and a source of recruits for shallow reef habitats. The extent of vertical connectivity of reef coral species, however, is largely unquantified. Using 10 coral host microsatellite loci and sequences of the host mtDNA putative control region, as well as ribosomal DNA (rDNA) ITS2 sequences of the coral's algal endosymbionts (Symbiodinium), we examine population structure, connectivity and symbiont specificity in the brooding coral Seriatopora hystrix across a depth profile in both northwest (Scott Reef) and northeast Australia (Yonge Reef). Strong genetic structuring over depth was observed in both regions based on the microsatellite loci; however, Yonge Reef exhibited an additional partitioning of mtDNA lineages (associated with specific symbiont ITS2 types), whereas Scott Reef was dominated by a single mtDNA lineage (with no apparent host-symbiont specificity). Evidence for recruitment of larvae of deep water origin into shallow habitats was found at Scott Reef, suggesting that recovery of shallow water habitats may be aided by migration from deep water refuges. Conversely, no migration from the genetically divergent deep slope populations into the shallow habitats was evident at Yonge Reef, making recovery of shallow habitats from deeper waters at this location highly unlikely.

Coral reefs are in decline worldwide, and marine reserve networks have been advocated as a powerful management tool for maximizing the resilience of coral communities to an increasing variety, number, and severity of disturbances. However, the effective design of reserves must account for the spatial scales of larval dispersal that affect the demography of communities over ecological time frames. Ecologically relevant distances of dispersal were inferred from DNA microsatellite data in a broadcast-spawning (Acropora tenuis) and a brooding (Seriatopora hystrix) coral at isolated reef systems off northwest Australia. Congruent with expectations based on life histories, levels of genetic subdivision among populations were markedly higher in the brooder than in the broadcast spawner. Additionally, significant subdivision for both species between systems (>100 km), and between (>10 km) or within reefs (<10 km) within systems, indicated that many reefs or reef patches are demographically independent. There was also a clear distinction in the scale of genetic structure between the different systems; at the more geographically complex of the systems, a much finer scale structure was detected in both species. This suggested that the hydrodynamics associated with these complex reefs restrict distances regularly traveled by larvae. The primary implication is that short-term recovery of these coral communities after severe disturbance requires the input of larvae from viable communities kilometers to a few tens of kilometers away. Therefore, to be self-sustaining, we suggest that coral reef protected areas need to be large enough to encompass these routine dispersal distances. Further, to facilitate recovery from severe disturbances, protected areas need to be replicated over these spatial scales. However, specific designs also need to account for size, complexity, and isolation of reefs, which will either restrict or enhance dispersal within this range.

Here we present nine novel, polymorphic microsatellite loci developed for the scleractinian coral Acropora millepora (Acroporidae) from the Great Barrier Reef. In addition, we have assessed the specificity and polymorphism of five microsatellite loci previously developed for a Caribbean congener and one locus developed for an Indo‐Pacific congener. Only one of the latter six loci produced reliable results, yielding a total of 10 polymorphic microsatellite loci for A. millepora. Variability was tested on 20–23 individuals from a single population, plus another ∼10 individuals from each of three different populations, resulting in five to 20 alleles per locus.

Understanding the evolutionary processes that have shaped existing patterns of genetic diversity of reef-building corals over broad scales is required to inform long-term conservation planning. Genetic structure and diversity of the mass-spawning hard coral, Acropora tenuis, were assessed with seven DNA microsatellite loci from a series of isolated and discontinuous coastal and offshore reef systems in northwest Australia. Significant subdivision was detected among all sites (FST = 0.062, RST = 0.090), with the majority of this variation due to genetic differentiation among reef systems. In addition, genetic divergence was detected between the coastal and offshore zones that cannot be adequately explained by geographic distance, indicating that transport of larvae between these zones via large-scale oceanic currents is rare even over time frames that account for connectivity over multiple generations. Significant differences in the amount of genetic diversity at each system were also detected, with higher diversity observed on the lower latitude reefs. The implications are that these reef systems of northwest Australia are not only demographically independent, but that they will also have to rely on their own genetic diversity to adapt to environmental change over the next few decades to centuries.

Coral spawning on the oceanic reef systems of north-western Australia was recently discovered during autumn and spring, but the degree to which species and particularly colonies participated in one or both of these spawnings was unknown. At the largest of the oceanic reef systems, the participation by colonies in the two discrete spawning events was investigated over three years in 13 species of Acropora corals (n = 1,855 colonies). Seven species spawned during both seasons; five only in autumn and one only in spring. The majority of tagged colonies (n = 218) spawned once a year in the same season, but five colonies from three species spawned during spring and autumn during a single year. Reproductive seasonality was not influenced by spatial variation in habitat conditions, or by Symbiodinium partners in the biannual spawner Acropora tenuis. Colonies of A. tenuis spawning during different seasons separated into two distinct yet cryptic groups, in a bayesian clustering analysis based on multiple microsatellite markers. These groups were associated with a major genetic divergence (G”ST = 0.469), despite evidence of mixed ancestry in a small proportion of individuals. Our results confirm that temporal reproductive isolation is a common feature of Acropora populations at Scott Reef and indicate that spawning season is a genetically determined trait in at least A. tenuis. This reproductive isolation may be punctuated occasionally by interbreeding between genetic groups following favourable environmental conditions, when autumn spawners undergo a second annual gametogenic cycle and spawn during spring.

No abstract

Here we report the isolation of 44 microsatellites from the brooding, pocilloporid coral, Seriatopora hystrix, developed from a partial genomic DNA library using a repeat enrichment protocol. A further eight previously published microsatellites were also tested; five of these were developed for S. hystrix, whereas three were isolated from corals of the closely related genus Pocillopora. Out of these, we incorporated nine and 10 primer pairs into two multiplex reactions that reliably amplified polymorphic microsatellites in populations from the west and the east coast of Australia, respectively. Number of alleles ranged from three to 22 per locus.