Includes bibliographical references (leaves 24-26). / Copepod abundance data were collected during the annual pelagic fish spawner and recruit stock assessment surveys, in summer and winter from 1988 to 2000. These data were used to examine temporal and spatial variability in abundance and community structure of copepods in the southern Benguela, by estimating changes in abundance (No. mol) of individual species and size-classes. The relative abundance of Calanus agulhensis, Calanoides carinatus, Centropages brachiatus, Metridia lucens. Oithona spp., On cae a spp., copepod nauplii and other small copepods show that these taxa were important in driving total copepod abundance changes in the region. A general linear model (GLM) was used to help control for different factors that might affect interannual variability in copepod abundance: season, area, chlorophyll concentrations and sea surface temperature. The southern Benguela was divided into six sub-areas. Results of the GLMs show the largest abundances of copepods on the northern and central West Coast, fewer copepods on the central and eastern Agulhas Bank and fewest copepods on the south West Coast and western Agulhas Bank. Relative abundance time-series of four cope pod size-classes (cyclopoid copepods «900 μm total length (TL )), small calanoid copepods (900-1000 μm TL), medium calanoid copepods (1000-2000 μm TL) and large calanoid copepods (2000-5000 μm TL)) show that there was a possible change in control mechanism of copepods in 1996 from bottom-up to top-down. Moreover, in winter 1999 there was a unique event during which the abundance of various species showed a marked decline, but returned to normal levels in 2000. The time series of relative abundance showed reduced numbers of all four size-classes in 1992 1994. These findings indicate that factors affecting interannual variability of copepods are complex, but that, similar to previous findings, the southern Benguela system has not undergone a major change over the past two to three decades in terms of total copepod abundance in the different sub-areas.

Includes bibliographical references (leaves 25-32). / GPS telemetry in conjunction with a recent diet time series and historical dietary informationwas used in this study to obtain novel insight into the dietary trends and feeding ecology of Cape gannets Morus capensis on the Namibian islands. particularly Ichaboe and Mercury. The gannet diet has changed substantially since the 1950s. refiecting spatio-temporal changes in the availability of commercially important pelagic prey species. The more recent diet time series for Ichaboe Island (Nov 1995 to Feb 2004) showed that trawler scavenged hakes Merluccius spp and naturally foraged saury Scomberesox saw'us dominated the diet by both contribution to mass (35 and 34 %, respectively) and frequency of occurrence (34 and 25 %, respectively). In a significant contrast, juvenile horse mackerel Trachurus trachurus capensis (40 % mass, 26 % frequency) and juvenile snoek Thyrsites atun (20 % mass, 20 % frequency) were the two main prey species at Mercury Island during Oct 1996 to Feb 2004. Multivariate analysis of data showed significant time and site effects in diet composition between Ichaboe and Mercury for the period Nov 1996 to Mar 1999. Twenty-five and 15 GPS field deployments were made on birds at Ichaboe and Mercury, respectively, during the 2003/4 breeding season. Birds from both locations showed significant differences in their foraging patterns. Birds from Ichaboe had shorter foraging trips (24.3 hrs vs 29.4 hrs),traveled shorter distances away from their island (130 km vs 197 km) and had shorter foraging path lengths (422 km vs 673 km). Birds from Ichaboe foraged in two dominant directions: west. to obtain mainly scavenged fish offal, and north to obtain forage fish. Mercury birds foraged only north, overlapping their foraging zone with birds from Ichaboe in a northerly direction, between 25.8° Sand 24.4° S. Birds at all colonies - especially in the south - appear to be constrained by lower quality food and generally poor feeding conditions which seem to be a limiting factor.

Includes abstract. / Includes bibliographical references. / The potential of phytoplankton particle spectra to be used as the basis for an indicator of the suitability of feeding habitat for fish recruits in the Southern Benguela was investigated. Phytoplankton samples collected on regular cruises on the St Helena Bay Monitoring Line (SHBML) off Elands Bay on the west coast of South Africa had been analysed with the Coulter Counter and formed the basis of this study. Chlorophyll a content of phytoplankton samples was also measured on monthly cruises and with total particle concentration (determined by the Coulter Counter) showed that overall phytoplankton cells represented a significant portion (76%) of samples collected. Surface particle spectra were constructed for the 12 stations on the SHBML for each of 15 monthly cruises between September 2000 and February 2007.

This study investigates the hydrographic and dynamic properties of the continental shelf region between Cape Juby (28.5 °N) and Agadir (30.4 °N) within the Moroccan Sub-region of the North West African Upwelling System. Data came from two cruises conducted in June (beginning of summer) and November (end of autumn) 2013. Coastal upwelling was obvious in both cruises in the in-situ temperature and salinity data as well as in remotely sensed sea surface temperature maps. ADCP data showed the presence of a strong jet like current associated with enhanced upwelling off Cape Juby. This strong quasi-permanent upwelling center was observed during both cruises. It results from the orientation of this portion of coast which is aligned with the dominant wind direction, as well as a wind intensification near the cape. The presence of a secondary upwelling front was also observed near the shelf break. It was accompanied by an intense baroclinic jet. The EK60 data showed evidence of internal waves as well as small and mesoscale turbulence that were probably strongly interacting with the mean upwelling circulation and made a straightforward interpretation of the data quite challenging. However, this study revealed the main physical processes of this poorly studied region, as well as their seasonal variability.

Sharks are vulnerable to exploitation as a result of their biological characteristics. Markrecapture models were applied to conventional tag recapture data and acoustic telemetry data to estimate abundance, apparent survival, recapture probability and temporary emigration for the pyjama shark, Poroderma africanum in Mossel Bay, South Africa over a five-year period. This study applied Pollock's robust design (with the conventional tag data) and Cormack-Jolly-Seber (CJS) models (with the acoustic tag data) to analyze the mark-recapture data. In addition, a von Bertalanffy model was fit to the data to estimate individual growth. The best-fit robust design model showed the population as having no temporary emigration, survival probability that is dependent on the length at first capture, and time-constant capture probabilities. The best-fit CJS model showed the population also having time-constant survival, but sex dependent capture probabilities. Robust design abundance estimates (with 95% C.I.) in Mossel Bay varied from 279 (102-787) sharks to 733 (320-1777) sharks, although confidence intervals were quite large. CJS apparent annual survival (95% C.I.; CJS) was estimated to be 0.271 year⁻¹ (0.04 to 0.56) and annual recapture probability (95% C.I.) was estimated to be 0.008 year⁻¹ (0.003-0.20), indicating that survival and recaptures for this endemic species are relatively low. Annual somatic growth rate (k) was estimated to be 0.213 year⁻¹, indicating that this population is slow growing, a characteristic common in most shark species. Overall, the results in this study provide baseline knowledge on this population in Mossel Bay and can be used to implement proper management techniques. This knowledge can be further expanded upon to give a more in-depth understanding of all size and age classes in the population and the role that the environment and anthropogenic activities play in the population structure.

Includes bibliographical references (p. 51-59) / A retrospective statistical analysis of catch and effort data for sharks captured in protective gill nets along the KwaZulu-Natal coastline was carried out with the objective to establish if the frequency of shark catches is related to different phases of the moon and secondly to season.

This thesis assesses the relationships between the El Niño-Southern Oscillation (ENSO), rainfall and South African commercial winter wheat yields from 1974 - 2000. The analysis is through a combination of the Pearson's product moment correlation coefficient and an assessment of the magnitude and consistency of rainfall and wheat anomalies in the year of, and the year following warm (El Niño) and cold (La Niña) ENSO events. The ENSO - rainfall relationship is analysed on timescales from 1 - 24 months and this study finds that there are more unusually dry and unusually wet months during El Niño events than during La Niña episodes (where dry and wet months are <75°/o and over 25~·'0 more than the 1921 - 2000 mean, respectively) and these conditions are consistently found in some rainfall areas during each El Niño or La Niño event; there is marked inter - El Niño and inter La Niña variation in late summer and annual rainfall; and that there has been a significant shift in the timings of maximum rainfall anomalies during El Niño episodes from an earlier investigation. The effects of rainfall on the wheat yield vary spatially, but are most apparent in the Free State where severe droughts have resulted in reduced yields. In the Northern Cape and Western Cape anomalously wet conditions, especially in late summer, coincide with reduced wheat yields. The ENSO - wheat yield relationship is not a simple linear one, and despite yields in the Northern Cape and Western Cape tracking Pacific sea surface temperatures by nearly a year, El Niño and La Niña years are not synonymous With increased or decreased yields in any province. ln fact, maximum and minimum yields in the Free State and Northern Cape are found in the year of, or year following a La Niño event, and consequently the present predictability of wheat yields by ENSO is limited. The relationships between ENSO, rainfall and wheat yield in South Africa is not readily apparent, which may be due to the short (<30 years) data set or mediating factors outside this study such as farm - management strategies or hemispheric variation in the evolution of El Niño and La Niña events.

Considering the continual dissemination of marine non-indigenous (NIS) species across biogeographic borders, little is known about the status of these species within the network of South African marine protected areas (MPAs). Using the most recent data, a target list of NIS was set up for each MPA. Nineteen of the 23 South African MPAs were surveyed intertidally for alien and invasive species. The intertidal zone was divided into high-, mid- and low-shore and surveys were conducted during spring low tide. The presence and location in the intertidal zone of alien and invasive species were recorded. Additionally, invasive species' biomasses were quantified at sites at which they were found. Classification and regression tree (CART) analysis was used to assess which factors result in high numbers of NISs within MPAs. The invasive Mediterranean mussel, Mytilus galloprovincialis, was the most widespread species, occurring in 13 of the 19 sampled MPAs. It was most prominent on the island MPAs of the West Coast National Parks and had an overall highest biomass in the mid-shore. The bryozoan, Bugula dentata, was also widespread - present in MPAs across all three ecoregions. The invasive acorn barnacle, Balanus glandula, and bisexual mussel, Semimytilus algosus, both exhibited high levels of invasion in MPAs on the west coast. Extensions of known ranges were recorded for a number of species: the bryozoans Bugula dentata, Watersipora subtorquata and Cryptosula pallasiana, the polychaete Neodexiospira brasiliensis, the amphipod Orchestia gammarellus and the hydrozoan Obelia dichotoma. Certain species were recorded outside of harbours for the first time: the hydrozoans Obelia dichotoma and Obelia geniculata, the bryozoan Cryptosula pallasiana and the ascidians Microcosmus squamgiger and Diplosoma listerianum. CART analysis indicated that the size of the nearest port was an important indicator of the number of alien species in an MPA. When the nearest port is bigger than 0.4 km², more alien species are likely to occur within that MPA. For invasive species, the distance to the nearest yacht marina was the most important factor, with MPAs within 3.7 km of a yacht marina being more likely to have more invasive species in their borders. For all NISs, the highest numbers in an MPA were expected when the nearest port was greater than 2.1 km² and the nearest yacht marina was within 3.7 km of that MPA. Using these findings, spatial planning of future MPAs can further be prioritised to minimise the risk of introduction and spread of NIS therein. MPAs at risk as defined by these findings require structured monitoring regimes. A proactive measure would be establishing an interdisciplinary forum between relevant management authorities in order to enable dissemination of information on NIS. While controlling established NIS is difficult, the creation of task groups to act as rapid response teams for NIS, and the possibility of small-scale fisheries from edible invasive species are practical avenues that could be explored as mitigation efforts.

Includes bibliographical references. / A total of 553 male southern African sardine Sardinops sagaxcollected between 2010 and 2012 from around southern Africa, including from the northern Benguela off Namibia and from the South African west, south and east coasts, were examined for the occurrence of a testicular coccidian Eimeria sardinae to assess its potential use as a "biological tag" in order to differentiate between hypothesized discrete sardine stocks within the region. Sardine from Namibia and from the South African east coast showed no infection by E. sardinae, whereas fish from the South African west and south coasts were infected. Average prevalence infection values of 48.9 ± 6.3 and 48.1 ± 9.7% were recorded for sardine from the west and south coasts. Average infection intensity of 49.5 ± 15.8 oocysts per sample was recorded for the west coast and 69.8 ± 33.5 oocysts per sample for the south coat sardine. There was no significance difference in E. sardinae infection (t=0.08; p=0.94) in S. sagax from the South African west coast and south coast. These results provide some support for the hypothesis of multiple sardine stocks off southern Africa and hence are important for fisheries management. The intensity of infection had a significant (Analysis of Variance-NOVA, p<0.05) negative effect on the gonadosomatic index values of infected fish from both the west and south coasts. There was a declining trend in condition factor values with increasing intensity of infection category, however no significant effect was observed. The marked negative effect of infection intensity on GSI, with heavy infection possibly leading to parasitic castration, has potentially important implications for sardine population dynamics and requires futher research.

Includes bibliographical references (leaves 27-29). / The silver kob, Argyrosomus inodorus, has been identified as a viable species for mariculture in South Africa due to its high demand and market value as well as the necessary biological traits such as high fecundity and fast growth rates. In order for the full scale commercial production to become a reality however, it is important that the environmental conditions required for optimum growth at each phase of the species development are fully understood. Until recently, this sciaenid species was misidentified along with the dusky kob, Argyrosomus japonicus, as Argyrosomus hololepidotus, a fact that resulted in a poor understanding of the life history characteristics of the species and therefore the implementation of a poor management plan. As a result, wild stocks have been heavily over-exploited with spawner biomass-per-recruit ratios estimated at 2.9-12.5% of the pristine level. This study hypothesized that juvenile silver kob would exhibit a preferred temperature when a choice was available to them, and that this preferred temperature might change with age. Two cohorts of juvenile silver kob were acclimatized at 18°C and then subjected to identical temperature preference trials using temperature gradient tanks where the temperature ranged between 16 and 21.5°C as well as to identical isothermal control trials where the temperature throughout the tanks was 18°C. Salinity was kept constant at 35ppt and photoperiod remained 12L: 120 throughout. After a day of acclimation, observations were done over a five day period until the distribution of fish within the tanks had stabilized and that final temperature preferendum could be determined. During this time, dissolved oxygen concentrations were measured continuously to ensure that oxygen did not become limiting but remained above 6.4ppt. Results showed that the juveniles did indeed exhibit a preferred temperature with a range of 16.3 to 17.6°C for the first cohorts, and 16.9 to 18.1°C for the second. This difference between the preferred temperature ranges of the two age classes was found to be significant (p<0.01) by a Mann Whitney U Test. Kolmogorov-Smirnov tests showed no significant difference (p<0.01) between the tanks of the isothermal control trials indicating that there was no shoaling behaviour or individual tank bias affecting the results.