Genetic isolation by distance reveals restricted dispersal across a range of life histories: implications for biodiversity conservation planning across highly variable marine environments

AIM: Isolation by distance (IBD) analyses are an effective tool for determining genetic connectivity among populations, providing a basis for estimating dispersal and thus contributing to spatial biodiversity planning. Here, we use an IBD approach to determine patterns of connectivity to infer dispersal distances for a phylogenetically diverse range of marine species. LOCATION: Two thousand and eight hundred kilometre of South African coastline. METHODS: mtDNA data sets for 11 rocky shore and reef‐dwelling marine species with diverse life history strategies (broadcast spawning, brooding and live‐bearing) were used to quantify intraspecific genetic connectivity by means of global and pairwise Φₛₜ. Mantel tests were used to assess the significance of genetic structure with geographic distance. Significant, positive slopes were used to estimate dispersal distances. For all species, we also plotted genetic diversity along the coast and compared this to patterns of species richness. RESULTS: For seven of the 11 species tested, we report positive IBD relationships, with patterns of IBD of similar magnitude in broadcast spawners and live‐bearers. Dispersal estimates were low, with effective distances of < 10 km per generation. We find discordance between pelagic larval development and population structure and that genetic structure and the physio‐geographic features of the coastline considered to be important biogeographic barriers do not overlap. Genetic diversity and species richness show congruent patterns, with both variables lowest on the west coast, and increasing towards the east. MAIN CONCLUSIONS: Our study highlights the complexity of marine systems and the clear need for regional assessments of connectivity, rather than more general management plans which may not otherwise effectively encompass area‐specific conservation needs. Importantly, we show that the current marine protected area (MPA) network in South Africa is not effective in protecting evolutionary processes and strongly recommend a network of more closely spaced MPAs in the region that reflects the low average connectivity between distant marine populations.