Application of the sequential t-test algorithm for analysing regime shifts to the southern Benguela ecosystem

Long-term ecosystem changes, such as regime shifts, have occurred in several marine ecosystems world-wide. Multivariate statistical methods have been used to detect such changes. A new method known as the sequential t-test algorithm for analysing regime shifts (STARS) is applied to a set of biological state variables as well as environmental and anthropogenic forcing variables in the southern Benguela. The method is able to correct for auto-correlation within time-series by a process known as prewhitening. All variables were tested with and without prewhitening. Shifts that were detected with both methods were termed robust. The STARS method detected shifts in relatively short time-series and identified when these shifts occurred without a priori hypotheses. Shifts were generally well detected at the end of time-series, but further development of the method is needed to enhance its performance for auto-correlated time-series. Since 1950, two major long-term ecosystem changes were identified for the southern Benguela. The first change occurred during the 1960s, caused predominantly by heavy fishing pressure but with some environmental forcing. The second change occurred in the early 2000s, caused mainly by environmental forcing. To strengthen these findings, further analyses should be carried out using different methods.     

Occurrence and distribution of Cape hakes Merluccius capensis Cast. and M. paradoxus Franca in the Cape of Good Hope area

A brief history of research into, and the fishery for, Cape hake is given. From twelve monthly surveys in 1972/73, abundance of the two species (Merluccius capensis and M. paradoxus) at depth was found to vary, larger individuals of both species being found deeper than the smaller ones. Off the Cape of Good Hope, M. capensis was never found deeper than 440 m, and there was significant species overlap between 150 and 440 m. M. paradoxus was numerically dominant except at the shallowest station sampled (150 m). Little overlap of mature specimens of each species was recorded. The research data were applied to commercial trawl statistics, revealing that M. capensis accounted for only 12,9 per cent by mass and 4,8 per cent by number of the commercially caught hake in the years 1955–1974. Such low percentages justify considering the Cape hake fishery as a single-species entity for the purposes of stock assessment and management.     

Diet of sardine Sardinops sagax in the southern Benguela upwelling ecosystem

The diet of sardine Sardinops sagax in the southern Benguela was investigated by microscopic examination of stomach contents. The relative dietary importance of prey size and prey type was assessed by calculating the carbon content of prey items. Sardine is an omnivorous clupeoid, ingesting both phytoplankton and zooplankton, with the relative importance of these two food types varying both spatially and temporally. Stomach contents were numerically dominated by small prey items, principally dinoflagellates, followed by crustacean eggs, cyclopoid copepods, calanoid copepods and diatoms. Virtually all prey items ingested by sardine were <1.2 mm maximum dimension, the particle size below which sardine only filter-feed. Despite the numerical dominance by phytoplankton, zooplankton contributed the major portion to sardine dietary carbon, small calanoid and cyclopoid copepods, anchovy eggs and crustacean eggs being the primary prey types. These results indicate that, like anchovy Engraulis capensis, sardine in the southern Benguela are primarily zoophagous, and contrast with earlier dietary studies on sardine in the region. However, the two species appear to partition their prey on the basis of size; sardine consume small zooplankton, whereas anchovy consume large zooplankton. This difference has been observed in other upwelling ecosystems where the two genera co-exist and is likely to contribute to the regime shifts observed between sardine and anchovy.     

Characterising and comparing the spawning habitats of anchovy Engraulis encrasicolus and sardine Sardinops sagax in the southern Benguela upwelling ecosystem

The spawning habitats of anchovy Engraulis encrasicolus and sardine Sardinops sagax in the southern Benguela upwelling ecosystem were characterised by comparing their egg abundances with environmental variables measured concomitantly during two different survey programmes: the South African Sardine and Anchovy Recruitment Programme (SARP), which comprised monthly surveys conducted during the austral summers of 1993/94 and 1994/95; and annual pelagic spawner biomass surveys conducted in early summer (November/December) from 1984 to 1999. Eggs were collected using a CalVET net. Physical variables measured included sea surface temperature (SST), surface salinity, water depth, mixed-layer depth, and current and wind speeds; biological variables measured included phytoplankton biomass, and zooplankton biomass and production. Spawning habitat was identified by construction of quotient curves derived from egg abundance data and individual environmental variables, and relationships between these variables were determined using multivariate co-inertia analysis. SARP data showed that anchovy spawning was associated with cool water and moderate wind and current speeds, whereas sardine spawning was related to warmer water and more turbulent and unstable conditions (i.e. high wind speeds and strong currents) than for anchovy. SARP data also showed significant differences in selection of spawning habitat of the two species for all environmental variables. The relationship between anchovy egg abundance and salinity was strongly positive, but strongly negative with water depth, phytoplankton biomass and zooplankton production. Sardine egg abundance was strongly positively related to current speed. The spawner biomass survey data demonstrated that the spawning habitat of anchovy was characterised by warm water and high salinity, whereas sardine spawning was associated with cool water and low salinity. The survey data showed significant differences in spawning habitat selection by anchovy and sardine for SST, salinity and zooplankton biomass, but not for the other environmental variables. There was a positive relationship between anchovy egg abundance and SST, salinity and mixed-layer depth, and a negative relationship with water depth, phytoplankton biomass and zooplankton production. For sardine there was a strong positive relationship between egg abundance and current speed and wind speed. Differences in the results between the two survey programmes could be attributable to differences in their spatio-temporal coverage. Spawning habitats of anchovy and sardine appear to be substantially different, with anchovy being more specific than sardine in their preference of various environmental conditions.     

Density-dependent changes in reproductive parameters and condition of southern Benguela sardine Sardinops sagax

The sardine Sardinops sagax population in the southern Benguela has undergone substantial fluctuations in size over the past 50 years, collapsing from an apparently large population in the 1950s to low levels in the mid-1960s, remaining low for the next two decades, and recovering from the late 1980s to a population size that is now similar to or larger than that which occurred during the 1950s. Marked changes in condition and reproductive parameters of sardine have also occurred during this period; condition and standardised gonad mass are higher and length-at-maturity is lower at low population size compared with high population size. The correspondence between the temporal patterns in condition, reproductive parameters and population size are strongly suggestive of density-dependence, and indicate a compensatory response arising from reduced intra-specific competition. This is likely to have resulted from greater per capita food intake, improved body condition and hence faster growth, thus enabling fish to achieve maturation at a presumably younger age and smaller size. Biological parameters did not vary in or out of phase with time-series of sea surface temperature in the southern Benguela, weakening the hypothesis of environmentally mediated changes in these parameters and hence providing support for the hypothesis of a direct density-dependent response by sardine.     

Vertical migration,catchability and acoustic assessment of semi-pelagic Cape horse mackerel Trachurus trachurus capensis in the southern Benguela

The vertical distribution and migration of the Cape horse mackerel Trachurus trachurus capensis were investigated on the basis of midwater and bottom trawl collections taken over diel cycles at two sites on South Africa's south coast. Attempts were also made to investigate the relationship between bottom trawl catch rates and acoustic abundance estimates. During the first cycle, horse mackerel were dispersed in midwater during the night, and towards sunrise they coalesced into schools and migrated to the seabed where they remained throughout the day. During that cycle, horse mackerel did not appear to feed, despite the prevalence of potential food, mainly Calanus copepods and euphausiid larvae, in the water column. In contrast, during the second cycle, horse mackerel remained near the seabed during the 24-h experiment, but they exhibited a marked feeding periodicity, with most feeding taking place in the late afternoon. They appeared to select for larger copepods in the water column. Several reasons are advanced to explain the difference in the vertical migratory behaviour of horse mackerel between the two sites, on the basis of the vertical distribution of their potential prey. Diel differences in acoustic estimates of horse mackerel abundance were significantly reduced after applying an acoustic deadzone correction factor. Trawl catches, however, could not be reconciled with acoustic estimates during the day, when horse mackerel were too close to the bottom for effective echo-integration.     

Mesoscale coupling between planktonic cnidarian distribution and water masses during a temporal transition between active upwelling and abatement in the northern Benguela system

The composition and spatio-temporal distribution of planktonic cnidarians in the northern Benguela system are described from plankton samples collected on four cruises between December 1981 and March 1982. The sampling period was characterized by initially active coastal upwelling, followed by abatement of upwelling. In all, 20 species of medusae and 24 species of siphonophores were identified. The most abundant medusae were Chrysaora hysoscella, Aequorea aequorea and Liriope tetraphylla, and the most abundant siphonophores were Muggiaea atlantica, Abylopsis tetragona and Bassia bassensis. Three assemblages of planktonic cnidarians were differentiated. The first was associated with coastal upwelling and consisted of the medusae Proboscidactyla menoni, Aglauropsis edwardsii and juvenile C. hysoscella, along with the siphonophore M. atlantica. The second assemblage was associated with oceanic water and the component species were indicative of the offshore influence of the Benguela system; L. tetraphylla and most of the siphonophores belonged to this assemblage. The third assemblage consisted of species typical of shelf fauna, namely A. aequorea, Mitrocomella grandis and adult C. hysoscella, all also typical of the northernmost waters of the Benguela system. Overall, there was coupling between mesoscale hydrographic features and the pattern of distribution of cnidarian populations, with a distinct cross-shelf gradient that was more pronounced during active upwelling.     

Abrupt environmental shift associated with changes in the distribution of Cape anchovy Engraulis encrasicolus spawners in the southern Benguela

Cape anchovy Engraulis encrasicolus spawners in the southern Benguela showed an eastward shift in their distribution on the Agulhas Bank that occurred abruptly in 1996 and has since persisted. We assessed whether this shift was environmentally mediated by examining sea surface temperature data from different regions of the Agulhas Bank, which showed that in 1996 the inner shelf of the Agulhas Bank to the east of Cape Agulhas abruptly became 0.5°C colder than in previous years and has since remained that way. In addition, signals, coherent with the 1996 shift recorded in sea surface temperatures, were also found in atmospheric surface pressure and zonal wind data for that region; interannual coastal SST variability is also shown to be correlated with zonal wind-stress forcing. As a result, increased wind-induced coastal upwelling east of Cape Agulhas is proposed as the main driver of the observed cooling in the coastal region. The synchrony between the environmental and biological signals suggests that the eastward shift in anchovy spawner distribution was environmentally mediated and arose from a change in environmental forcing that altered the relative favourability for spawning between regions to the west and east of Cape Agulhas. The results highlight how a relatively minor change in environmental conditions can lead to a drastic spatial reorganisation of the life history of one species in an ecosystem.     

Impact and implications of large-scale environmental anomalies on the spatial distribution of spawning of the Namibian pilchard and anchovy populations

Trends in the spatial pattern of spawning of pilchard Sardinops ocellatus and anchovy Engraulis capensis during the period 1971–1985 were examined in relation to large-scale environmental anomalies outside the spawning area. Shifts in the locality of peak spawning of anchovy often coincided with that of pilchard, although sometimes anchovy were slower to respond to an environmental change. With negative sea surface temperature (SST) anomalies spawning shifted northwards, whereas with positive anomalies spawning took place both in the north and farther south. With a transition from negative SST anomalies early in the spawning season to zero or positive anomalies later in the spawning season, the impact on anchovy and pilchard spawning differed. Anchovy spawning was either reduced or absent during the negative phase of transitional seasons, thus avoiding unfavourable conditions. Pilchard, on the other hand, were spawning mainly in the north during the negative phase of the transitional seasons, but also in the south during the zero to positive phase. The shifts in spawning locality in response to environmental anomalies outside the spawning area imply that explanations for such shifts need not be sought in changes in biomass or age structure or in the fact that stocks may be genetically different, all theories previously advanced.     

An evaluation of artificial shelters in the artisanal spiny lobster fishery in Gazi Bay,Kenya

Two designs of artificial shelters were evaluated for use in the artisanal spiny-lobster fishery in Gazi Bay, Kenya. Both types of shelter were effective in aggregating the spiny lobster Panulirus ornatus in nearshore sea-grass beds. Lobsters aggregated at the shelters were caught by free-diving fishermen, using spearguns and hand nets. Mean lobster catches taken from the shelters ranged from 0,38 to 0,83 kg·trip^?1, around 50% of those taken farther offshore in natural reefs. No significant difference was found in the size or sex composition between shelter and reef-caught lobsters.     

Remotely sensed variability of temperature and chlorophyll in the southern Benguela: upwelling frequency and phytoplankton response

High-resolution (1km) satellite data from the NOAA AVHRR (Advanced Very High Resolution Radiometer) and OrbView-2 SeaWiFS (Sea-viewing Wide Field-of-view Sensor) are used to investigate the upper layer dynamics of the southern Benguela ecosystem in more detailed space and time scales than previously undertaken. A consistent time-series of daily sea surface temperature (SST) and chlorophyll a concentration images is generated for the period July 1998–June 2003, and a quantitative analysis undertaken. The variability in SST, upwelling and phytoplankton biomass is explored for selected biogeographic regions, with particular focus on intra-seasonal time scales. The location and emergence of upwelling cells are clearly identified along the length of the southern Benguela, being distinct on the narrow inner and the mid-continental shelves. Most notable is the rapidly pulsating nature of the upwelling, with intense warm/cold events clearly distinguished. The phytoplankton response to this physical forcing is described. Chlorophyll concentration on the inner shelf largely mirrors the pattern of SST variability, similarly dominated by event-scale processes. Over the mid-shelf, higher chlorophyll is observed throughout all seasons, although low biomass occurs during winter. The variability of the offshore extent of SST and chlorophyll is identified at locations of differing shelf width. Cooler upwelled water is confined primarily to the narrow inner-shelf, with event-scale pulses extending considerable distances offshore. Agulhas Current influences are readily observed, even on the Cape Peninsula inner-shelf. Chlorophyll concentrations vary considerably between the locations of differing shelf width. SST, upwelling and phytoplankton indices are derived for selected locations to quantify the intra-seasonal variations. The SST indices show marked temperature changes associated with rapid pulsation on the event scale. No strong seasonal signal is evident. In contrast, the upwelling indices display a strong seasonal signal, with most intense upwelling occurring in spring/summer in the south. The phytoplankton response to the seasonal upwelling index differs between the selected locations. This study concludes that, although low-resolution SST and chlorophyll data may be useful for investigating general patterns over large scales, higher resolution data are necessary to identify finer scale spatial and temporal variability, especially in the inshore coastal zones.     

Southward extension of the reproductive range of Macroramphosus scolopax in the upwelling area off North-West Africa

Ichthyological literature gives the distribution of M. scolopax as along the whole coast of West Africa. However, analysis of this literature and the results of trawling surveys in the North-East Atlantic suggest only exceptional occurrences in the tropics. Plankton surveys off N.W. Africa, up to and including 1977, rarely yielded M. scolopax larvae off the coast of the southern Sahara and not at all off Mauritania, though a few specimens were collected after having been swept southwards in the open ocean by the Canary Current. Larvae did occur off northern Mauritania in February 1983. Details of distribution, abundance and length composition of early stages caught during the period January–April 1983 are given. It is concluded that the reproductive range of M. scolopax (as well as that of some other temperate species) extends south of that documented and that its spawning season there may be shorter than in the centre of this species' distribution. The findings are discussed in relation to interannual variation of upwelling intensity and the current system off N.W. Africa.     

Comparing trophic flows in the southern Benguela to those in other upwelling ecosystems

A balanced trophic flow model of the southern Benguela ecosystem is presented, averaging the period 1980–1989 and emphasizing upper trophic levels. The model is based largely on studies conducted within the framework of the Benguela Ecology Programme and updates the results of an expert workshop held in Cape Town in September 1989. Small pelagic fish other than anchovy Engraulis capensis and sardine Sardinops sagax, mainly round herring Etrumeus whiteheadi and mesopelagic fish, were important components of the food web in the southern Benguela. Severe balancing difficulties were encountered with respect to the semi-pelagic resources (hake Merluccius spp.) and demersal top predators (sharks), indicating the need for further research on the interaction of these groups with their ecosystem. The model is compared to other existing trophic flow models of ecosystems in major upwelling areas, i.e. the northern Humboldt Current (4–14°S), the California Current (28–42°N) and the southern Canary Current (l2–25°N), and to two independently constructed models of the northern Benguela ecosystem. These models are compared using network analysis routines of the ECOPATH software, focusing on the interactions between the five dominant fish species (anchovy, sardine, horse mackerel Trachurus trachurus capensis, chub mackerel Scomber japonicus and hake) that support important fisheries in all systems. The upwelling systems rank by size rather than species dominance. The ratio of catches and primary production differs between systems, partly because of differences in fishing regimes. Predation on the five dominant fish groups by other fish in the system was the most important cause of fish mortality in all models. Fishery catches are generally a larger cause of mortality for these groups than predation by mammals. The ecological cost of fishing appears to be comparatively low in the southern Benguela, because catches are low compared with the primary production, but also because the fishery is relatively low in the foodweb. However, in view of the very tight foodweb demonstrated in the model. it is likely that an increase in fishing pressure would cause severe trade-offs with respect to other components of the southern Benguela ecosystem.     

Influence of oceanographic fronts and low oxygen on the distribution of ichthyoplankton in the Benguela and southern Angola currents

The study focuses on ichthyoplankton populations in the southern Angola Current, the Angola-Benguela Front and the coastal upwelling area of the northern and central Benguela Current. The horizontal and vertical distributions of eggs and larvae of sardine Sardinops sagax, anchovy Engraulis encrasicolus and horse mackerel Trachurus trachurus capensis were examined in relation to distribution patterns of temperature, salinity and dissolved oxygen. Samples were collected during February–March 2002 using a Bongo net and a multiple opening/closing net in five depth layers (0–30m, 30–60m, 60–100m, 100–140m and 140–220m). Eggs and larvae were restricted mainly to the shelf areas in the Angola-Benguela Font and southern Angola Current waters. Their vertical distributions in the Angola-Benguela Front and northern Benguela upwelling region showed a sharp decline in abundance at about 60m depth. This decline appeared to be correlated to the distribution of dissolved oxygen in the water column It is hypothesised that low oxygen concentrations have a strong impact on the development and survival of the early life-cycle stages of fish, and that recruitment of sardine and other pelagic species relies more on the upward extension of the oxygen minimum layer than was previously thought.     

Influence of buoyancy and vertical distribution of sardine Sardinops sagax eggs and larvae on their transport in the northern Benguela ecosystem

In recent years, sardine Sardinops sagax spawning has been recorded inshore off central Namibia. Field observations on eggs and laboratory measurements show that spawning, demonstrated by the distribution of newly spawned eggs, takes place just below the upper mixed layer. The high positive buoyancy of the eggs causes them to ascend rapidly to the surface layer, where they are moved offshore by upwelling-induced offshore transport. However, increased wind-induced mixing also influences the vertical distribution of eggs, causing them to be partly mixed down below the layer moving offshore and into the layer moving inshore. This mechanism acts to retard the transport and offshore loss of eggs from the spawning areas. The vertical distribution of sardine larvae, with highest concentrations deeper than 20 m, indicates active movement out of the layer moving offshore, and this tendency seems to be more pronounced for older larvae. Hence, vertical migration of larvae is an additional factor mitigating their loss from nearshore. Taken together, these features seem to minimize the offshore loss of offspring, particularly in periods of low stock biomass when spawning close to the shore seems to be common.     

Changes to regulations and the gear used in the South African commercial fishery for Jasus lalandii

Alterations to the operational fishing gear used in the South African west coast rock lobster Jasus lalandii fishery and their resultant impacts are investigated. The most important developments have been: (i) a change during the 1960s from hand-hauled hoopnets to winch-hauled traps, with a concomitant modification of vessels; (ii) the introduction of deck-grid sorters in 1975; (iii) an increase during 1984 of minimum mesh aperture from 62 to 100 mm (stretched), with a concomitant decrease in the length of the trap codend; (iv) a decrease in the minimum legal size during the early 1990s; and (v) the introduction of bottom-grid traps in 1994. Most of these alterations have been driven by a trade off between the need for greater operational efficiency and concerns surrounding the issue of how best to deal with the problem of catching and releasing specimens smaller than the minimum legal size.     

Comparison of assemblages and some life-history traits of seabirds in the Humboldt and Benguela systems

There are 21 and 15 species of seabirds that breed in the Humboldt and Benguela upwelling systems respectively. Only two species of gull are common to both systems, one as an endemic subspecies to the Benguela system. Eleven species and two subspecies are endemic (or nearly so) to the Humboldt system; seven species and one subspecies to the Benguela system. Each system has an endemic penguin, sulid, cormorant and tern that feed mainly on anchovy Engraulis spp., sardine Sardinops sagax or both these fish. The Peruvian pelican Pelecanus thagus also feeds primarily on these prey items. A plentiful availability of food has resulted in many of these seabirds attaining high levels of abundance. For the four pairs of species that feed on anchovy and sardine, those in the Humboldt system all have a biology that enables them to increase more rapidly than their Benguela counter-parts. This reflects the higher frequency of environmental perturbations that depress seabird populations in the Humboldt system. In addition, both systems have a small endemic cormorant that feeds near the coast and a small endemic tern that breeds in the adjacent mainland desert and feeds at the sea surface. Several seabirds endemic to a system have no obvious ecological equivalent in the other system: the pelican, a diving-petrel, four storm-petrels and a gull in the Humboldt system; a cormorant and a gull in the Benguela system. Some species with tropical or subantarctic affinities breed at the boundaries of the systems. Others breed also in freshwater systems. The grey gull Larus modestus, which feeds in the Humboldt system, breeds in montane deserts.     

Ecosystem modelling in the southern Benguela: comparisons of Atlantis,Ecopath with Ecosim,and OSMOSE under fishing scenarios

Ecosystem-based management of marine fisheries requires the use of simulation modelling to investigate the system-level impact of candidate fisheries management strategies. However, testing of fundamental assumptions such as system structure or process formulations is rarely done. In this study, we compare the output of three different ecosystem models (Atlantis, Ecopath with Ecosim, and OSMOSE) applied to the same ecosystem (the southern Benguela), to explore which ecosystem effects of fishing are most sensitive to model uncertainty. We subjected the models to two contrasting fishing pressure scenarios, applying high fishing pressure to either small pelagic fish or to adult hake. We compared the resulting model behaviour at a system level, and also at the level of model groups. We analysed the outputs in terms of various commonly used ecosystem indicators, and found some similarities in the overall behaviour of the models, despite major differences in model formulation and assumptions. Direction of change in system-level indicators was consistent for all models under the hake pressure scenario, although discrepancies emerged under the small-pelagic-fish scenario. Studying biomass response of individual model groups was key to understanding more integrated system-level metrics. All three models are based on existing knowledge of the system, and the convergence of model results increases confidence in the robustness of the model outputs. Points of divergence in the model results suggest important areas of future study. The use of feeding guilds to provide indicators for fish species at an aggregated level was explored, and proved to be an interesting alternative to aggregation by trophic level.     

A Statistical Model for Stock Assessment of Southern Bluefin Tuna with Temporal Changes in Selectivity

Assessment of the status of southern bluefin tuna (SBT) by Australia and Japan has used a method (ADAPT) that imposes a number of structural restrictions, and is similar to methods used for a number of stocks world-wide. A flexible method for assessment of the SBT population is presented that is much less restrictive and has potentially wide applicability. The three key features are: (1) all fitting to data is within the context of maximum likelihood, (2) catch-at-age data are not assumed to be without error (as in existing methods), but rather to be random variables, while age-specific selectivity is allowed to change over time within the bounds of specific structure, and (3) autocorrelation in recruitment processes is considered within the likelihood framework of the model. While the results suggest the stock has been depleted considerably from its virgin biomass, and are generally consistent with previous assessments, they also indicate that it is not as much below the biomass that will produce maximum suitable yield as previously estimated and that the extent of stock rebuilding necessary may not be as large as has been argued. The available data are shown to provide little information on the steepness parameter of the stock-recruitment function, and hence on sustainable catch levels for the stock.