Estimating the biodiversity of the sub-Antarctic Indian part for ecoregionalisation: Part I. Pelagic realm of CCAMLR areas 58.5.1 and 58.6

Two studies are tabled for the Subantarctic pelagic area of CCAMLR areas 58.5.1 and 58.6, which covers the Crozet Basin and the north of the Kerguelen Plateau: one based on marine pelagic species (plankton and fish), and the second on top predators.

In this report, we followed three types of methodology: (1) a taxonomic approach based on communities only, (2) a physiognomic approach used for the bioregionalisation of the Southern Ocean based on abiotic and chlorophyll-a characteristics of the habitat only, and (3) a mixed approach, we called “ecoregionalisation”, which includes taxonomic, ecological and physiognomic data.  The mixed approach is a relatively new method and is demonstrated on mesopelagic fish here. It will be also applied to pelagic fish and zooplankton as work progresses.

The Kerguelen Plateau influences the position of the major frontal zones and acts as a barrier to the Antarctic Circumpolar Current which passes along its northern escarpment and through troughs to the south. North of the Kerguelen shelf, the subantarctic zone does not really exist. Instead, there is a narrow latitudinal band, the Transition Frontal Zone (TFZ), caused by the juxtaposition of the Subtropical (STF) and the Subantarctic Front (SAF). The Polar Frontal Zone (PFZ), between the Antarctic Polar Front and the SAF, is closer to the TFZ in the Kerguelen shelf area than it is in the Crozet area. The Agulhas Return Current zone, north of the CCAMLR area, is composed of warm subtropical waters. Our studies indicate that the change in communities and regions is very sharp at the PFZ and the TFZ. The main biogeographical barriers in this area are the SAF and the Agulhas Front. Over the island shelves and seamounts, essential habitats for fish larvae were related to currents and retention zones.

In this report, we showed the existing information that we can use to achieve the pelagic ecoregionalisation of this area. There are numbers of difficulties as samples are from different historical surveys and mainly concentrated in the Kerguelen area.

Four variables were used for the epipelagic physical regionalisation: mean dynamic topography, sea surface height variability, depth, and near-surface chlorophyll-a. Eleven environmental types were apparent. The results show a general latitudinal banding structure broadly consistent with the frontal structure of the region, with the Agulhas Return Current clearly apparent. Shallow topography around the region’s island groups and bathymetric features (Conrad Rise, Walters Shoal) give rise to smaller-scale structure generally associated with higher productivity.

Some examples on ecoregionalisation based on mesopelagic fish are presented. The first one concerns the modelling of the myctophid community at the scale of the Indian sector of the Southern Ocean using Generalised Dissimilarity Modelling. 12 groups delineated ecoregions which were associated with frontal zones and the topography. A boosted regression tree model was used to predict potential habitat of target myctophids. The series of fronts and their associated eddies developed south of the Subtropical Front influences these species habitat. One of the target species was E. antarctica. The models indicated that the probability of presence was expected to be low in the north, consistent with the limit of the main distribution of the species. The quasi-absence of E. antarctica from island shelves is explained by its pelagic nature and diel behaviour. A modelling based on abundances showed that the preferential habitat towards the east differs in intensity from one year to the next.

Application of these methods on plankton and pelagic fish will be carried out in 2012. A report showing the application of these methods on the pelagic realm of the study area will be presented in 2012 at the EMM working group.