Population structure, connectivity and ecological dynamics of the Antarctic silverfish, Pleuragramma antarctica

The Antarctic silverfish (Pleuragramma antarctica) is a keystone species in the continental shelf waters around the Antarctic, performing an essential role of connecting higher and lower trophic levels in the Southern Ocean ecosystem. Its early life history is dependent on the platelet ice layer found below sea ice, thus intimately intertwining its fate with that of sea ice extent. Antarctic silverfish belong to the family Nototheniidae, part of the Notothenioidei suborder whose species radiation in the Southern Ocean 24 million years ago is one of the most expansive among teleost fish. Most notothenioids inhabit a benthic niche as adults, though many experience a pelagic egg and larval phase. Antarctic silverfish are unique among notothenioids in that they are pelagic throughout their life history. Larvae develop in the platelet ice layer near the surface beneath sea ice, descending into deeper waters as they grow in size as juveniles, finally reaching their maximum depth range as adults at 400 – 700 m below the surface. While they lack a swim bladder, Antarctic silverfish manage to remain in the water column as adults by a type of paedomorphy in which they retain lipids from larval and juvenile life stages, allowing them to achieve neutral buoyancy. Despite their presence in the water column as adults, they practice a similar energy-efficient life strategy to their benthic counterparts. Their feeding strategy involves hanging in the water column and passively consuming prey. Remaining in the water column throughout their life history combined with their passive life strategy renders Antarctic silverfish especially susceptible to transport via local and circumpolar current systems. Thus, local and circumpolar current systems form the hydrographic framework in which hypotheses regarding Antarctic silverfish population connectivity must be tested. How populations of fish are defined, and the extent to which separate populations exchange individuals forms the basis of marine fish population biology. The extent to which Antarctic silverfish, which have a circumpolar distribution, represent one fully connected, panmictic population around the Antarctic continent, remains an open question. It is reasonable to presume that, given their pelagic larval phase, many species of notothenioids with a circumpolar distribution represent large, homogeneous populations. This presumption remains the null hypothesis to test when investigating population structure in notothenioids, and it is especially salient when considering the fully pelagic Antarctic silverfish. The first investigation into Antarctic silverfish population structure employed mitochondrial DNA markers on a circumpolar scale and did not find evidence to reject the null hypothesis of panmixia throughout the Southern Ocean. Intriguingly, while comparisons between regions failed to indicate differentiation, comparisons within regions between years hinted at inter-annual variation in patterns of connectivity. Genetic differentiation within the same geographic area that is lost and gained between sampling years due to variations in recruitment, mortality and hydrography is known as chaotic genetic patchiness. The extent to which chaotic genetic patchiness is relevant to understanding Antarctic silverfish population structure was further studied in a more recent investigation, restricting its geographic focus to the Antarctic Peninsula, and employing a set of highly polymorphic EST-linked microsatellite markers to understand population connectivity around the Antarctic Peninsula. Based on both its more focused regional scale and sampling scheme, as well as the use of genetic markers more adept at capturing population differentiation, this study was able to detect genetic structuring on the scale of the Antarctic Peninsula. Building upon these initial studies, this thesis aimed to characterize the circumpolar population structure of Antarctic silverfish, integrating aspects of life history and hydrography in order to describe mechanisms of connectivity between populations. The first aim was to understand the hydrography underlying life history connectivity on the scale of the Ross Sea region in order to better understand what may be occurring on the circumpolar level. Silverfish larvae were collected from areas in the Ross Sea coincident with hydrographic features hypothesized to influence their connectivity. While a microsatellite-based analysis was precluded due to the poor state of preservation of the larvae, it was possible to confirm species identification using mitochondrial sequence analysis. The genetic confirmation of species was especially important given that this study proposed a new spawning ground for silverfish in the Ross Sea based on size at collection and established growth rates from the time of hatching. Importantly, this study provided renewed support for the life history hypothesis in silverfish, emphasizing the impact of trough circulation in transporting early life stage fish from the ice shelf edge to the continental slope, where retention back towards the coast or entrainment in shelf-long currents modulates connectivity between neighboring populations of silverfish. The Ross Sea investigation was then expanded on a circumpolar scale, now carried out using a suite of highly polymorphic EST-linked microsatellite markers developed in a closely related notothenioid species and shown to successfully amplify in silverfish in a previous study. This analysis was carried out on fish collected over 25 years from six different regions: the western Ross Sea, the eastern Weddell Sea, Larsen Bay, the northern Antarctic Peninsula, the South Orkney Islands, and the western Antarctic Peninsula. The data analyzed included samples from the two previous investigations of silverfish population structure described earlier, the first using mitochondrial markers on a circumpolar scale that had not found evidence of population structuring, and the second using the same suite of microsatellite markers employed in this thesis on a regional scale around the Antarctic Peninsula. The integration of these previous datasets into the present analysis allowed for an increase in the resolving power of the previous mitochondrial marker-based study, as well as for the integration of the Antarctic Peninsular work into the greater circumpolar context. The circumpolar investigation of Antarctic silverfish population structure confirmed that the population structure of silverfish on a circumpolar scale is characterized by high levels of gene flow, and suggested that the Antarctic Slope Front and Current System (AFS) plays an integral role in connecting populations in the Southern Ocean. The importance of the AFS was evident in that reductions in gene flow were only observed in the South Orkney Islands and west Antarctic Peninsula, which were the only two areas in the study where the AFS has not been shown to arrive. This result also expanded to a circumpolar scale the earlier Ross Sea study, which had emphasized the importance of the AFS in connecting Ross Sea populations between local trough systems. It remained however, that small scale population differentiation which had been observed in the Ross Sea based on larval distributions, as well as in the eastern Weddell Sea based on the distribution of older and younger cohorts between sampling areas, was unable to be resolved using genetic techniques. Thus, the final aim of the main project of this thesis was carried out in order to resolve population structure on the regional scale, this time in the Weddell Sea, employing otolith chemistry. Analysis of trace element deposition in otolith nuclei, reflective of oceanographic conditions to which fish were exposed in early life, has been shown to delineate population structure in the Southern Ocean, in both silverfish and related notothenioids. Of the stations for which samples were available in the Weddell Sea, five stations were selected based on their locations with respect to hydrographic features hypothesized to influence population structuring in the region, in Atka Bay, Halley Bay, off of Coats Land, and west and east of the Filchner Trough. Previous studies, as well as data on biomass and abundance from the sampling expedition during which the silverfish were collected, emphasized the importance of the Filchner Trough in supporting a local population of silverfish in the eastern Weddell Sea continental shelf area. Furthermore, hydrographic data collected in the Weddell Sea emphasized the importance of warm water mass intrusion onto the continental shelf carried from the east into the Weddell Sea region by the AFS. These warm water intrusions from the AFS not only have the potential to carry fish from other regions into the Weddell Sea area, but regulate circulation patterns and the strength and directionality of coastal currents in the region, modulating local connectivity. The results of the otolith nucleus chemistry analysis revealed significant population structuring along the eastern Weddell Sea, in contrast to the structure revealed using genetics. The population structure revealed by the otolith chemistry analysis supported the importance of warm water intrusions from the AFS in transporting fish between areas, while highlighting the role of the Filchner Trough circulation in supporting a coherent population in the southeast Weddell Sea. These results emphasize the importance of the integration of multidisciplinary techniques in the context of local hydrography in addressing questions of population structure and life history connectivity in Antarctic silverfish in the Southern Ocean, and for that matter, any pelagic species inhabiting a continental shelf ecosystem.