The study area, the KwaZulu-Natal Bight, is an oceanographically important area, which, regardless of having two of the most important fisheries off the east coast of South Africa, has received little research attention regarding its biological functioning. Until now chiefly oceanographic processes have been considered the drivers of this generally oligotrophic system. This study seeks to understand which of three important processes, a topographically induced oceanic upwelling cell near Richards Bay, a cyclonic eddy near Durban, or fluvial fluxes centred around the Thukela River, forces ecological functioning through their nutrient or organic matter input. The overall aim of the thesis is to understand the pelagic and benthic ecosystems of the Bight in terms of these drivers through the use of stable isotope (δ¹³C and δ¹⁵N) analyses of a range of biotic and abiotic samples. These were collected on board of a number of research cruises – forming predominantly part of the larger African Coelacanth Ecosystem Programme suite of studies – in the wet and dry seasons of 2010. Isotopic analyses found distinctions between fluvial and oceanic particulate organic matter and indicate that upwelling was not occurring in either sampling season. Organic matter originating from the Thukela River did not play a significant role in the wet season, although it dominated the planktonic pelagic food web in the dry season. The organic matter of the most productive region in the Bight, the Middle Shelf, was of riverine origin in the dry season, but of indeterminate origin in the wet season when it may have been an artefact of an old upwelling event which had previously occurred to the north of the Bight. There is, however, some evidence suggesting that this organic matter may rather have been of riverine origin, with its δ¹³C signals subsequently having been modified by the diatom bloom occurring there. In the demersal ecosystem, sediment isotopic data show organic matter to be well-mixed throughout the Bight in both seasons, with riverine organic matter dominating most of the Bight except its northern and southern edges, where oceanic organic matter increases in importance. Sediment organic matter (most likely via the macrobenthic biota) was deemed an important food source for demersal animals and omnivory an important feeding strategy. Seasonal studies from 2008 to 2010 in the Thukela Bank area indicate that the demersal animals' stable isotope signatures responded to the seasonal isotopic changes in riverine organic matter, indicating the cross-seasonal importance of this food source to the demersal ecosystem. Parallel methodological studies examined how routine isotopic sample handling procedures could have affected the results of the ecological studies. These studies suggest that i) effects of preservation/fixation methods and the use of dyes are species-dependant; ii) acidification has no effect on zooplankton isotopic signatures, and that iii) drying methods alone and interactively with multiple thawing and refreezing of samples affect the stable isotope values offish muscle tissues. Recommendations are made for further improvements in methodology and considerations to be taken when processing samples. Overall, it is concluded that riverine input to the Bight has a more important biological role than previously thought, and that organic matter from this source is an important driver of ecosystems within the Bight throughout the year for the demersal and pelagic ecosystems. / Thesis (Ph.D.)-University of KwaZulu-Natal, Westville, 2012.

Ecosystem modelling allows for an understanding of the structure and functioning of ecosystems. During this study, the oligotrophic KwaZulu-Natal (KZN) Bight, a data-limited system on the east coast of South Africa, was modelled. A framework for modelling data-poor systems, incorporating the construction of multiple models, sensitivity analyses and comparative analyses was applied to the Bight using literature data. Models converged on general trends of ecosystem functioning showing 99% of flows originated from detritus, primarily imported from rivers. The largest source of riverine detritus is the Thukela River which flows into the central Bight. This area supports a shallow-water prawn trawl fishery which targets penaeid prawns. Fisheries time series‘ were incorporated into the model framework to study the effects of prawn trawling and the decrease in prawn recruitment, caused by estuarine nursery loss, on the central Bight ecosystem. Dynamic simulations suggest the biomass of biotic groups were more affected by prawn recruitment level than trawling effort level. To understand the importance of nutrients in more detail, nutrient content, biomass and stoichiometric ratios were documented for various pelagic and demersal functional groups, and compared between areas in this oligotrophic system. Results showed the central Bight had the highest carbon, nitrogen and phosphorus biomasses, due to riverine nutrient sources, and the southern Bight had the lowest. In addition, the demersal community had higher biomasses than the pelagic community for all nutrients. Nutrient dynamics and limitations within the Bight were explored through the construction and analysis of trophic flow networks of carbon, nitrogen and phosphorus for the southern, central and northern Bight. Network analyses suggest nutrient cycling was lowest in the central Bight, and highest in the southern Bight. Cycling of nitrogen was highest in all areas due to the dominance of benthos, in terms of biomass, which was nitrogen-limited. Higher trophic levels were found to be phosphorus-limited. However many pelagic groups were co-limited by nitrogen and phosphorus, probably due to the oligotrophic nature of the bight. This suite of ecosystem models provides the first holistic view of the KZN Bight and an understanding of ecosystem functioning in the southern, central and northern Bight. / Thesis (Ph.D.)-University of KwaZulu-Natal, Westville, 2012.

The KwaZulu‒Natal Bight is formed from a narrow indentation in the SE coast of South Africa with the waters within considered to be oligotrophic. These waters therefore depend on both allochthonous sources of nutrients such as intermittent upwelling of deeper water and nutrients supplied by riverine inputs, as well as the autochthonous nutrients supplied by phytoplankton production, microbial fixation and recycling of nutrients by the microbial loop. Two African Coelacanth Ecosystem Programme cruises were undertaken during 2010, during the wet summer, and dry winter months. During each cruise, the waters of the KZN‒B were sampled rapidly to provide spatial scales (synoptic) of bacterial abundance and biomass, as well as at four predetermined locations to determine temporal scales (focus) of bacterial abundance, biomass and productivity. During the synoptic section, samples were taken in surface waters, close to F‒max (the depth at which phytoplankton were at their most dense as determined by in situ fluorometry), below the F‒max (where depths exceeded 50 m), and near the bottom. These samples were fixed with formaldehyde, stained with DAPI and cells were visualised by epifluorescent microscopy. During the focus section, samples were taken in surface waters, close to F‒max and below F‒max and incubated with 3H‒Thymidine to determine bacterial productivity. Bacterioplankton dynamics (numbers, biomass and productivity) for both cruises, synoptic section, were higher within the photic zone and near riverine influenced waters, with summer showing higher dynamics than winter. Irrespective of season, bacterioplankton dynamics decreased with increasing distance from the coast as well as with increasing depth, potentially via bottom‒up control mechanisms. Results obtained from the focus section of both cruises showed a significant difference between seasons for the Thukela Mouth and Richards Bay North, while no difference at the Durban Eddy. These results from the focus section suggest that bacterioplankton temporal dynamics were more top‒down controlled, rather than environmentally influenced, resulting in fluctuating dynamics over time. Overall, it is proposed that the degree of inorganic nutrient supply to the phytoplankton, resulted in the formation of DOM for use by the heterotrophic bacteria, resulting in a bottom‒up control mechanism, where Chl‒a concentrations within the euphotic zone induces either top‒down or bottom‒up control mechanisms on the heterotrophic bacteria directly affecting their numbers, biomass and productivity. / Thesis (M.Sc.)-University of KwaZulu-Natal, Westville, 2013.