Phase III – A molecular consideration of Western Indian Ocean marine fish connectivity and differentiation at multiple spatial and temporal scales and its implications for conservation and resource management

There is insufficient understanding of the contemporary, ecological connectivity among MPAs across the Western Indian Ocean (WIO) region, which is an important knowledge for effective conservation and management of marine resources in the region. This deficiency hampers the integration of MPAs into an effective reserve network. Furthermore, the lack of clarity on the biogeographic and historical origins of the WIO fauna, and little knowledge of phylogeographic and biogeographic breaks prevents an understanding of the features that influence connectivity. Since it is difficult to study connectivity directly, a genetic approach provides a valuable and widely accepted, indirect means of examining connectivity across the region. The general aim of this study was therefore to examine the connectivity and regional intraspecific differentiation in several reef fish species across the WIO at various spatial scales, considering processes over ecological and evolutionary timeframes. Specific aims included: (i) To determine patterns and scales of genetic connectivity across the WIO (ii) To determine the relationships among the various regions of the WIO for nine target species with their shaping historic and biogeographic processes (iii) To contribute to the identification based on patterns emerging from the analyses of included taxa, unique biogeographic regions or phylogeographic provinces within the WIO, and (iv) To consider, incidentally, whether widespread species are biological realities and, consequently, address diversity and endemism in the WIO.

This study targeted the scientific community, conservationists as well as policy and decision makers. The study areas included localities in the WIO and the Pacific as follows: Maldives, Seychelles, Kenya, Tanzania (mainland and Zanzibar Island), Comoros (Mohéli and Moroni), Mayotte, Geyser, Glorieuses, Madagascar (St. Marie and Nosy Bé), Juan de Nova, Europa, Mozambique, South Africa, Réunion, Mauritius and Rodrigues

During this study a several scientific investigations were carried out. (i) Investigation of patterns of fine-scale genetic structure and scales of connectivity in three species of importance to regional fisheries, namely (Epinephelus merra, Lutjanus kasmira and Myripristis berndti) in order to identify open and closed systems and localities that are critically isolated using mtDNA sequence data and genotype data from eight to 13 microsatellite loci. (ii) Broad-scale patterns of differentiation and relationships among the regions of the WIO were investigated in nine reef species (Abudefduf vaigiensis, Cheilodipterus quinquelineatus, Dascyllus trimaculatus, E. merra, L. kasmira, M. berndti, Pristiapogon kallopterus, Scarus rubroviolaceus, and Sufflamen fraenatum), with a view to revealing unique biogeographic regions in the WIO. Genetic data were generated from two mitochondrial (mtDNA) gene fragments and one nuclear gene intron for six of the species.

Genetic studies revealed a number of important information as summarised below. The wider sampling range and global approach of the study allowed for the detection of structure and restricted connectivity that may have been missed by other studies. The study of fine-scale structuring in E. merra indicated low connectivity between Mascarene localities (Madagascar, Mauritius, Réunion and Rodrigues), Maldives and other localities in the WIO. The overall pattern for L. kasmira indicated a high connectivity and little differentiation across the WIO, and indication of a single genetic population. M. berndti on the other hand revealed more restricted connectivity within the south-western Indian Ocean.

Phylogenetic analyses revealed three unique lineages corresponding to species-level diversity within A. vaigiensis, with two lineages restricted geographically (to Kenya and Seychelles), and the third one widespread across the region. Two highly divergent lineages were identified in C. quinquelineatus, with the two occurring sympatrically and presenting some evidence of hybridization in the Kenyan locality. Low diversity and very little differentiation among most WIO sites were revealed within D. trimaculatus. High levels of diversity for all markers considered were revealed within P. kallopterus, but no significant structure among three widely separated localities (Glorieuses, Réunion and Zanzibar).

Scarus rubroviolaceus was characterised by low variability in mtDNA markers coupled with extreme diversity for the nuclear intron, however, there was neither evidence of differentiation among WIO localities, nor evidence of genetic structure between the WIO and the West Pacific. Sufflamen fraenatum showed a marked divergence between Pacific and WIO specimens, in which four groups (Mauritius, Seychelles, Tanzania and the remainder of the east African localities) could be defined.

The patterns documented here and the approach have application in conservation and management, particularly with regard to reserve selection and designation.

Future studies identified from this project include the adoption of a region-wide or global approach, the inclusion of multiple taxa, careful selection of molecular markers and analyses, with appropriate sample sizes, a more detailed consideration of physical features and the life histories of the taxa under study, and integration with other approaches for examining connectivity.