Variability in the diet of the Cape gannet at Bird Island,Algoa Bay,South Africa

Results are presented following a long-term study of the diet of the Cape gannet Morus capensis at Bird Island, Algoa Bay, examining 4 178 regurgitations representing 36 351 prey items collected during 12 breeding cycles from April 1979 to March 1991. In all, 32 teleost and two cephalopod species were recorded. The pelagic shoaling pilchard Sardinops ocellatus, anchovy Engraulis capensis and saury Scomberesox saurus scombroides were identified as the main prey species. The diet composition showed marked interannual fluctuations during the sampling period, which appear to be related to changes in the abundance of the main prey species. The relative abundance of pilchard, anchovy and saury was also found to change intra-annually, the first two species occurring more frequently in the diet during the breeding season and saury dominating the diet in the non-breeding season. These changes are thought to originate from temporal variation in the oceanographic regime within the foraging range of the gannets.     

Seasonal variation in the relationship between otolith radius and fish length in the pilchard off South West Africa

The relationship between otolith radius and fish length in the pilchard Sardinops ocellata off South West Africa is shown to vary seasonally. This phenomenon may explain the sinuous nature of the relationships between body part and fish length in other studies. The seasonal variation is too large to be accounted for solely by ingrowth of the hyaline ring by opaque material. The otolith radius: fish length ratio shows a weak negative correlation with fish growth rate (r = ?0,7547, p = 0,1–0,05).     

Back-calculation and time of hyaline ring formation in the otoliths of the pilchard off South West Africa

The frequency of hyaline and opaque edges, back-calculation, and examination of otoliths bearing a natural tag were used to determine the time of ring formation for the pilchard off South West Africa. The extent of the error in the back-calculation method caused by seasonal variation in the relationship between otolith radius and fish length is assessed. The times at which hyaline rings were laid down between 1970 and 1981 are documented. Formation of hyaline rings is related to sea surface temperature (p < 0,001), prominent rings forming when the temperature is colder than the annual average. Generally, a prominent hyaline ring is formed in September and secondary rings in May and December.     

Socio-economic characteristics of gillnet and beach-seine fishers in the Western Cape,South Africa

Data collected by questionnaire and telephone surveys conducted during 1998 and 1999 are used to describe the socio-economic characteristics of inshore netfishers in the Western Cape. Approximately two-thirds of netfishers work or have worked in other fishing sectors and a further 6–50%, depending on the area surveyed, are retired. Very few (0–11%) permit-holders in most areas classified their occupations as netfishers and the majority claimed to make <5% of their income from netfishing. Estimated costs and returns to net permit-holders suggest that, in most areas, commercial netfishing at current levels of catch and effort is not economically sustainable in the long-term. Only Saldanha-Langebaan gillnetters and beach-seine permit-holders, on average, manage to cover their opportunity costs and make an economic profit. The lack of profits in other areas is compelling evidence that the net fisheries are at or beyond the open access equilibrium point, suggesting that effort reduction in the order of 60% is necessary if maximum economic yield is to be obtained from the fishery. The netfisheries provide part-time employment for approximately 2 000 crew in the Western Cape. Additional economic benefits and employment directly related to the fishery in the form of equipment and fuel purchases made by fishers, maintenance of fishing gear and the sale of fish are estimated to contribute at least R15 million to the regional economy annually. Between 42 and 76% of respondents felt that their catches had declined since they had started netfishing and most felt that no new permits should be issued. Knowledge of catch restrictions among respondents was low (53–73%), indicative of a lack of communication between management and fishers, poorly defined permit conditions and a lack of enforcement. Many fishers interviewed feel it is unfair that they are restricted to catching only low-value target species and do not adhere to the catch restrictions, even if they do know them. The importance of the netfishery for participants varies greatly between and within areas. In order to reduce effort equitably, current and potential new permit-holders should be assessed on an individual merit basis.     

Growth of Cape horse mackerel Trachurus trachurus capensis off South Africa

Recent studies have provided evidence that the Cape horse mackerel Trachurus trachwus capensis in South African waters is a unit stock. Therefore, an age and growth study of the entire stock throughout its distributional range was undertaken in 1988/89. Growth is described by the equation

L_t = 487,989 (1 – e^{?0.556(t – 0,229)}) mm.     

Evidence of recovery of the linefishery in the Berg River Estuary,Western Cape,South Africa,subsequent to closure of commercial gillnetting

A total of 248 roving creel surveys along the length of the Berg River Estuary, in the Western Cape, South Africa, recorded 626 shore-angler and 88 boat-angler outings over the period December 2002–November 2005. Catch-and-effort information was obtained from catch inspections with 360 handline and 246 rod-anglers. Average total annual linefishing effort was estimated at 449 ± 29 (mean ± SE) boat-angler days, 1 299 ± 118 recreational shore-angler days and 1 394 ± 57 subsistence (handline) shore-angler days. The estimated total shore-based linefish catch (excluding boat-based catches) from the estuary for the years 2004 and 2005 was 37 231 ± 1 326 fish and 26 938 ± 706 fish (approximately 8 t and 7 t) respectively. Compared with other estuaries along the South African east coast where angler catches have been surveyed, species diversity in catches from the cool-temperate Berg River Estuary was low, with only 15 species caught, of which three, elf Pomatomus saltatrix (56%), harder Liza richardsonii (31%) and carp Cyprinius carpio (11%), dominated the catch. Average linefish catch per unit effort of most species increased significantly in the two years subsequent to the closure of the long-existing commercial gillnet fishery in March 2003. Length frequency distributions revealed significant increases in the average size and an increased contribution of larger size class elf and harder to the linefish catch over the monitoring period, suggesting a degree of recovery of the estuarine icthyofauna after more than a century of intensive gillnet fishing.     

Influence of the annual sardine run on catches of large sharks in the protective gillnets off KwaZulu-Natal,South Africa,and the occurrence of sardine in shark diet

Shark catches in the protective nets set off the beaches of KwaZulu-Natal (KZN), South Africa, are strongly influenced by the sardine run, the winter influx of shoals of Sardinops sagax from the south-west. The effect of the sardine run, which is highly variable from year to year, is greatest in June and July at beaches south of Durban. Total annual shark catch and effort are presented for the period 1952–2005, and total monthly shark catch on the KZN south coast for the period May–August, 1965–2005. Measures to reduce catches of sharks associated with the sardine run have been introduced and have been increasingly successful. Reliable species-specific catch data are available for the period 1978–2005 only. For this period, the spatio-temporal distribution of each of 14 species of shark and the frequency of occurrence of sardine in their diets is documented. Occurrence varies according to species, as does the apparent influence of the sardine run on shark distribution. During June and July on the KZN south coast, sardine were found in the diet of all but two species and frequency of occurrence was 40% or greater in eight species. The presence of copper or bronze whaler sharks Carcharhinus brachyurus in KZN waters appears to be strongly associated with the sardine run, as does that of certain life-history stages of dusky sharks C. obscurus. Spinner sharks C. brevipinna and smooth hammerhead sharks Sphyrna lewini are caught in greater numbers in summer than in winter, but appear to shift their spatial distribution seasonally to feed on sardine.     

Distribution and abundance of eggs,larvae and early juveniles of saury Scomberesox saurus scombroides (Richardson) off the south-western Cape,South Africa, 1977/78

The distribution patterns of saury eggs, larvae and early juveniles have been investigated over the continental shelf off the South-Western Cape from August 1977 to August 1978. Young saury tend to occupy water of 18°C or warmer in all seasons and are found offshore of the Benguela oceanic front when inshore upwelling is evident. Their distribution in the region appears to be linked to seasonal hydrography. Spawning in the study area occurs mainly between Cape Columbine and Cape Point, and to a lesser extent east of Cape Point. The ratio of large larvae and early juveniles to small larvae is greater off the West Coast than off the South Coast. Most spawning appears to take place in summer, with a peak abundance of larvae and early juveniles towards the end of this season. Instantaneous natural mortality was calculated: 5 – 30 mm, Z = 0,14·day^?1; 30 – 100 mm, Z = 0,02·day^?1. There is therefore a marked but unexplained decline in mortality shortly after the transition from the larval to the juvenile phase.     

Bycatch in the gillnet and beach-seine fisheries in the Western Cape,South Africa,with implications for management

Interview questionnaires and access point surveys were conducted in order to describe and quantify the catch composition of the inshore net-fisheries in the Western Cape, South Africa. A total of 138 562 fish, representing 29 species from 20 families, was recorded in 141 monitored commercial gillnet fishing operations between February 1998 and October 1999. Numerically, the legal target species, harders Liza richardsonii, dominated the catches, contributing 94.87% of the total gillnet catch. Elf, Pomatomus saltatrix, horse mackerel Trachurustrachurus capensis, gurnard Chelidonichthys capensis and barbel Galeichthys feliceps were the most common bycatch species, and contributed 4.2% to the total catch numerically and occurred in 12–47% of the marine 44–64-mm gillnet catches that were monitored. Five species most frequently targeted by shore-anglers on the West Coast: galjoen Dichistius capensis, white stumpnose Rhabdosargus globiceps, hottentot Pachymetopon blochii, silver kob Argyrosomus inodorus and white steenbras Lithognathus lithognathus, also occurred in gillnet catches. Most of the bycatch consisting of immature, undersized fish that were often injured during entanglement and were not released alive. L. richardsonii also numerically dominated the beach-seine hauls that were monitored (>99%) with only four bycatch species being recorded in low numbers. Beach-seine questionnaire respondents, however, reported sporadic catches of at least 17 bycatch species, including occasional appreciable catches of the important linefish species L. lithognathus and A. inodorus.     

Aspects of the upwelling regime between Cape Point and Cape Agulhas,South Africa

A progressive increase in mean ovary mass (standardized by the cube offish length) over time was observed in specified length classes of pilchard Sardinops ocellatus sampled from purse-seine catches off the west coast of South Africa. This, in earlier investigations, was considered to have been a density-dependent response following the collapse of the stock in the mid 1960s, and it appeared to result mainly from the increasing frequency of occurrence of pilchard entering the batch-spawning cycle, particularly among fish of 16–20 cm L_c . The possibility that the observed trend was an artifact of changing distribution of fishing or shifting seasonality of spawning relative to the three-month period of sample collection was rejected following analysis of variance on spatially disaggregated data and inspection of seasonal patterns in ovary mass. Other causes of the increase in mean ovary mass could have been density-dependence, declining age structure, changes in the environment and selection for early maturity under prolonged high rates of mortality. The data are critically examined to evaluate the likelihood of each of these explanations.     

Population dynamics and potential yield of three species of giant winkles in the Western Cape,South Africa

Of the southern African intertidal and shallow subtidal trochid and turbinid gastropods, Turbo sarmaticus, T. cidaris and Oxystele sinensis, are the most abundant large species, and therefore obvious targets for a winkle fishery. T. sarmaticus is harvested by recreational snorkel divers, and an application has been made for a permit to experimentally harvest the other two species commercially. This study involves four major aspects, namely morphometrics and flesh yield, shore-based abundance estimates, subtidal abundance estimates, and an estimate of the potential yield for each of the three species off the South-Western Cape, South Africa. Abundance and biomass of the winkles increased from west to east, highest densities being recorded in the low intertidal regions to 2 m depth. Low biomasses of the three species were recorded west of Cape Hangklip and are unlikely to support a fishery there. East of Cape Hangklip, invasion of rock lobsters Jasus lalandii has had a considerable impact on T. cidaris and O. sinensis populations. Consequently, a boat-based fishery with an estimated annual total allowable catch of 75.5 tons for T. cidaris, with a bycatch of 9.9 tons for O. sinensis, can only be considered in the easternmost fisheries-management area between Kleinbaai and Quoin Point. Because previous studies on the harvesting potential of T. sarmaticus off the Western Cape concluded that it was not commercially sustainable, a total allowable catch for that species was not considered. Management options deserving considerations before the commencement of a giant winkle fishery are discussed.     

Distribution of and predation on saury Scomberesox saurus scombroides in continental shelf waters off the Cape Province,South Africa

Although the size distribution of larvae and early juveniles of the saury Scomberesox saurus scombroides in continental shelf waters off the Cape Province, South Africa, is consistent with a south-north passive dispersal by known currents, the size of late juveniles and adults increases from north-west to south-east. Occurrence of these stages is highly seasonal: they are found in summer off the Western Cape and mainly from late summer to winter in Algoa Bay in the Eastern Cape, possibly reflecting longitudinal migration. Large numbers of late juveniles and adults are often found downstream of upwelling plumes off the Western Cape when warmer waters lie close inshore. They are sometimes taken in purse-seines in association with adult round herring Etrumeus whiteheadi and pilchard Sardinops ocellatus. Important predators of late juvenile and adult saury are fast-moving, surface-feeding species: yellowfin tuna Thunnus albacares, Cape gannet Morus capensis and Cory's shearwater Calonectris diomedea. Cape gannets consume an estimated 2 277–6 044 tons of saury annually in South African waters. Predator diets provide important time-series of occurrence, abundance and length-frequency distributions of late juvenile and adult saury.     

The feeding ecology of skates (Batoidea: Rajidae) off the Cape south coast,South Africa

Six species of skates were trawled commonly on the continental shelf off the Southern Cape during routine surveys of demersal fish between 1986 and 1990. Raja miraletus is a small skate found principally in shallow (<50 m) water that feeds mainly on small crustaceans, such as mysids and macrurans. The large Raja alba occurs over most of the continental shelf, is a piscivore throughout its life and preys heavily on flatfish, especially Cynoglossus zanzibarensis. R. cf. clavata is ubiquitous on the shelf but most abundant between 50 and 150 m deep. Feeding predominantly on crustaceans, this skate switches from carids, penaeids and mysids when small to brachyurans and small benthic fish when larger. R. pullopunctata is found over a similar depth range but attains a much larger size. It eats small mysids and macrurans when small, but brachyurans and fish become more important in the diet with increasing size of the predator. R. wallacei is most commonly taken at depths of 80–>200 m, feeds initially on small crustaceans, but switches to fish, including eels, as it grows. The small Cruriraja parcomaculata is found at the edge of the continental shelf and over the upper slope; it preys on small crustaceans, including mysids, isopods and macrurans. The distribution pattern of the skates and their prey suggest overlap in prey taken, the overlap being greatest in those species which feed on crustaceans. The large biomass of skates in the area suggests that they are an important component of the demersal faunal assemblage of the continental shelf there, and also that they may have a significant impact on the benthic fauna.     

Impact of predation by Cape fur seals Arctocephalus pusillus pusillus on Cape gannets Morus capensis at Malgas Island,Western Cape,South Africa

Cape fur seals Arctocephalus pusillus pusillus were estimated to kill some 6 000 Cape gannet Morus capensis fledglings around Malgas Island in the 2000/01 breeding season, 11 000 in 2003/04 and 10 000 in 2005/06. This amounted to about 29%, 83% and 57% of the overall production of fledglings at the island in these breeding seasons respectively. Preliminary modelling suggests this predation is not sustainable. There was a 25% reduction in the size of the colony, the second largest of only six extant Cape gannet colonies, between 2001/02 and 2005/06. There has been a large increase in predation by Cape fur seals on seabirds around southern African islands since the mid-1980s, coincidental with both an increase in the seal population, altered management of the islands and an altered distribution of prey for gannets and seals. At Malgas Island, most gannet fledglings were killed between 10:00 and 18:00, the period when most are in the water around the island, from mid-January to mid-March, the main fledging period. The Cape gannet is classified as Vulnerable.     

Distribution,population size and conservation of the Cape gannet Morus capensis

The Cape gannet is endemic to the southern African coast where it currently breeds at six islands: Mercury, Ichaboe and Possession off South West Africa and Bird (Lambert's Bay), Malgas and Bird (Algoa Bay) off South Africa. Previously, breeding also occurred at Hollams Bird, Halifax and Seal (False Bay) Islands. Equivocal records for Marcus, Dassen and Dyer Islands are not accepted. Off South West Africa, gannets were breeding at Hollams Bird, Mercury and Ichaboe Islands at least as early as 1828, but they only occupied Halifax and Possession Islands sometime between that date and c. 1885, possibly as a result of displacement of gannets from Ichaboe Island during exploitation of accumulated guano deposits in the early 1840s. Gannets bred at Hollams Bird Island until at least 1938, but had ceased breeding at Halifax Island by 1928. Off South Africa the earliest records of breeding are 1648, 1687, 1755 and 1912 for Malgas, Seal (False Bay), Bird (Port Elizabeth) and Bird (Lambert's Bay) Islands respectively. Gannets have not been reported at Seal Island since the late 17th century. On the west coast of Africa the Cape gannet is a regular nonbreeding winter visitor as far north as 4°20′N 6°00′E, but west of 6°E it is rare. On the east coast of Africa it is a common winter visitor as far north as Delagoa Bay, but farther north it is rare. Within its normal range the Cape gannet seldom occurs farther off shore than 100 km; it hardly ever moves inland. Aerial censuses of Cape gannets at breeding islands in 1967, 1969, 1978, 1980 and 1981 are compared with an aerial census conducted in 1956 and other published estimates of abundance. Between 1956 and 1980 the estimated number of breeding pairs at all colonies decreased from c. 150 000 to c. 80 000 and numbers decreased at all three extant gannetries off South West Africa. These decreases are attributed to a greatly diminished food resource following the collapse of the South West African pilchard stock after the late 1960s. The number of gannets decreased at Bird (Lambert's Bay) and Malgas Islands between 1956 and the late 1960s but subsequently increased, trends that are related to performances of the Western Cape pilchard and anchovy resources respectively. At Bird Island (Algoa Bay) gannets were up to 3,5 times more abundant in the late 1970s than in 1956. Other marine resources located east of Cape Point have shown similar large increases in recent years. Rates of increase of gannets at islands off South Africa during the 1970s would have required an unrealistically high survival for the first year had other population parameters remained constant. It is possible that birds emigrated from the South West African Islands. Few gannets have been reported oiled, and conservation of the species seems to be mainly affected by greatly reduced prey availability and injudicious guano harvesting. Human exploitation of juvenile gannets off the West African coast is difficult to assess.     

Information on currents in the Cape Peninsula area,South Africa

Statistical information and some interpretative detail, based on currents measured at 28 mooring sites between Cape Point and Yzerfontein since 1981, are presented. The most striking feature is the continual presence of wavelike reversals of current close to the coast and in deeper water offshore. These currents have a periodicity of 3–4 days and exhibit a net poleward transport. The energy at these frequencies is greater than that at tidal frequencies and dominates the flow. The observation of equatorward flow over the 230-m contour confirms the presence of a jet there, identified previously by drogue tracks and current profiling studies.     

A note on the diet of Cape cormorant Phalacrocorax capensis at Mercury Island,South West Africa,in November 1978

During November 1978, Cape cormorants Phalacrocorax capensis at Mercury Island, off South West Africa (Namibia), consumed practically exclusively pelagic gobies Sufflogobius bibarbalus. The mean number of fish per regurgitation was 52 with a mean caudal length of 48 mm and an estimated mean mass of 1.7 g. The estimated mean original mass of stomach contents was 75 g. These Cape cormorants took significantly smaller pelagic gobies than did bank cormorants P. neglectus sampled concurrently at the same locality.     

Identification of and distinction between the scales of South African pilchard Sardinops ocellatus and Cape anchovy Engraulis capensis

Fossil fish scales hold potential for elucidating past fluctuations in fish populations. Only certain pilchard Sardinops ocellatus and anchovy Engraulis capensis scales can be distinguished easily from each other. Misidentification of the less-typical scales of the two species can introduce a bias in scale-based population studies and, in order to overcome this bias, scales removed from pilchard and anchovy reared in captivity were examined. A complete set of reference photographs of the scales of each species was compiled and study of these showed that both pilchard and anchovy have five distinctive scale types, each of which is found on different parts of the fish. It was also shown that the "typical" scales of each species, i.e. those referred to in the literature, constitute less than 50 per cent of the scales found on the fish.     

Long-term trends in catch and effort of commercial linefish off South Africa's Cape Province: snapshots of the 20th century

The Cape commercial linefishery, established during the first half of the 19_th century, currently consists of about 2 500 vessels (±20 000 crew), which land some 15 500 tons of fish each year. In spite of a long history, a lack of a long-term catch and effort data series has severely hindered the management of the fishery. This paper providcs commercial catch and effort data for three periods during the 20_th century: 1897–1906, 1927–1931 and 1986–1998. Trends in catch per unit effort (cpue) were verified with additional data from the inshore trawl-fishery, from fishery-independent surveys and, where possible, stock assessment. According to a Linefish Management Protocol developed for the Linefishery, any stock demonstrating a historical reduction in cpue or catch contribution of more than 75% is to be regarded as overexploited. Evaluations based on present datasets support this arbitrarily determined reference point for cpue, but reveal that catch composition is a poor indicator of stock status and should be used with caution. In spite of technological advances such as the advent of combustion engines, nylon lines. echo-sounders, electronic navigational aids, onboard freezer facilities and larger vessels, declines in catch rate indicative of severe overexploitation (i.e. 75–99%) were observed for many important linefish species during the 20_th century. Most of these were higher-trophic-Level species from the warm/temperate East Coast, several of which are also endemic. Life-history characteristics of the vulnerable species (i.e those demonstrating declines in cpue of >75%) include predictable location in time and space (either coastal migrant or resident), longevity (>15 years) and late maturity (relative to maximum age). Apart frorreduced productivity associated with stock depletion, other setbacks, such as ecosystem alteration, loss of genetic diversity and shor-term commercial extinction, are also anticipated. In order to rebuild depleted linefish stocks it is deemed essential to create additional marine reserves, dramatically improve enforcement, develop extensive public awareness programmes and substantially reduce commercial effort.