Browsing by Subject "Applied Ocean Science"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Open Access
    An investigation of some key physico-chemical water quality parameters of an Integrated Multi-Trophic Aquaculture (IMTA) system operating recirculation methodology in the Western Cape of South Africa
    (2020) De Prisco, Joseph Anthony; Bolton, John
    Over the last few decades, Integrated Multi-Trophic Aquaculture (IMTA) in South Africa has developed from early experimental designs to large scale, commercially operating farms. This was in response to uncertainty regarding food availability for stock (primarily kelp in the case of abalone farms) and a desire to recirculate water whilst reducing the environmental footprint of the abalone farms. The growing prevalence of IMTA as a commercially viable activity has brought about a need for an expansion of the knowledge pool regarding the physico-chemical processes at work in such systems. Of particular interest to researchers are mechanisms and dynamics of nutrient transfer between components of the system and how these could be manipulated to increase efficiency and reduce running cost of farms. This work was conducted to try and quantify some of the changes in some physical and chemical characteristics of the water stream on a large-scale IMTA farm cultivating seaweed of the genus Ulva (Ulva rigida) and the locally named perlemoen abalone (Haliotis midae) on the south west coast of South Africa (Viking Abalone Farm at Buffeljagsbaai, Western Cape, South Africa) (34.7550° S, 19.6154° E). Experiment one was a three-day experiment taking place in December of 2018, there was no particular reason for the choice of month, analyses of this nature are potentially useful on any given day of any given month as although the literature contains plenty of gaps, there is no single identifiable data gap sufficient to encourage the use of particular timeframes. The sampling regime involved single sample point testing of three modular clusters each operating a different rate of water recirculation (50%, 75% and 100%) with 50% recirculation being standard farm operation, 75% and 100% tested to gauge effect of increasing recirculation, 75% tested as a potential standard farm operation to reduce load on pumps and reduce volumes of water pumped in, 100% tested in case of emergency situation which requires farm to be isolated from the inbound water stream arriving from the immediate coastal water, ambient conditions were also tested for reference and comparison. Parameters tested were those which the farmers already tested periodically to gauge changes in water quality which may effect the abalone or seaweed, though slightly different methods were used for the testing of ammonia. On the farm the standard method is the Nesler photometric test (Lovibond photometer), whereas this research was conducted using a calibrated indophenol blue spectrophotometric technique (Modified Grasshoff, 1976). Results showed no statistically significant differences (Mood's Median Test, p>0.05) between the 50% and 75% recirculation cluster for temperature, pH, Total Ammonia Nitrogen (TAN) or Free Ammonia Nitrogen NH3 (FAN). At 100% recirculation, statistically significant differences (Mood's Median Test, p0.05) occurred for temperature. At 100% recirculation, TAN and FAN increased rapidly, though the commensurate rapid and considerable decrease in pH meant the FAN increase was not as high in magnitude as it would be at 4 a normal seawater pH of around 8.2. Abalone suffered no mortalities at 100% recirculation for three days and later reports from the farmers suggested no noticeable drop in growth rate that could be attributed to this test in the months following the experiment. From the regulatory perspective, the TAN levels breached WWF guideline maximum effluent concentrations for abalone aquaculture (600µM/l) only in the 100% recirculation cluster, and only then during three of the thirteen sampling runs. The TAN concentrations in 50% and 75% recirculation treatments were far below the WWF guideline maximum effluent concentration with maximum concentrations of 7.15 µM/l in 50% and 13.46 µM/l at 75%, the increase in maximum concentration was large but not egregious and resulted from a more pronounced build-up of ammonia as residence time of water in the cluster increases at 75% recirculation. Experiment two was an intensive 24-hour sampling run; the primary aim was to test the effectiveness of the seaweed biofilter in an Integrated Multi-Trophic Aquaculture (IMTA) farm culturing perlemoen abalone and a green macroalga. Parameters tested were temperature, pH, dissolved oxygen, salinity, TAN, nitrate and nitrite as these are relevant parameters for the farmer and the necessary equipment to test them was available. Samples were stored in a freezer for this experiment due to intensity of sampling regime, and spiked standards were prepared to check shifts in concentration of TAN, nitrate and nitrite that may have resulted from the freezing and thawing processes. Spike recoveries were good in the case of TAN (87%-98%) and nitrite (92%-96%), but random and widely dispersed in the case of nitrate. As such, nitrate and nitrite were removed from the analysis as nitrite values only really held value if taken in conjunction with nitrate values. Minimal and non-useful variation in salinity observations meant that salinity was also discounted from the analysis. Temperatures varied minimally between sampling points during the experiment, though they rose in all sampling points during daytime as would be expected. pH was higher in abalone inbound and Ulva effluent water compared to the abalone effluent water. Total ammonia nitrogen percentage removal across the seaweed biofilters ranged from 65%-85% with the mean and median at 73% and 71% respectively. Free ammonia nitrogen percentage removal across the seaweed biofilters ranged from 41%-80% with the mean and median removals at 63% and 66% respectively. A regression analysis demonstrates a strong positive linear relationship between TAN removal and TAN load to the seaweed biofilter (r2= 0.90). Principal component analysis revealed a strong negative correlation between FAN removal and pH, as pH increased across the seaweed biofilters, the level of FAN removal decreased. This suggests that the perceived benefit of increasing pH in seaweed biofilters during the day-time may have some negative repercussions.
  • Thumbnail Image
    Item
    Open Access
    Diet of albacore (Thunnus alalunga) and yellowfin tuna (Thunnus albacares) off the south-west coast of South Africa
    (2022) Allie, Mogammad Ighsaan; Attwood, Colin
    Considering the magnitude of albacore and yellowfin biomass, these two economically vital tuna species apply a considerable amount of predation pressure on prey communities off the South African coast. Nevertheless, little is known about their respective diets in this upwelling region. Fish Aggregating Devices (FADs), otherwise used globally, are banned in South Africa therefore giving us a unique opportunity to examine the natural diet of the tuna surrounding our shores. Sixty-one stomachs were sampled from recreational fishing competitions in May 2016 and 2018 off the south-west coast of South Africa to investigate the diets of albacore and yellowfin tuna. The fork length (mm) and wet weight (g) of the fish were recorded. The importance of each prey in the diet was estimated by the Index of Relative Importance (IRI). Each sample specimen was cut open, had its entire stomach carefully removed and frozen for later dissection and analysis. Prey items were initially grouped into fish, cephalopods and crustaceans and later identified to the lowest possible taxonomic level, counted, and weighed. The difference in diet between the species was investigated and modelled by Analysis of Similarity (ANOSIM), Non-Metric Multidimensional Scale (NMDS) and Principal Component Analysis (PCA). The feeding strategy was determined by Costello's Diagram. The 43 albacore and 18 yellowfin tunas ranged in fork length from 740 to 959 mm and 803 to 1720 mm, respectively. The most important prey class based on IRI in the diet of albacore was crustaceans (6162.01), followed by cephalopods (3672.47). For yellowfin the highest IRI was cephalopods (6269.39) followed by fish (3977.53) and unlike the diets of yellowfin in other parts of the world, crustaceans were numerically a very low prey item making up just 15.84% of the diet. Yellowfin, the larger of the two tuna species consume larger prey items and are more opportunistic than albacore, showing a greater vertical feeding range, by diving deep for cephalopods and surface feeding on offal from hake trawlers. Intra-species variations for both tuna proved to be smaller than the difference between the two species, with no substantial changes in either diet based on fish size. Unlike diets of tuna species in other parts of the world, the data suggest that southern African albacore and yellowfin were less dependent on fish (possibly due to the lack of FADs in the case of yellowfin), but more likely due to the higher availability of crustaceans and cephalopods at the upwelling front. The magnitude of the role of cephalopods in the Benguela ecosystem is likely to be underestimated.
  • Thumbnail Image
    Item
    Open Access
    Reproductive biology, morphology and diet of Raja straeleni (spotted skate) in South Africa
    (2022) Azevedo, Alexandra; Attwood, Colin
    The spotted skate, Raja straeleni, is the most abundant and frequently caught skate in the waters of southern Africa. This study aims to add information on the morphology, reproductive biology and feeding habits of R. straeleni in southern Africa to better manage the species and to understand its role in the ecosystem. Samples were obtained from trawlers operating on the west and south coasts of South Africa during 2020. A total of 72 skates ranging in total length (LT) from 392 to 695 mm were examined. The 19 males ranged from 437 to 630 mm and the 53 females ranged from 392 to 695 mm. The following relationships were quantified (1) disk width (D)=0.77LT-11.96, (2) D=0.94Length to first dorsal-14.66, (3) male weight Wmale = 3.61LT-22.74, (4) Wfemale = 3.32LT-20.85. The length-at-50% maturity (L50) of the males was estimated at 57.1 cm (95% confidence interval: 54.2-59.4 cm), or 91% of the maximum male length recorded in this study. The total length-at-50% maturity (L50) of the females was estimated at 62.5 cm (95% confidence interval: 57.0-64.0 cm), or 90% of the maximum female length recorded in this study. Adult skates were collected during a period of sexual dormancy, as deduced from the low sperm presence and lack of oocytes. The diet of 57 skates were examined of which four had empty stomachs two contained only unidentifiable remains. Prey items spanning six phyla were recorded. In terms of %frequency they were Arthropoda (17%), Chordata, (8.2%), Mollusca (4.5%), Echinodermata (1.9%), Annelida (1.9%), and Nemertea (1.9%). The trophic level of R. straeleni was 3.98 ±0.05 SE. The trophic level was previously calculated as 3.74 (n = 421) (Ebert and Bazzarro 2007). The difference in findings is a result of a higher incidence of squid in the diet of skates examined in this study. The average trophic level of Raja species is 3.76 (n = 12, 95% Cl = 3.71 to 3.80), and for all skates it is estimated at 3.78 ±0.05 SE (n = 60) (Ebert and Bazzarro 2007). The diet of R. straeleni is very similar to that of other skates. Approximately 1000 t are landed per annum, predominantly from the inshore Agulhas Bank, which suggests a standing biomass at least ten times this amount (Attwood et al. 2011). R. straeleni, like other skates, is an important generalist carnivore in the Benguela system.
  • Thumbnail Image
    Item
    Open Access
    Seasonal characteristics of phytoplankton bloom phenology in the northern Benguela Upwelling System
    (2019) Matlakala, Mmakabele Lebogang; Thomalla, Sandy; Smith, Marie; Vichi, Marcello; Louw, Deon
    Seasonal phytoplankton blooms in the Benguela Upwelling System (BUS) play a crucial role in ecosystem services and driving climate change through air-sea gas exchanges. Upwelling systems are particularly are sensitive to effects of climate change including the transport of nutrients, which influence the composition of phytoplankton communities. This is important because species composition affects a number of key processes that have significant climate feedbacks. This study uses historical long-term in situ data (at 10 and 70 NM stations) as well as OC-CCI satellite ocean colour data to investigate seasonal phytoplankton bloom phenology and community structure of diatoms, dinoflagellates and coccolithophores in the northern BUS. The seasonal cycle of satellite chlorophyll was used to determine the timing of bloom initiation at inshore and offshore boxes that overlapped the in situ stations. An ocean colour algorithm to detect coccolithophore presence and absence was used to determine the inshore-offshore seasonal cycle of coccolithophores. Results indicate a gradual decrease in chlorophyll concentration further offshore as well as high intra-seasonal, inter-annual and spatial variability. Offshore blooms initiate later and last longer than inshore blooms which have a higher magnitude. Diatoms are dominant over dinoflagellates and coccolithophores throughout the northern BUS, with higher concentrations observed at the inshore station (10 NM) for all three phytoplankton groups. However, satellite results show a higher presence of coccolithophores in the offshore region during spring and summer that is associated with periods of strong stratification. This study provides a better understanding of the characteristics of the phytoplankton seasonal cycle in the northern BUS which is useful for detecting trends and possible change associated with climate change forcing in response to global warming.
  • Thumbnail Image
    Item
    Open Access
    Sexual segregation and abundance trend of whale sharks in southern Mozambique
    (2022) Auditore, Lisa-Marie; Pierce, Simon; Rohner, Chris; Reed, Cecile
    Coastal aggregations of endangered whale sharks (Rhincodon typus) consist mainly of juveniles, and most have a strong male bias. Generalised additive models were constructed on a 15-year dataset (2005–2019), from Praia do Tofo, Mozambique, to investigate sexual segregation in relation to environmental conditions. Temporal (year, day of year), and biophysical (sea surface temperature, time from high tide, moon illumination, Indian Ocean dipole index, wind, and chlorophyll a concentration) predictors were modelled in relation to the total, male, and female sightings. Encounter histories of individual sharks were then used in three capture-markrecapture model parameterisations to test for sex-specific demographic variables, and evaluate trends within the constellation. No differences in selected environmental variables were found between males and females, and demographic parameters were similar. Segregation does not occur from a preference for localised habitat conditions; while fewer females are present, they make use of the area similarly to males. Declining annual sighting trends corroborate trends in the seasonal abundance from multi-state open robust design models. Total abundance decreased over the study period (89%), with females (92%) declining at a greater rate than males (81%) suggesting exposure to different conditions or threats. The rapid decline indicates a need for a detailed threat assessment to assess potential causes.
  • Thumbnail Image
    Item
    Open Access
    Using seasonal mass-balanced models of the Algoa Bay ecosystem to investigate African penguin and small pelagic fish interactions
    (2019) Weigum, Emily; Shannon, Lynne; Jarre, Astrid
    Ecopath with Ecosim is a modeling software that allows the creation of mass-balanced models of the trophic flows of an ecosystem to explore the past and present impacts of fishing and environmental change on the trophic flows of a given food web. Currently, Algoa Bay supports the world’s largest breeding colony of endangered African penguins (Spheniscus demersus). The breeding success of African penguins is largely dependent on the availability of their food, mainly anchovy (Engraulis encrasicolus) and sardine (Sardinops sagax). African penguins breed year-round however, peak breeding season begins during the summer months (January to March) in Algoa Bay, when penguins begin building up fat reserves and laying eggs. In order to investigate the interaction between the small pelagic fish and penguin population, mass-balanced trophic models of the Algoa Bay ecosystem were constructed using the software Ecopath with Ecosim. Input parameters were derived from data compiled from published literature and survey data collected by DAFF (Department of Agriculture, Forestry, and Fisheries; formerly Marine and Coastal Management, MCM) and SAEON (the South African Environmental Observation Network). Two seasonal models were created to depict the summer (January to March) and winter (June to August) seasons from 2010-2014. Additionally, two seasonal models were created to represent a marine protected area where all fishery catch was set to zero. These static marine protected area models were created to investigate how the removal of fishery predation impacts the trophic structure of the Algoa Bay ecosystem by comparing ecotrophic efficiency values to those of the fished system. Two Ecosim simulations were used in fishery sensitivity analyses. The first, by setting all fishery catches to zero over a period of 30 years for each season to investigate the sensitivity of trophic groups to the removal of fishing pressure. The second, by setting fishery mortalities of anchovy and sardine to levels reported in the west coast (also over a period of 30 years) to investigate the sensitivity of trophic groups to an increase in small pelagic fishing pressure. Seasonal differences were observed with an increase in primary production and zooplankton biomass in the summer season compared to the winter season. The increase in plankton biomass resulted in an approximately 3x greater summer total system throughput, and total net primary production. Transfer efficiencies were higher than the average of 10% for aquatic ecosystems at trophic levels II and III with the seasonal averages being 17% and 19% respectively. The greatest seasonal change in modeled biomass occurred at trophic level III with summer biomass being 72% larger, indicating that small pelagic fishes benefit the most from the increase in summer plankton biomass. Linefish ecotrophic efficiency values were smaller in the marine protected area models in the summer and winter respectively, reflecting the reduced mortality on linefish under MPA conditions. Although the removal of fisheries resulted in an overall drop in predation pressure of 27% in the static MPA models, predation by other predatory trophic groups in the bay increased by 13% as a result of their subsequent biomass increases after the removal of fishery predation. Ecosim results showed the greatest change in biomass occurred in linefish which increased in both seasons when fishery catch was set to zero. The increase in linefish biomass can be attributed to the removal of fishery pressure on themselves and their prey, and may be an indication that linefish are overexploited in the bay. As a result of the biomass increase of some predator groups, some lower trophic level fish groups decreased despite the removal of fishery pressure. The results of the marine protected area analyses indicate that fisheries do not necessarily only have direct impacts on a target species as food-webs can have moderating effects. When fishing mortalities for sardine and anchovy were increased to west coast values, African penguins had the largest decrease in modeled biomass followed by other seabirds. Although, the observed decrease in modeled biomass of African penguins over the 30-year period does not fall within IUCN criteria for major concern, African penguins are already listed as endangered. An added decline of 10% on the largest breeding colony could have major implications on the future of the species. This study was a preliminary attempt at constructing mass-balanced trophic models of the Algoa Bay ecosystem, highlighting seasonal differences while investigating the possible impact of implementing a marine protected area in the bay and the sensitivity of trophic groups to fishing. Future research is needed to improve the more uncertain model parameters; however, these models are a good base for future work and the application of spatialized modeling of the bay using EcoSpace