The panga, Pterogymnus laniarius (Cuvier, 1830), is a South African endemic sparid fish species. On the Agulhas Bank, South Africa it is a commercially important species, caught as bycatch in the hake directed trawlfisheries and targeted by offshore hook-and-linefishers. Recently there has been considerable interest shown in directing further fishing effort on this species. The lack of a suitable management procedure for teleost bycatch in South Africa was the principal reason for undertaking this study. This thesis investigates aspects the panga's life history, particularly those aspects that have management implications. A full knowledge of this species' distribution and abundance was necessary as this could highlight the existence of any nursery areas, ontogenetic migratory patterns and areas of high spawner biomass. The derived parameter estimates were then included as inputs into stock assessment models to determine the status and productivity of the resource. Growth studies based on sectioned sagittal otoliths revealed that the panga was a relatively slow growing fish with ages of 16 years being recorded. Growth was best described by the von Bertalanffy growth model as Lt=379.4(1-e⁻°·¹³⁽t ⁺ ¹·⁷⁸⁾). Total, natural and fishing mortalities were estimated at 0.36 year⁻¹, 0.28 year⁻¹ and 0.08 year ⁻¹, respectively. Detailed histological examination of the gonads revealed that panga is a late gonochorist, males and females maturing after a non-functional intersexual stage. Females mature at approximately 200 mm fork length or 4 years of age. Reproduction occurs throughout the year although there is a slight peak in winter. Gametogenesis was found to be similar to that of other sparid fishes and marine teleosts in general. The panga feeds predominantly on crustaceans with a distinct ontogenetic shift in feeding habits. Juvenile fish feed predominantly in the water column on mysids after which they move to the benthos. Subadult fish feed principally on ophiuroids and amphipods. Adult fish remain on or near the benthos, feeding predominantly on crabs, and on polychaetes, ophiuroids and fishes to a lesser extent. Several aspects of the panga's biology contribute to its ability to sustain a higher fishing pressure than other sympatric sparid species. These include its late gonochoristic reproductive style, protracted spawning season, maturation before recruitment and preference for soft substratum prey that enables it to utilise large areas of the Agulhas Bank. The panga's longevity, slow growth and low natural mortality rate, however, mitigated against these factors and were considered in the stock assessments. A heterogeneous Geographical Information System (GIS) was developed to analyse the distribution and abundance patterns of the panga. The GIS developed in this thesis makes a significant contribution towards the development of a South African Fisheries Information System to analyse and manage fish resources in general and bycatch resources in particular. The GIS developed in this study combines statistical Generalized Additive Modelling and standard GIS methods. Analysis of fourteen biannual fishery independent biomass surveys, disaggregated by life history stage, revealed that a nursery area for immature fish (<23 cm TL or < 4 years of age) exists on the Central Agulhas Bank. After sexual maturation, approximately 40% of the biomass migrated eastwards, colonising large areas of the Eastern Agulhas Bank.ilie location of the nursery area appears to be a result of the pelagic eggs and larvae being advected towards the coast in an anti-cyclonic gyre, stemming off the Agulhas current and later deposited over the Central Agulhas Bank. The weak bottom currents on the Central Agulhas Bank prevented juvenile loss to the Benguela system. The eurytopy of adult fish to various physical variables such as temperature, dissolved oxygen and stronger currents enabled it to reduce conspecific competition and migrate eastwards with distribution primarily determined by depth. Relative biomass estimates revealed a gradual increase in biomass of 5.5% per annum between 1988 and 1995. Predictions from yield-per-recruit, biomass-per-recruit and spawner biomass-per-recruit analyses showed that there was scope for further exploitation. A FSB₅₀ fishing strategy was considered to be the most appropriate fishing strategy as it did not reduce the spawner biomass-per-recruit to less than 50% of unexploited levels. Effort control was considered the most effective management tool as the age-at-50%-selectivity occurred after age-at-sexual maturity and releasing undersized fish was undesirable due to heavy mortalities resulting from severe barotrauma. The panga resource was also assessed using an age-structured production model. The values for the free parameters of the model were estimated using biomass indices derived from fishery-independent trawl surveys. Although the data were fairly uninformative about the productivity of the resource, the results indicated robustly, that the population has recovered from low levels in the mid-1970's and could sustain higher levels of fishing intensity. Risk analysis calculations were used to assess the sustainability of different catch scenarios. The level of sustainable catch was found to be sensitive to the selectivity pattern of the gear utilised. Both stock assessment methods used in this study to investigate the status of the panga resource showed that the resource could theoretically sustain higher catches. It was found that although the stock could be harvested using available fishing methods, the sympatry of this species with other commercial species was of concern as the latter would form a significant bycatch in a panga directed fishery. The failure of current harvesting methods to address the bycatch problem highlights the management problem in South Africa and stresses the need for creativity by both scientists and fishers in designing new and improved methods for selectively harvesting bycatch fish resources. Considering that no suitable method is currently available to fish the panga stock in a directed fishery the fishery should be managed as status quo until some suitable and efficient gear is developed.

Over the past few decades it has become recognised that an ecosystem approach is required to manage world fisheries. Management strategies must ensure that non-target (bycatch) as well as target catches are sustainable. To achieve this, detailed commercial catch and biological information is required. The composition of catches made by trawlers operating off the south and west coasts of South Africa was investigated. Distinct fishing areas were identified on each coast, based on target species and fishing depth. Catch composition differed markedly among the areas defined. Although hake Merluccius sp. dominated South Coast catches, a large proportion of the catch was composed of bycatch. On the West Coast, hake dominated catches and this domination increased with increasing depth. On both coasts approximately 90% of the observed nominal catch was processed and landed. Estimates of annual discards suggested that the fishery discarded 38 thousand tons of fish per annum (16% of the nominal trawl catch). The data also indicated that hake discarding, the capture of linefish and the increased targeting of high value species might be cause for concern. Spatial analysis indicated that a variety of factors such as trawling position, catch size and catch composition affects bycatch dynamics. The monkfish Lophius vomerinus is a common bycatch species that has been increasingly targeted by demersal trawlers. This study showed that L. vomerinus is a slow-growing, long-lived species (West Coast males L∞ = 68.50cm TL, t₀ = -1.69yr, K = 0.10yr⁻¹; West Coast females L∞ = 110.23cm TL, t₀ = -1.54yr, K = 0.05yr⁻¹; South Coast sexes combined L∞ = 70.12cm TL, t₀ = - 0.80yr, K = 0.11yr⁻¹), that matures at approximately 6 years of age. These traits could have serious management implications for the species. Per-recruit analysis suggested that the stock might be overexploited, although further investigation is required to confirm this. Solutions were suggested for each of the concerns raised, taking cognisance of the differences observed between the South and West Coasts and the economic dependence of South Coast companies on bycatch. The needs of future research were considered.

The bycatch of cetaceans in fishing gear is considered to be one of the biggest conservation threats to these species. Gear modifications have the potential to reduce these bycatches in global fisheries but there is little available information on how such modifications may change the fishing performance of gear, or indeed the behavior of cetaceans interacting with fishing gear. Generalized linear models (GLMs) were used to identify factors related to cetacean bycatches in UK bottom set gillnets. Rigged net height had a significant positive relationship with harbour porpoise (Phocoena phocoena) bycatch in ICES Area VII suggesting that lowering the profile of gillnets may have the potential to reduce bycatch rates. Modifications to gillnets, such as changing the amount of floatation or increasing the density of the meshes, were found to have significant effects on the active fishing heights of these nets. However, results from a bycatch mitigation trial in Argentina showed that the reduced fishing profile of one experimental net did not result in a concurrent reduction in the bycatch rate of Franciscana dolphins (Pontoporia blainvillei). While there was no significant difference in the rate, length or intensity of harbour porpoise encounters in the presence or absence of gillnets, the proportion of fast echolocation click trains were significantly higher when a net was present, indicating that porpoises either increased acoustic inspection of the net or foraging in the vicinity of the net. An analysis of underwater video footage collected inside trawl nets in an Australia fishery showed that bottlenose dolphins (Tursiops sp.) were present inside nets more frequently than they were caught and were actively foraging inside these nets. The orientation of dolphins inside these nets indicates that the current design of excluder devices used in this fishery could be improved to further reduce bycatch rates.

Incidental catch, or bycatch, of harbour porpoise (Phocoena phocoena) in fishing operations is an international conservation issue. The main objective of this thesis was to develop methods for determining the impact of bycatch on the state and dynamics of porpoise populations and for calculating bycatch limits that will achieve conservation objectives in the future. I applied these methods to the North Sea as a case study. First, I analysed sighting rates of harbour porpoise on seabird surveys in the North Sea during 1980-2003 to determine whether these data could provide informative time-series of relative abundance. Some general patterns and trends in sighting rates were consistent with previous studies. However, the standardised indices of abundance were relatively imprecise and thus have limited value for a monitoring framework that relies on statistical detection of trends. Second, I used a population model to integrate available data on harbour porpoise in the North Sea and to assess the dynamics of the population during 1987-2005. There was a high probability that bycatch resulted in a decrease in abundance. The estimated life history parameters suggested a limited scope for population growth even in the absence of bycatch. The model and data were not informative about maximum population growth rate or carrying capacity. The model suggested that dispersal was the most plausible explanation for observed changes in distribution within the North Sea. Third, I considered management procedures for calculating bycatch limits. I performed simulations to compare the behaviour of the procedures, to tune the procedures to specific conservation objectives and to test the robustness of the procedures to a range of uncertainties regarding population dynamics and structure, the environment, observation and implementation. Preliminary annual bycatch limits for harbour porpoise in the North Sea ranged from 187-1685 depending on the procedure, tuning and management areas used.

Two species of sea turtle, loggerheads (Caretta caretta) and leatherbacks (Dermochelys coriacea) are caught frequently as bycatch in longline fisheries. These fisheries use hooks baited with fish or squid. Yet, leatherbacks feed on gelatinous prey while loggerheads are carnivores. I investigated the responses of these two species to bait odors in controlled laboratory experiments to better understand their feeding behavior and why they interact with longlines. Both species initiated feeding behavior in the presence of squid bait odors and just C. caretta showed feeding behavior with sardine odors; neither responded to mackerel odors. The turtles are hooked differently on longlines. Loggerheads are usually hooked in the mouth while leatherbacks are usually hooked in the shoulder or flippers. Comparisons of prey attack behavior and accuracy in apprehending a stimulus in the presence of waterborne food odors identified speciesspecific differences that may predispose the turtles to particular kinds of hooking. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2014. / FAU Electronic Theses and Dissertations Collection