Stable isotope (SI) ratios of carbon (δ13C) and nitrogen (δ15N) are now widely used as biomarkers in ecological studies to provide information about food web structuring. However, understanding trophic relationships using SI analysis requires not only knowledge of SI values of predator and prey, but also accurate discrimination factors (DFs), which can differ among species and by physiological state. This thesis examined three questions using captive birds from the South African Foundation for the Conservation of Coastal Birds (SANCCOB). First, the effects of ontogeny on δ13C and δ15N ratios of African penguins (Spheniscus demersus) were assessed. Blood samples were collected from penguins in four age classes (P3 chicks, blues, juveniles and adults) concurrently with their diet (sardine (Sardinops sagax) and formula). Second, to assess the influence of breeding physiology on SI ratios, the blood of ten breeding pairs of penguins was sampled over a five-month period from June to October 2016. Following laying, each pair was categorised into one of three (four for whole blood) egg production phases (initial yolk deposition, rapid yolk deposition and post-laying) and their influences on SI ratios were tested. Third, species differences in DFs were evaluated for African penguins, kelp and Hartlaub’s gulls (Larus dominicanus and L. hartlaubii), greater crested terns (Thalasseus bergii) and Cape cormorants (Phalacrocorax capensis). Flying birds were mostly fed sardine with a small but unknown amount of sardinella (Sardinella aurita), DFs were therefore estimated for a 50:50 sardine:sardinella diet, a 75:25 sardine:sardinella diet and a 100% sardine diet for each flying bird species. The DFs were assessed for the whole blood (WB), red blood cells (RBC), plasma (PL) and delipidated plasma of the penguins, and only WB for the flying birds as well as flesh, whole fish, delipidated flesh and delipidated whole fish for fish species, and for formula. Results indicated that age influenced both the δ13C and δ15N of WB, only the δ15N of RBC and the δ13C of delipidated PL. The assessment of breeding physiology yielded a significant interaction between the effects of egg production phase and sex on the δ13C of WB; females had significantly lower δ13C in the rapid yolk deposition phase than the other two phases and all males. The δ13C of PL was affected only by sex, with females having a significantly lower δ13C value than males. Neither physiological state nor sex influenced the other blood components. Differences were found among the three DFs in the non-penguin species, but not for all consumer – prey tissue combinations. There were also significant differences among species with a DF calculated from a diet with the most probable prey proportions eaten. Depending on the combination of consumer and prey tissue used to calculate the DF, a different conclusion regarding trophic information can be reached. A literature review updated with the present data showed that no general pattern or grouping of similar species with regards to DF values could be drawn, highlighting the importance of determining species- and tissue-specific DFs. Thus age, egg production, tissue and species all influenced the SI values of bird blood and therefore their DFs. Not all physiological conditions affect all blood components in the same way, making different components more or less sensitive to physiological influences. Though their influence is at a small enough scale that it is unlikely to hamper correct conclusion in ecological studies, it is crucial that these factors are considered when using SI analysis (SIA). When uncertainties exist for some coefficients in wild studies, SIA should therefore be combined to other dietary techniques to determine the food web structure as best as possible.

This study investigated the potential impacts of several anthropogenic stressors on the movement and survival of African penguins in Algoa Bay, taking into account prey availability and environmental conditions (sea surface temperature and chlorophyll productivity). The primary drive for the study was based on the seismic exploration, which took place in the habitat of the endangered African penguin. The research was also carried out to understand how pelagic fishing and prey availability influence the broader behavioural and mortality aspects of penguins. To achieve this goal, I used monthly beach surveys to assess seabird mortality, admissions of penguins in rehabilitation centres and the monitoring of breeding penguins’ movement at sea through individual GPS tracking. In addition, prey distribution and abundance were monitored with acoustic surveys to assess potential changes in relation anthropogenic stressors. Breeding African penguins did not react dramatically to seismic activities, except on St Croix Island where they changed their foraging direction during the seismic period. There was no evidence of a siginficant increase in African penguin carcasses encountered or strandings along the beaches during the period of seismic activity. However, a sharp increase in encounter rate was recorded soon after seismic operations, which may potentially indicate a delay between the effect of seismic activities on penguin mortality and the encounter of the related carcasses after the operation. However, the impact of seismic surveys on penguins may be long-term rather than short term, and additional studies would be needed to reveal long-term effects of seismic activities, if any. Foraging effort of breeding African penguins increased substantially in the presence of commercial fishing activities within close proximity to the colonies. To a greater extent, at-sea movement patterns and counts of beach cast carcasses were largely influenced by prey availability. Penguins spent more time at sea, traveled longer and covered a larger foraging area during periods of relatvely low pelagic fish abundance. There was also a sharp increase in African penguin mortality during periods of low prey availability in the Bay. However, environmental conditions showed some influence over these interactions. Foraging effort of breeding African penguins increased substantially in the presence of commercial fishing activities within close proximity to the colonies. To a greater extent, at-sea movement patterns and counts of beach cast carcasses were largely influenced by prey availability. Penguins spent more time at sea, traveled longer and covered a larger foraging area during periods of relatvely low pelagic fish abundance. There was also a sharp increase in African penguin mortality during periods of low prey availability in the Bay. However, environmental conditions showed some influence over these interactions. For example a red tide event during in period of low prey availability, made it difficult to understand impacts of prey on penguins difficult. Both anthropogenic stressors and environmental conditions influenced African penguins’ behaviour and survival. Due to the worrying trend of African penguins in South Africa for the past few years, all conservation management efforts to increase penguin numbers and limit their mortality are necessary. This study highlights the negative impact of industrial fishing on this Endangered species and the rapid increase in mortality of penguins during times of low prey availability. A network of Marine Protected Areas would certainly increase food availability to African penguins by limiting competition with fisheries, and contribute to increasing the population numbers.

Cape St Francis, Eastern Cape, has been identified as one of four potential sites for establishing a mainland-based African penguin (Spheniscus demersus) colony. This thesis comprises three main components: a verification of a preparation method for stable isotope samples from penguin feathers; a dietary analysis of the penguins on Bird Island, Algoa Bay, though stable isotope analysis of whole blood and feathers (2012 and 2013); and an estimation of available fish surplus that could potentially support a colony of penguins at Cape St Francis. Each component contributes towards the next, all building towards answering the main research question: Will there be enough food around St Francis Bay to support a colony of penguins and sustain the already established fisheries industry within the bay? Stable isotope analysis of whole blood and feathers from breeding adults and whole blood from juveniles provided insight into the variability of African penguins’ diets at different stages in their life history. Stable isotope mixing models indicated that the predicted proportions that each prey species could potentially contribute to diet conflicted with published stomach sample data. This might arise from inaccurate trophic enrichment factors used in the model, or from systematic biases in the published stomach sampling techniques, or both. Dietary sexual dimorphism was not demonstrated by the isotope signatures of breeding penguins. Based on official catch data, the fisheries activity on the south coast, and especially around the potential colony site at St Francis, is much lower than around the west coast’s penguin colonies. The model provided a first-order estimate for fish supply around the potential colony site at St Francis both at a large coastal scale and a local small scale. At both scales the estimate indicated an ample availability of fish at current fishing levels. The model in Chapter 4 can also be applied to refining the assessments of other potential colony sites on the south coast. In conclusion, the south coast is a promising area for a new colony of penguins in terms of food availability. There is relatively low fishing activity in the area and, as suggested by the large-scale model in Chapter 4, an ample fish resource. The final chapter briefly discusses factors that need to be considered before attempting to establish a mainland-based colony of African penguins.