Technical ReportPDF Available

Oceans and Coasts Annual Science Report 2021

Authors:
  • Oceans and Coasts
  • Department of Forestry Fisheries and the Environment Cape Town, South Africa

Abstract

Reports on research, monitoring and related activities of the Department of Forestry, Fisheries and the Environment's Chief Directorate: Oceans and Coastal Research, in support of the Departmental mandate of conserving and managing South Africa’s coastal and marine environment.
Oceans and Coasts
Annual Science Report
2021
Report No. 21
Oceans and Coasts
Annual Science Report
2021
Department of Forestry, Fisheries and the Environment
Oceans and Coasts
Annual Science Report
2021
Department of Forestry, Fisheries and the Environment
CONTENTS
SUMMARY FOR DECISION AND POLICY MAKERS...................................................... i
MONITORING PROGRAMMES
1. Continuing decline of the African penguin population in South Africa.............................................. 4
2. Long-term variation in the population and reproductive performance of
macaroni and eastern rockhopper penguins at Marion Island (1994–2019)....................................... 5
3. Long-term observations of currents on the Prince Edward Islands shelf............................................. 6
4. Long-term variability in bottom temperature on the Prince Edward Islands shelf............................. 7
5. Variability of wind speed and direction on the west coast of South Africa.......................................... 8
6. Long-term ocean acidication trends in St Helena Bay.......................................................................... 9
7. Chlorophyll variability on the west and south coasts.............................................................................. 10
8. Surface chlorophyll a concentrations along the St Helena Bay Monitoring Line................................ 11
9. Microplankton community structure and diversity along the
St Helena Bay Monitoring Line.................................................................................................................. 12
RESEARCH HIGHLIGHTS
10. Does localised cooling occur at the Prince Edward Islands?.................................................................. 13
11. Does large-scale climate variability inuence oceanography around the
Prince Edward Islands?................................................................................................................................ 14
12. New moored observations reveal contrasting oxygen seasonalities along
the southern Benguela coast........................................................................................................................ 15
13. Current reversals o Port Edward on the east coast of South Africa.................................................... 16
14. Oceanographic triggering of South Africa’s sardine run......................................................................... 17
15. Stormwater contribution to microplastics in coastal zones around Cape Town................................. 18
16. DNA metabarcoding of marine zooplankton in South Africa – how good is
the reference database?................................................................................................................................ 19
17. Macrofaunal community of a large temporarily closed estuary during
prolonged mouth closure............................................................................................................................. 20
18. A compromised immune system: the Cape urchin in a rapidly acidifying world............................... 21
19. Behavioural responses of Cape fur seals to swim-with-seal tourism activities
in the Robberg MPA..................................................................................................................................... 22
20. Ecological eectiveness of South Africa’s Marine Protected Areas........................................................ 23
21. e polycentric governance approach of the Benguela Current Commission..................................... 24
22. Interactions between Cape fur seals and Cape gannets at Malgas island –
a need for urgent management intervention.............................................................................................. 25
Oceans and Coasts Annual Science Report, 2021
23. e rst satellite tracking of movements of long-nned Pilot whales
in South Africa.............................................................................................................................................. 26
24. Eects of limited forage sh availability on African penguins............................................................... 27
25. Mortality event of Cape fur seals in South Africa during 2021.............................................................. 28
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION
AND TRAINING
26. OCIMS data management response eorts to the avian inuenza outbreak....................................... 29
27. New MIMS web portal improves data sharing and discovery................................................................ 30
28. Underway temperature measurements from the ermosalinograph
on the SA Agulhas II.................................................................................................................................... 31
29. GLORYS Ocean Model captures event-scale mesoscale eddies on the
southeast coast of South Africa.................................................................................................................. 32
30. Hydrography of the southeast coast of South Africa as determined
from the GLORYS Ocean Model................................................................................................................ 33
31. e South African Continuous Plankton Recorder Survey – mapping
plankton communities at the basin scale.................................................................................................. 34
32. e 2021 Western Indian Ocean Regional Benthic Imagery Workshop.............................................. 35
33. Training workshop on Biological Observations in the Indian Ocean................................................... 36
OUTPUTS FOR 2021
Peer-reviewed publications.................................................................................................................................... 37
Popular articles......................................................................................................................................................... 38
Presentations at symposia, conferences and workshops.................................................................................... 38
Published datasets.................................................................................................................................................... 39
Published reports..................................................................................................................................................... 41
Published training material.................................................................................................................................... 42
eses........................................................................................................................................................................ 42
Oceans and Coasts Annual Science Report, 2021
ACKNOWLEDGEMENTS
Most sta members of the Chief Directorate: Oceans & Coastal Research contributed in one way or another to
the contents and production of the Oceans and Coasts Annual Science Report, 2021. e Department wishes to
express its appreciation to the many other agencies that have contributed to the work presented in this report. e
at-sea, ship-based work and many coastal eld trips for data collection and community engagements undertaken
by the Branch: Oceans and Coasts are facilitated by the Chief Directorate’s science managers and made possible
by the various units within the Branch’s Corporate Management Services and Financial Management Services.
EDITORS
SP Kirkman, JA Huggett, T Lamont, T Haupt.
CONTRIBUTING AUTHORS
Ariefdien R, Basson R, Crawford RJM, Dyer BM, Halo I, Haupt T, Huggett J, Jacobs L, Kirkman SP, Kotze PHG,
Krug M, Lamont T, Louw GS, Maduray S, Makhado AB, Masotla MM, McCue SA, Naidoo AD, Nhleko J,
Rasehlomi T, Seakamela SM, Soeker MS, Tsanwani M, Tyesi M, van den Berg MA, Upfold L, Visagie L, Williams
L, Worship M (OC Research), Benjamin S (Animal Ocean), Pieterse J (Cape of Good Hope SPCA), Tan Shau
Hwai A (CEMACS), Findlay KP, Sparks C (CPUT), Monteiro PMS, Smith M (CSIR), de Goede J, van der Lingen
CD (Fisheries Management, Fisheries Research and Development), Kumar MN (INCOIS), Rixen T (Leibniz
Centre for Tropical Marine Research), Groeneveld JC, Singh S (ORI), Chiloane L (SAEON), Gridley T (Sea-
Search), Dakwa FE, Pfa M, Ryan P, Toolsee T (UCT), Cedras R (UWC), Lahajnar N (Universität Hamburg),
Teske PR (University of Johannesburg), Willows-Munro S (UKZN), Anthony T (Western Cape Department of
Agriculture)
CONTACT INFORMATION
Branch: Oceans and Coasts
Physical Address:
2 East Pier Shed, East Pier Road, Victoria & Alfred Waterfront
Cape Town,
Western Cape, South Africa
Tel: 021 819 2410
Website: https://www.environment.gov.za
Chief Director, Oceans and Coastal Research – Ashley D Naidoo (anaidoo@de.gov.za)
Director, Oceans Research – Ashley S Johnson (ajohnson@de.gov.za)
Director, Biodiversity and Coastal Research – Gerhard J Cilliers (gcilliers@de.gov.za)
Editors – Stephen P Kirkman (skirkman@de.gov.za), Jenny A Huggett (jhuggett@de.gov.za),
Tarron Lamont (tlamont@de.gov.za), Tanya Haupt (thaupt@de.gov.za).
COVER IMAGE: Compilation created by Gavin Tutt, with original photographs provided by Darrell
Anders and Tanya Haupt
RP13/2022
ISBN: 978-0-621-49993-3
Oceans and Coasts Annual Science Report, 2021
SUMMARY AND PERSPECTIVES FOR DECISION AND POLICY MAKERS
i Oceans and Coasts Annual Science Report, 2021
Introduction
e oceans constitute the largest component of the
Earth’s system and play a crucial role in stabilis-
ing climate and supporting life on earth and human
well-being. However, as indicated by the United Na-
tions’ First World Ocean Assessment (released in
2016), much of the world’s ocean is already severely
degraded, with changes and losses in the structure,
function and benets from marine systems. Fur-
thermore, with projected climate changes and hu-
man population growth, stressors on the ocean will
intensify, not only in localised coastal regions with
high human population densities, but also at a global
scale. us, the Decade of Ocean Sciences for Sustain-
able Development (2021–2030) was proclaimed by
the United Nations to support eorts to reverse the
cycle of decline in ocean health and create improved
conditions for sustainable development of ocean
economies while conserving the ecosystem. In this
decade, all countries are therefore urged to maintain
and increase science investments in describing and
understanding the ocean and the role it plays in the
planet system.
e Oceans and Coasts Annual Science Report (ASR),
2021, presents evidence of South Africas ongoing
investment in this regard. e 33 contributions
(termed report cards) herein, report on research,
monitoring and related activities of the Department’s
Chief Directorate: Oceans and Coastal Research
(OC Research), in support of the Departmental
mandate of conserving and managing South Africa’s
coastal and marine environment. e various sci-
ence programmes of OC Research focus on a num-
ber of fundamental physical, chemical and biological
aspects of oceans and coasts (including estuaries),
and are guided by a medium- to long-term ecologi-
cal research and monitoring plan that was devel-
oped for the period 2016–2030. is plan is focused
mainly on describing and documenting marine and
coastal biodiversity and complex ecosystem function-
ing and processes, to support the Department’s ocean
mandate.
us, the data and information products generated
from the research and monitoring activities must
ultimately be useful for informing managers and
policy makers. Underlying the plan is the understand-
ing that the most valuable scientic data collections
or observations are those taken within a long-term
context. However, within a framework of long-term
programmatic work aimed at providing continuous
or sustained observations (monitoring) and descrip-
tions of key aspects of the marine environment, some
shorter-term research elements are conducted as
projects. Such projects allow for deeper understand-
ing of marine ecosystems and processes. It is against
this backdrop of continuous applied shallow and
deep ocean science that sta are able to provide im-
mediate and relevant management advice and rec-
ommendations. Such management advice is required
on a range of historical issues such as shing or ship-
ping and new and emerging issues such as seismic
surveys or where to place the increasing number of
undersea internet cables.
As in the previous issue (2020), the contributions to
this report are presented under three main sections.
ese are: Monitoring Programmes (9 report cards);
Research Highlights (16 report cards), showcasing
research achievements for 2021; and Technologi-
cal Innovation and Training (8 report cards), which
was previously called Tools and Technologies. In ad-
dition to showcasing technological developments in
support of OC Researchs various programmes, the
latter section now incorporates evidence of training
and capacity building, another key element of the
research and monitoring plan. At the end of the
report is the list of scientic outputs for 2021, in-
cluding peer-reviewed publications and other prod-
ucts that reect both the volume and quality of work
accomplished by OC Research in 2021.
Overview of the 2021 Annual Science Report
From the perspective of OC Research, there are
a few issues that stand out in 2021. One of these is
the alarming continuance of the decline in African
penguin numbers, as debate on a management in-
tervention aimed at securing sucient foraging sh
availability in range of important breeding colo-
nies has been drawn out. Report cards in this issue
demonstrate the continuing penguin decline and the
link with forage sh availability. Just as important
as securing urgent management interventions, argu-
ably, is understanding why there does not seem to be
enough prey to sustain the population. In our coastal
waters, it has proven dicult to disentangle eects
of shing from environmental changes on the prey
resources of penguins and other top predator species.
For macaroni and Eastern rockhopper penguins at
the Prince Edward Islands (also reported on), this is
less problematic because shing activity can be ruled
out as the cause of short- to long-term variability in
their foraging and demographic parameters. How-
ever, to properly understand the causes of declines
in reatened seabird populations in South Africa,
we will need to advance our understanding of the
oceanographic and atmospheric variations. Such
understanding of the physical environment allows for
insights into factors that drive (or are symptomatic of)
spatial and temporal changes in the productivity of
marine ecosystems, and their implications for higher
trophic levels. In this issue, updates are provided for
monitoring programmes on several Essential Ocean
Variables (EOVs) including currents and bottom
temperature (both on the Prince Edward Islands
shelf), upwelling-favourable winds, ocean acid-
ity, chlorophyll a concentration and microplankton
community structure and diversity (all o the west
coast of South Africa).
SUMMARY AND PERSPECTIVES FOR DECISION AND POLICY MAKERS
SUMMARY AND PERSPECTIVES FOR DECISION AND POLICY MAKERS
Oceans and Coasts Annual Science Report, 2021 ii
Another standout issue in 2021 was the occurrence
of extreme events, in the form of large-scale mortal-
ity events aecting seals and seabirds, both referred
to in report cards in this issue. Several hundred
dead or dying seals were recorded on the west coast
between September and December. Inspection of
the environmental variables at the time revealed no
obvious trigger for the seal mortality. Pathological
and toxicological testing of tissue samples also yielded
no clear results, but given the malnourished condi-
tion of most aected seals, reduced availability of prey
is considered likely to be the primary cause of the
die-o. Seal mortality events of the scale observed
in 2021 are unprecedented in South Africa in recent
decades, but are not that uncommon in the North-
ern Benguela (Namibia), where the ecosystem is in a
depressed state having undergone trophic changes
that are thought to be irreversible. e event may
therefore provide a warning sign of an ecosystem
shi associated with lower trophic level changes, un-
derlining the importance of continuing and enhanc-
ing the monitoring of oceanographic and biological
EOVs in the region. It also underlines the utility of top
predators such as seals as bio-indicators of ecosystem
variability and change.
Unrelated to the seal mortality, nearly 20,000 birds,
most of them Endangered Cape cormorants, were
recorded to have died from avian inuenza in the
Western Cape. e scale of these two mortality events
has highlighted the need for preparedness, to be able
to detect mortality events at an early stage, diagnose
them, prevent spread in the case of epidemics, and
to collect sucient, adequate scientic data. In both
cases, the mortality events provided learning experi-
ences that will lead to improved responses to further
events. In the case of seals, a monitoring protocol is
now under development. For the seabirds, the de-
velopment of an app allowing for coherent data re-
porting and capture to the Department’s Oceans and
Coastal Management Information System (OCIMS)
is described in this report. e app enabled eective
tracking of the spread of the epidemic and assisted
the coordination of responses, highlighting the
benets of utilizing innovative technology to facilitate
the capture and ow of information in such events.
A range of research highlights are presented in the
report. ese include ndings of studies on oceano-
graphic processes in the Southern Ocean, the south-
ern Benguela and the east coast of South Africa. One
of these report cards describes the oceanographic
triggers for South Africa’s “sardine run” on the east
coast, an event that is considered to be one of Earth’s
most spectacular marine migrations, and that has
considerable ecological and economic signicance.
Further research highlights report on studies as
diverse as zooplankton barcoding, estuarine commu-
nities, coastal pollution, marine ecotourism, marine
protected area eectiveness, ocean governance, wild-
life interactions and tracking of animal movements
at sea. An experimental study shows how reducing
pH of seawater compromises the immune system
of Cape urchins, drawing attention to the types
of challenges that this creature can be expected to
face under the two-fold increase in acidity of sur-
face ocean waters that is anticipated by the end of
this century. A decline in the biomass of this spe-
cies is likely to have knock-on ecosystem eects for
other species that are dependent on it such as rock
lobsters (as prey) and juvenile abalone (for refuge),
while kelp bed ecosystems are also likely to be altered
if these important grazers are reduced. ese types
of studies are essential for forecasting the eects of
predicted oceanic or atmospheric changes on the
structure and functioning of our marine ecosystems,
and their potential social and economic implications.
A range of innovative tools or technologies that have
been developed in support of research, monitoring,
data capture or dissemination are reported on. In
addition to the OCIMS app that was developed for
responding to the avian inuenza epidemic, the de-
velopment of a new web portal has improved ca-
pacity for data information and discovery in the
Department’s Marine Information Management
System. Technology for underway sampling and
measurement of EOVs including temperature and
plankton are described, as well as testing of a so-
phisticated ocean model, in terms of its ecacy in
simulating the variability and distribution of cer-
tain oceanographic processes in our waters. ere
is a widespread lack of adequate long-term in situ
observations at appropriate spatial scales to identify
and monitor oceanographic processes and features
and their impacts through the water column. Towards
countering the lack of observations to some degree,
the global community has, in recent decades, become
reliant on ocean models to elucidate such changes
and variations. e southeast coast of South Africa
is one example of a region where limited in situ ob-
servations have severely constrained our understand-
ing of ecosystem variability and change, necessitat-
ing the use of ocean models to generate improved
knowledge. However, as described in the report card,
it is very important to recognise that the accuracy of
all ocean models is highly dependent on the amount
and quality of in situ and satellite observations used
to parameterise them. is further highlights the
critical need for continued and improved in situ
observations and investments in ship and satellite
technologies for coastal countries like South Africa
that are bordered by large ocean spaces.
e report concludes with report cards on two train-
ing workshops spearheaded by OC Research for the
Indian Ocean region. Ironically, the need to conduct
these workshops through a virtual platform, due to
physical restrictions for the ongoing global Covid-19
pandemic, increased the participation and impact of
these workshops. In addition to training and capacity
building, these workshops brought about enhanced
regional networking, collaboration and mentor-
ship opportunities, and moved the region closer to
adoption of standardised methodology and data
collection protocols that will enable region-wide
iii Oceans and Coasts Annual Science Report, 2021
comparisons and integrated data analyses. Imagery
from one of these training initiatives inspired the
cover image of this report.
Concluding remarks
e range of report cards to the Annual Science
Report, 2021, attests to OC Researchs dedication to
development of sta capacity, with contributions
accepted from students, interns, scientists, technical
sta and senior managers. Also evident is the Chief
Directorate’s openness to collaborating with other
organisations, including partnering with the various
national marine research nodes in other departments
and universities, and with regional and international
institutions. No less than 17 external organisations
are represented in the list of contributors to this
report. As an index of research productivity, the
number of peer-reviewed publications achieved in
2021 showed a pleasing increase relative to previous
years. Also gratifying is that three-quarters of these
outputs were accepted by international publications,
attesting not only to the volume of work, but also the
signicant contribution of South African scientists to
global knowledge.
At the time of writing this, Parties to the Conven-
tion on Biological Diversity (CBD) are in the process
of developing the Post-2020 Global Biodiversity
Framework (GBF), which will set out an ambitious
plan to implement broad-based action to bring about
a transformation in society’s relationship with biodi-
versity. e overarching aim of this plan is to ensure
that by 2050 the CBD’s vision of “living in harmony
with nature” is achieved. Several Parties (includ-
ing South Africa) and organisations have been lob-
bying for better inclusion of marine and coastal
biodiversity considerations in the GBF, compared to
the CBD’s Strategic Action Plan, 2011–2020, which
the GBF will replace. Greater representation of the
oceans and coasts in the goals, targets and monitoring
framework of the GBF will increase the investment
in marine and coastal monitoring and research that
is required of Parties, including for tracking the suc-
cess of the framework’s implementation. e moni-
toring programmes, research output and techno-
logical innovations that are described in this report
demonstrate that South Africa, with OC Research
at the forefront, is well placed to address this
challenge.
SUMMARY AND PERSPECTIVES FOR DECISION AND POLICY MAKERS
Oceans and Coasts Annual Science Report, 2021 4
MONITORING PROGRAMMES
1 . CONTINUING DECLINE OF THE AFRICAN PENGUIN POPULATION IN SOUTH AFRICA
In two of the three geographical regions in which they occur
in South Africa, i.e. the west coast (Western Cape colonies
to the north of Cape Town) and southwest coast (Western
Cape colonies south and east of Cape Town), the ongoing
decline in numbers has stabilised during this assessment,
slowing from ca. 9%, to ca. 3% per year in the past decade
(Figs. 2A and B). However, the ca. 3,300 breeding pairs lost
in these regions between 2019 and 2021, represents almost
25% of the remaining South African population. In Algoa
Bay in the Eastern Cape, colonies have been in decline for
the past 30 years (Fig. 2C). Recently (2019), the long-term
decline in this region was measured at 66%, but in just over
two years, the decline since 2001 has increased to 82%. e
magnitude of this decline would be sucient to qualify the
species to be up-listed to Critically Endangered (CR) if the
IUCN Red List criterion A2 were applied at a regional level.
e decline was exacerbated by the loss of more than 50%
of the breeding pairs from the St Croix Island colony. is
has been directly linked to food scarcity, with numerous
chicks observed to be malnourished, and high chick mor-
tality.
Increased ship trac, seismic surveys and oil leaks or spills
associated with ship-to-ship bunkering are other threats to
African penguins in this region, but prey shortages are con-
sidered to be the main cause of the decline. Two of the main
prey species of the African penguin, sardine and anchovy
(forage sh) have rarely been detected in recent diet sam-
ples collected at the colony. Local and regional availability
or abundances of forage sh have been linked to breeding
pair numbers of penguins, the ratio of adult to immature
penguins, and foraging performance. Urgent management
recommendations that may benet the African penguin
population include improving food availability, for exam-
ple by preventing or reducing shing eort in key foraging
areas, and mitigating the impacts of oil spills.
e African penguin Spheniscus demersus (Fig.1) was once South Africa’s most abundant seabird. It is Africa’s only penguin
species and is endemic to the Benguela upwelling region, as this species breeds only in Namibia and South Africa. In a 2019
assessment, it was estimated that the overall population had declined by 68% between 1991 (when there were ca. 42,500
breeding pairs) and 2019. Based on that assessment, the Endangered status of the population, rst conferred in 2010, was
upheld. However, with inclusion of more recent counts up to 2021 (ca. 10,500 breeding pairs), the overall decline since
1991 has increased to 73%. Should this decline persist, the African penguin population may reach a critical point of no
return.
Figure 2. Changes in the African penguin breeding population since
1979 within three regions of South Africa: (A) the west coast, (B) the
southwest coast, (C) the Eastern Cape. e solid line is the number
of breeding pairs based on nest counts, the grey polygon represents
variability in the probability of the estimate, and dashed lines are the
10-year generation lengths up to 2021.
Authors: Makhado AB, Dyer BM, Masotla MJ, Upfold L (OC Research)
Year
35
35
35
30
30
30
25
25
25
20
20
20
15
15
15
10
10
10
5
5
5
0
0
0
Breeding Pairs x 1000
A
B
C
1979 1985 1991 1997 2003 2009 2015 2021
-3G -2G -1G 2021
-3G -2G -1G 2021
-3G -2G -1G 2021
Figure 1. African penguins, photo from the DFFE photo library.
5 Oceans and Coasts Annual Science Report, 2021
MONITORING PROGRAMMES
2. LONG-TERM VARIATION IN THE POPULATION AND REPRODUCTIVE
PERFORMANCE OF MACARONI AND EASTERN ROCKHOPPER PENGUINS AT
MARION ISLAND (1994–2019)
e mass of breeding adults upon arrival at the islands
was sampled each year. No consistent long-term trend was
found for macaroni penguins (Fig. 2A), but there was a de-
creasing trend in the average mass of both male and female
eastern rockhopper penguins from 1994 to the mid-2000s
(Fig. 2B). is may explain the declining population trend
of eastern rockhopper penguins until 2001. eir decline
stabilised in the mid-2000s before increasing to the end of
the time series. Average mass on arrival of breeding pairs
was positively correlated with breeding success for both
species (r ≥ 0.34, p < 0.1). e two species, males in par-
ticular, undergo long fasting periods during breeding, with
these results underscoring the importance of pre-breeding
condition for breeding success, and hence the importance
of winter foraging conditions.
ere was no obvious long-term trend in breeding suc-
cess of macaroni penguins (measured by successful rear-
ing of chicks to edgings) but breeding success of eastern
rockhopper penguins increased over the latter part of the
time series (Fig. 3), corresponding with stabilisation of
population numbers. For macaroni penguins, there was a
positive correlation between average annual edgling mass
and breeding success (r = 0.40, p = 0.04), while for eastern
rockhopper penguins this relationship was weak (r = 0.12,
p = 0.57). is underscores the importance of ecient for-
aging during the chick rearing period, to produce larger,
fatter edglings with good chances of survival and recruit-
ment back into the breeding population.
e high interannual variability in breeding success is
therefore likely aected by the environmental conditions
that aect prey availability and thus feeding eciency, both
before and during the breeding period. Long-term changes
in environmental conditions that may have aected feeding
conditions for the Eudyptes spp. at the PEIs include a south-
ward shi in the sub-Antarctic front and an increase in sea
surface temperatures recorded around Marion Island since
the mid-20th century.
Penguins have a conservative life history, characterised by long generation times and high adult survival rates. Research on
their population dynamics can provide insights into variability and change in ocean systems. e Prince Edward Islands
(PEIs; consisting of Prince Edward and Marion Islands) in the Southern Ocean supports ca. 300,000 and 80,000 breeding
pairs of macaroni Eudyptes chrysolophus and eastern rockhopper E. lholi penguins, respectively. Long-term monitoring
by DFFE shows that macaroni penguin numbers at Marion Island declined by 45% from 1994–2019, at a rate of 1.9% per
year. Eastern rockhopper penguins declined even more over the same period, by 66% at a rate of 13% per year. However,
their numbers then stabilised, uctuating between 55,000 and 85,000 pairs (Fig. 1).
Figure 1. Time series of the number of breeding pairs of macaroni and
eastern rockhopper penguins at Marion Island, 1994–2019.
Figure 2. Time series of the average mass on arrival of (A) macaroni and
(B) eastern rockhopper penguins at Marion Island, 1994–2019.
Figure 3. Time series of the breeding success (chicks edged per breed-
ing pair) of macaroni and eastern rockhopper penguins at Marion Island,
1994–2019.
Authors: Dakwa FE (UCT, OC Research), Ryan P (UCT), Crawford RJM,
Dyer BM, Masotla MM (OC Research), Makhado AB (OCResearch, UCT)
Oceans and Coasts Annual Science Report, 2021 6
MONITORING PROGRAMMES
During 2014–2021, daily mean current speeds at mooring
M1 ranged between 0.01 and 50.90 cm s-1, while those at
M2 varied from 0.03 to 67.32 cm s-1 (Fig. 2). e eastward-
owing Antarctic Circumpolar Current results in much
stronger zonal (east/west) than meridional (north/south)
ow at the PEIs. A Taylor column (stationary anticyclonic
circulation over the shelf) is indicated by westerly ow in
the bottom waters at M2 (Fig. 2). is westerly ow is per-
sistent throughout the time series, but can be enhanced or
interrupted for short periods when fronts or mesoscale ed-
dies interact with the island shelf. Retention of nutrients
and biota by the Taylor column maintains enhanced pro-
ductivity on the shelf, accounting for the high concentra-
tions of marine biota at the PEIs.
Between May and July 2019, the close proximity of the
southern branch of the sub-Antarctic Front (S-SAF) to the
islands resulted in the strongest ow to date (>50 cm s-1)
at M2. A similar situation occurred between August and
November 2020, when the S-SAF resulted in surface cur-
rent speeds at M2 exceeding 40 cm s-1 (Fig. 2). During this
period, currents in the upper 100 m were directed strongly
southeastwards. is contrasts with the situation in winter
2019, when advection of shelf waters was strongly north-
and northeastwards.
During the winter months between 2014 and 2019, data
loss (vertical extensions of white shading interrupting
the time series) was common in the upper 60 m at M2
(Fig. 2). is reects the turbulent surface conditions driv-
en by strong wind mixing associated with winter storms.
Interestingly, during winter 2020, such data loss was not as
evident, possibly suggesting less intense storm events than
during previous years. is seems to be supported by the
generally warmer bottom shelf waters in the inter-island
region at this time (see Report 4).
Changes in the direction of current ow can be expected to
inuence the distribution of preferred prey, and hence the
feeding patterns of seabirds and marine mammals breeding
at the PEIs. Further detailed comparisons between currents
and feeding behaviour, diet, and reproductive performance
of selected predators are required to evaluate the impact of
these changing ow dynamics on top predators.
3. LONG-TERM OBSERVATIONS OF CURRENTS ON THE PRINCE EDWARD
ISLANDS SHELF
e Prince Edward Islands (PEIs) are a remote island archipelago in the sub-Antarctic zone of the Southern Ocean. e
islands provide crucial breeding habitat for vast populations of seabirds and marine mammals. It is well-known that there
are strong links between the oceanography and biological communities, but observations have been largely limited to
periods coinciding with annual relief voyages to re-supply the research base. Since April 2014, two moorings on the inter-
island shelf (Fig. 1) have been providing continuous measurements of water column current speed and direction in the
region.
Figure 1. Bathymetry on the PEI shelf. Mooring positions M1 and M2 are
shown in red and blue, respectively. e pink dot shows the location of M1
between April 2016 and April 2017.
Figure 2. Daily mean zonal and meridional current components (cm s-1)
at (a) M1, and (b) M2. Positive values denote eastward (zonal) and north-
ward (meridional) ow; negative values denote westward (zonal) and
southward (meridional) ow. Dashed red lines show the period of strong
eastward ow at M2 between August and November 2020. e period
between January and February 2021 has also been marked with dashed
red lines (see Report 4 for details on this event).
Authors: van den Berg MA, Lamont T (OC Research)
Contributors: Jacobs L, Louw GS (OC Research)
7 Oceans and Coasts Annual Science Report, 2021
MONITORING PROGRAMMES
4. LONG-TERM VARIABILITY IN BOTTOM TEMPERATURE ON THE PRINCE EDWARD
ISLANDS SHELF
Substantial daily, intra-, interseasonal and interannual
variability was observed at both moorings, but bottom
temperatures at the deeper mooring M2 (260 m depth)
were consistently lower than those at M1 (174 m depth).
On a seasonal scale, highest temperatures typically occur
in autumn (March–May) and the lowest in spring (Sep-
tember–November). Meridional (north/south) meanders
of the southern branch of the sub-Antarctic Front (S-SAF)
resulted in lower temperatures when the S-SAF was north
of the PEIs, and higher temperatures when the S-SAF was
to the south.
Since January 2019, bottom temperatures have been mainly
higher than the 7-year mean (Fig. 2). ese elevated tem-
peratures reect a more southerly location of the S-SAF
relative to the islands. Persistent strong northward ow
at M2 (see Report 3) between May and October 2019 was
associated with bottom temperatures that were 0.5–1.5°C
above average (Fig. 2). In contrast, elevated current speeds
observed at M2 (see Report 3) between August and No-
vember 2020 were associated with waters that were on
average 0.5–1.0°C cooler than the 7-year mean (Fig. 2). A
similar temperature decrease occurred during January and
February 2021. Interestingly, there was no clear change in
current speed during this cooling event (see Report 3).
Notably, bottom temperatures during late autumn and win-
ter months (May–July) in 2020 were on average 0.5–1.0°C
warmer than previous years (Fig. 3). Although generally
warmer waters were expected due to the more southerly
position of the S-SAF, such warming was not evident in
2019 (Fig. 2), when the S-SAF was also mainly south of the
PEIs. is suggests less intense wind mixing and cooling
during May–July 2020.
Water temperatures inuence geographical distributions
of prey species on which the vast numbers of seabirds and
marine mammals breeding at the PEIs depend. Tempera-
ture variations are thus likely to aect the distances that
these animals have to travel from the islands to nd food.
is has consequences for time and energy spent foraging,
for survival of young that are dependent on foraging adults,
and ultimately for the reproductive success and abundance
trends of these populations. e cooling events during late
2020 and early 2021 likely resulted in elevated productivity
on the PEI shelf.
Despite their small size, the Prince Edward Islands (PEIs) provide crucial breeding habitat for vast populations of marine
mammals and birds. Many of these animals depend strongly on the ambient oceanographic conditions at and around the
islands. While annual relief voyages to re-supply the research base only allow hydrographic data collection during April/
May each year, two moorings on the inter-island shelf (Fig. 1) have been providing continuous measurements of bottom
temperature since 2014.
Figure 1. Bathymetry on the PEI shelf. Mooring positions M1 and M2 are
shown in red and blue, respectively. e pink dot shows the location of M1
between April 2016 and April 2017.
Figure 2. Daily mean bottom temperature (°C) at moorings M1 (red) and
M2 (blue). Dashed lines indicate measurements while solid lines show
low-pass ltered values. Horizontal lines show mean temperatures for
each time series. Grey shading highlights cooling events between August
and November 2020 and during January–February 2021.
Figure 3. Monthly mean bottom temperature (°C) at mooring M2.
Authors: van den Berg MA, Lamont T (OC Research)
Contributors: Jacobs L, Louw GS (OC Research)
Oceans and Coasts Annual Science Report, 2021 8
MONITORING PROGRAMMES
5. VARIABILITY OF WIND SPEED AND DIRECTION ON THE WEST COAST OF
SOUTH AFRICA
Figure 1 shows typical upwelling conditions in the SBUS
during summer, with southeasterly winds (black vec-
tors) prevailing and sea surface temperatures as low as
ca. 14°C inshore, with warmer (ca. 25°C) water further o-
shore. While daily averages (grey lines) of wind speed and
direction from the AWS at Elands Bay and Cape
Columbine show large variability, monthly averages (red
and blue lines respectively) indicate strong seasonality
(Fig. 2). During summer, upwelling-favourable (southeast-
erly) winds are strongest (>5 m s-1 on average), while in
winter, winds are weakest and from the west (Figs. 2
and 3). In general, wind speed at Elands Bay seems to be
weaker than at Cape Columbine. us, upwelling is
generally greater at Cape Columbine.
Over the past ve years, the strongest summer-autumn
winds were observed in 2017, implying that upwelling was
most intense during this year. At Cape Columbine, these
winds were ca. 1 m s-1 higher than the ve-year average.
In Jan–Feb 2021, winds at Elands Bay were ca. 0.5 m s-1
higher than the ve-year average (Fig. 3). ese results
show that AWS are providing valuable information to en-
hance our understanding of short-term, localised upwelling
variations.
In upwelling systems, ocean productivity along the coast is largely inuenced by wind-controlled processes. Coastal up-
welling is generally dened by cold and nutrient-rich sea surface water that occurs in summer. In the Southern Benguela
Upwelling System (SBUS) on the west coast of South Africa, these cold waters are associated with the occurrence of strong
southeasterly wind forcing, which is controlled by the seasonally varying latitudinal (north-south) migrations of the South
Atlantic Atmospheric High Pressure System. erefore, wind speed and direction are important variables for describing
upwelling-related coastal features and monitoring long-term changes in coastal upwelling. In this report, daily in situ data
obtained from Automatic Weather Stations (AWS) at Elands Bay and Cape Columbine over the period 2015–2022 (Fig. 1)
are used to describe the variability of wind speed and direction at these locations.
Figure 1. Map of satellite-derived sea surface temperature with wind vec-
tors (10 m above sea level), averaged over the 2018–2021 period, depicting
the average upwelling conditions during summer (December to Febru-
ary). e size of the wind vectors shows wind speed (longer/shorter vec-
tors indicate stronger/weaker winds) and the wind direction is indicated
by the direction of the vector. e map inset shows the locations of AWS
stations at Elands Bay and Cape Columbine.
Figure 2. Time series (2015–2022) of daily (grey lines) and monthly (bold
lines) averages of wind speed (red) and wind direction (blue) at Elands Bay
(top) and Cape Columbine (bottom). Northerly, easterly, southerly and
westerly winds are represented by 0/360°, 90°, 180° and 270°. Gaps
indicate periods when data was corrupted or not available.
Figure 3. Annual variability of daily wind speed at Elands Bay (top) and
Cape Columbine (bottom). e 2017 data are highlighted in red, the 2021
values are indicated in green, and the ve-year (2016–2021) mean is shown
in blue. Black vectors indicate the monthly average wind strength and
direction.
Author: Tyesi M (OC Research)
9 Oceans and Coasts Annual Science Report, 2021
MONITORING PROGRAMMES
Dissolved inorganic carbon and total alkalinity observa-
tions, collected quarterly from 2013 to 2019, were used as
a baseline to reconstruct a 20-year record (2000–2019) of
surface pH and Ωarag. Extended Multiple Linear Regres-
sion (eMLR) was applied to monthly averages of reanalysis
surface temperature and salinity (0.25° spatial resolution),
chlorophyll a (4 km spatial resolution), and atmospheric
CO2. Here we present time series of eMLR reconstructed
surface pH and Ωarag at Stations 3, 7, and 12 of the SHBML
(Fig.1).
Figure 2 shows that over the 2000–2019 period, surface pH
and Ωarag in the inner shelf region (Station 3) declined
from 7.91 to 7.81 and 1.89 to 1.52, respectively. Surface
pH and Ωarag at Station 7 (Fig. 3) declined from 8.07 to
7.96 and 2.69 to 2.19, respectively. In the outer shelf region
(Station 12), surface pH and Ωarag declined from 8.17 to
8.06 and 3.47 to 2.80, respectively (Fig. 4). ese decreasing
trends were all statistically signicant.
6. LONG-TERM OCEAN ACIDIFICATION TRENDS IN ST HELENA BAY
ese results indicate that due to the uptake of anthro-
pogenic CO2 emissions by the oceans, surface waters on
the west coast of South Africa are becoming more acidic
(decreased pH) and more corrosive (decreased Ωarag).
ese changes need to be monitored due to their detrimen-
tal eects on calcifying marine organisms.
Figure 1. Map showing Stations 3, 7, and 12 (red squares) of the St Helena
Bay Monitoring Line (SHBML).
Figure 2. Time series (solid lines) and linear trends (dashed lines) of
surface pH and aragonite saturation state (Ωarag) from 2000 to 2019 at
Station 3 of the SHBML.
Figure 3. Time series (solid lines) and linear trends (dashed lines) of
surface pH and aragonite saturation state (Ωarag) from 2000 to 2019 at
Station 7 of the SHBML.
Figure 4. Time series (solid lines) and linear trends (dashed lines) of
surface pH and aragonite saturation state (Ωarag) from 2000 to 2019 at
Station 12 of the SHBML.
Coastal upwelling regimes inshore of Eastern Boundary Currents continue to be the most biologically productive ecosys-
tems in the global ocean. However, changes resulting from human activities have already begun to emerge in these regions.
Among these, carbon dioxide (CO2) derived from fossil fuel combustion is lowering the pH and aragonite saturation
state (Ωarag) levels in seawater. e reduction in Ωarag, a measure of the availability of calcium carbonate (CaCO3), is
detrimental for calcifying marine organisms such as corals, pteropods, molluscs and foraminiferans. ese organisms are
unable to develop their shells or skeletal structures when seawater is under-saturated with respect to CaCO3 (Ωarag <1).
St Helena Bay is the most productive region of the southern Benguela and is an important nursery area for pelagic sh. It
is also an area that is subject to hypoxia and anoxia, which has occasionally severely impacted marine resources.
Authors: Tsanwani M (OC Research), Monteiro PMS (CSIR)
Contributors: Mtshali T, Mdokwana BW, Vena K, Kiviets G, Siswana K,
Britz K (OC Research)
Oceans and Coasts Annual Science Report, 2021 10
7. CHLOROPHYLL VARIABILITY ON THE WEST AND SOUTH COASTS
Higher values are associated with greater phytoplankton
biomass and a more productive ecosystem, while lower
values indicate lower biomass and a less productive eco-
system. Highest index values are usually found o Namibia
(16–26°S; Fig. 2). In 2018, biomass was the lowest since
2013. While phytoplankton biomass was elevated during
2019 and 2020, the 2021 values were once again lower.
Persistent upwelling and oshore transport at Lüderitz
(ca. 27°S) are typically associated with very low index val-
ues, but elevated values during summer over the past two
years suggested less upwelling than usual. Along South
Africa’s west coast (28–34°S), index values are elevated
around the Namaqualand, Cape Columbine and Cape Pen-
insula upwelling cells. O Namaqualand (28.5–30°S), 2018
showed the highest values since 2013. While the 2019 and
2020 values were only slightly lower, index values during
January to August 2021 were notably much lower, reect-
ing even less productive conditions. Along South Africa’s
south coast (18.5–29°E), index values are generally lower
than on the west coast. During 2013-2021, the highest val-
ues occurred at 22°E in January–February 2014, with re-
duced peak biomass levels in subsequent years. Low values
in 2016 suggested that this was the least productive year on
the south coast during 2013–2021. While index values were
elevated east of 22°E during 2019 and 2020, lower values
overall in 2021 suggested lower productivity. e region-
ally-varying trends in productivity (a small but signicant
long-term increase in chlorophyll a o Namibia, and a de-
crease o the west coast of South Africa and on the Agulhas
Bank) appear to have continued over the past year.
Author: Lamont T (OC Research)
Phytoplankton are crucial for a number of key marine processes, such as food web modulation, CO2 exchanges, and the
cycling of carbon and other nutrients. On the west and south coasts of southern Africa, the Benguela upwelling system and
the Agulhas Bank are economically and ecologically signicant as they host productive ecosystems with complex trophic
structures that support numerous commercially harvested resources. To monitor environmental conditions, an index
of chlorophyll a is computed by integrating satellite-derived surface values from the coast to the 1 mg m-3 level further
oshore (Fig. 1).
Figure 1. Annual average chlorophyll a concentration and location
of the 1 mg m-3 contour (thick line).
Figure 2. Monthly chlorophyll a indices (1997–2021) for the west coasts of Namibia and South Africa (top panel) and for South Africa’s
south coast (bottom panel).
MONITORING PROGRAMMES
11 Oceans and Coasts Annual Science Report, 2021
MONITORING PROGRAMMES
8. SURFACE CHLOROPHYLL A CONCENTRATIONS ALONG THE ST HELENA BAY
MONITORING LINE
Along the west coast, southeasterly winds drive upwelling,
which transfers surface waters oshore, resulting in cool,
nutrient-rich waters being uplied to the surface from
deeper depths. On a seasonal scale, higher chlorophyll a
concentrations coincide with larger amounts of upwelling,
which occur during the upwelling season (October–March)
each year. Satellite-derived surface chlorophyll a clearly
illustrate this seasonal signal, with maxima in spring/early
summer and late summer/autumn (Fig. 2).
Higher chlorophyll a is usually associated with greater
phytoplankton biomass and a more productive ecosystem,
which largely results from the higher availability of nutri-
ents in the upper layers during upwelling. In contrast, lower
chlorophyll a indicates lower phytoplankton biomass and a
less productive ecosystem, usually associated with less up-
welling and nutrient availability in the surface layers during
late autumn to early spring (April–September) each year.
Generally, higher chlorophyll a occurs close to the coast
and decreases with distance oshore (Fig. 2).
During 2015, high values (>20 mg m-3) extended ca. 20 km
oshore in autumn (March) and late spring/early summer
(September to November). In contrast, values >20 mg m-3
were observed much closer to the coast during 2016–2019
and in 2021. During 2020, such high values extended about
10 km oshore in February-March. Elevated chlorophyll
(>5 mg m-3) extended approximately 110 km oshore in
March 2015 – the farthest oshore extent for such elevated
values since March 2010. In 2016, the farthest oshore ex-
tent (ca. 80 km) of values above 5 mg m-3 was observed in
February and March, but in subsequent years, such values
did not extend beyond 70 km oshore.
Chlorophyll a in 2017 was lower than in 2016 but remained
elevated throughout the year. Even lower values in 2018
suggested a less productive ecosystem. Higher values for
most of 2019 and 2020 suggested increased productivity.
Peak values in 2019 occurred during April–August, sug-
gesting a more productive autumn/winter than usual. In
contrast, peak values in 2020 and 2021 occurred during
March and October, in agreement with the usual seasonal
maxima.
St Helena Bay on the west coast of South Africa is one of the most productive areas of the Benguela ecosystem and has been
the focus of environmental research and monitoring for several decades (Fig. 1). It is a well-known retention area, with
signicantly elevated plankton biomass compared to other areas o South Africa, and is an important region for many
species such as small pelagic sh, hake, whales, and rock lobster.
Figure 1. Map of sea surface temperature, illustrating cooler waters
typically found inshore and warmer waters oshore, as well as the
location of the St Helena Bay Monitoring Line.
Figure 2. Time series of monthly chlorophyll a (mg m-3) along the
St Helena Bay Monitoring Line between 1997 and 2021.
Author: Lamont T (OC Research)
Oceans and Coasts Annual Science Report, 2021 12
MONITORING PROGRAMMES
St Helena Bay is one of the most productive regions in the
southern Benguela. e St Helena Bay Monitoring Line
(SHBML; Fig. 1) is sampled quarterly to allow for seasonal
microplankton variations to be monitored. Surface mi-
croplankton abundance and composition at each station
were assessed using FlowCam imaging soware. Dominant
microplankton groups included diatoms, dinoagellates,
ciliates and copepod nauplii (Fig. 2). Dierences in abun-
dance were observed in August (winter) and November
(spring) during 2018 and 2019 (Figs. 3 and 4).
Total microplankton abundance was high during spring
2019. Nearshore stations (up to Station 5) were generally
more productive than oshore stations. Diatoms dominat-
ed the microplankton community in both winter and spring
(Figs. 3 and 4). Chaetoceros species (spp.) were the most
abundant diatoms during both years. Copepod nauplii and
ciliates were more abundant at the oshore stations. Both
tintinnids and dinoagellates tended to be more prevalent
during winter, especially at inshore stations. Some dino-
agellates have been shown to negatively aect invertebrate
larvae, which could lead to negative impacts in the ecosys-
tem. Further monitoring is necessary to assess long-term
variability of microplankton in relation to the environment.
Microplankton are a diverse group of phyto- and zooplankton in the size range of 20–200 μm. ey are important ecologi-
cally as they form the foundation of the food web, are instrumental in nutrient and carbon cycling, generate at least 50%
of global oxygen, and facilitate energy ow to higher trophic levels. Due to their high sensitivity and rapid response to
changes in the environment they can be used as indicators of climate change and eutrophication.
Figure 1. Map indicating sampling stations along the St Helena Bay
Monitoring Line (SHBML). Black contours and shading indicate depth.
Figure 4. Relative microplankton abundance (%) along the SHBML from inshore (Station 1) to oshore (Station 12).
Authors: Maduray S, Worship M, Soeker MS (OC Research)
Contributors: Kakora H, Mdazuka Y, Maseti T (OC Research)
9. MICROPLANKTON COMMUNITY STRUCTURE AND DIVERSITY ALONG THE
ST HELENA BAY MONITORING LINE
Figure 2. Microplankton groups observed during SHBML cruises.
Figure 3. Total abundance of microplankton groups in August and
November 2018 and 2019.
13 Oceans and Coasts Annual Science Report, 2021
RESEARCH HIGHLIGHTS
10. DOES LOCALISED COOLING OCCUR AT THE PRINCE EDWARD ISLANDS?
Figure 1. Surface temperature (°C) from underway TSG measurements
around the PEIs during April 2015.
e oceanic environment of the PEIs is highly dynamic due
to frequent meandering of the southern branch of the sub-
Antarctic Front (S-SAF). When the S-SAF is positioned
north of the PEIs, cooler Antarctic surface waters are ob-
served at the islands. In contrast, when the S-SAF occurs
south of the islands, slightly warmer sub-Antarctic surface
waters are present at the PEIs. During April 2015, these
slightly warmer (>6.5°C) sub-Antarctic surface waters were
clearly observed in underway in situ surface temperatures
(Fig. 1) obtained from a ermosalinograph (TSG), as well
as Sea Surface Temperature (SST) maps (Fig. 2).
Overall, there was good agreement between patterns ob-
served from in situ and reanalysis surface temperature.
However, underway measurements during April 2015, and
other cruises, did not reveal any clear signals of localised
cooling (Fig. 1). is is because the underway observations
are usually collected over periods of several weeks. As such,
they are unable to provide a detailed synoptic view of sur-
face temperature variations at and around the PEIs.
As illustrated in Fig. 2, such a synoptic view is necessary
to identify the localised cooling. Daily SST maps revealed
a clear circular region of cooling (<6°C) over the PEI shelf
for a period of 7 days between 17 and 23 April 2015. ese
cooler waters were distinct from the cooler waters south of
the S-SAF (Fig. 2), and likely resulted from the combined
eects of several mechanisms that drive upwelling at the
islands. ese mechanisms include a Taylor Column (see
Report 3), persistent negative wind stress curl (i.e. clock-
wise rotation of the overlying air column which drives
upli of deeper, nutrient-richer water to the surface), and
the interaction of fronts and mesoscale eddies with the
island shelf.
Although localised cooling is expected to occur continu-
ously at the PEIs, daily reanalysis SSTs (Fig. 2) suggest that
it is infrequent. For the ve cruise periods investigated, the
only other observation of clear localised cooling from daily
SST was from 12 to 15 April 2018 (not shown). Surface
expression of this cooling may depend on the position of
the S-SAF. During April 2015, surface cooling was likely
more obvious in comparison to the generally warmer sur-
rounding surface waters because the S-SAF was located
south of the islands (Fig. 2). e S-SAF was also south
of the islands during the 2018 cruise. When the S-SAF is
located north of the PEIs, such surface temperature dif-
ferences are expected to be less obvious, and not easily
detectable from surface observations.
It is necessary to monitor the persistence and intensity of
such cooling at the PEIs, since it likely plays a crucial role
in sustaining enhanced productivity and ecosystem func-
tioning. e limited ability of contemporary in situ and
reanalysis datasets to detect the localised cooling can only
be overcome by collecting continuous in situ observations
throughout the water column across the entire PEI shelf.
e Prince Edward Islands (PEIs) are situated in the direct route of the strong eastward owing Antarctic Circumpolar
Current. Considering the persistence of physical mechanisms that drive upwelling, we expected to observe clear localised
cooling at the PEIs. Previous studies using widely spaced in situ Conductivity-Temperature-Depth (CTD) measurements
and coarse spatial (0.25°) resolution Sea Surface Temperature (SST) showed no evidence of such cooling. In order to deter-
mine if such cooling was evident in higher spatial resolution datasets, we examined underway in situ observations (Fig. 1),
collected during annual re-supply voyages each April from 2015 to 2019, and compared these to higher spatial resolution
(0.05°) daily reanalysis SST (Fig. 2). For the sake of brevity, we only show the 2015 data.
Figure 2. Selected daily reanalysis SST (°C) maps showing localised cool-
ing at the PEIs between 17 and 23 April 2015. e solid black contour in-
dicates a mesoscale eddy located west of the PEIs on 16 and 17 April. e
dashed black contour indicates the S-SAF. e solid white contour shows
the northern branch of the Antarctic Polar Front (APF), while the dotted
white contour shows the middle branch of the APF. Geostrophic vectors
indicate the strength (length of arrow) and direction of current ow.
Authors: Toolsee T (UCT), Jacobs L, van den Berg MA, Lamont T (OC
Research)
Oceans and Coasts Annual Science Report, 2021 14
RESEARCH HIGHLIGHTS
11. DOES LARGE-SCALE CLIMATE VARIABILITY INFLUENCE OCEANOGRAPHY
AROUND THE PRINCE EDWARD ISLANDS?
First, temporal variability of selected oceanographic pa-
rameters at the PEIs was analysed, namely sea surface
temperature (SST), wind speed, wind stress curl (WSC;
rotations of vertical air columns that inuence upwelling),
and surface geostrophic (circulation-driven) and Ekman
(wind-driven) currents. For each parameter, monthly data
at a spatial resolution of 0.25° were extracted and averaged
within a 2° x 2° region around the PEIs (Fig. 1). Wavelet
analysis (a statistical technique commonly used to identify
temporal signals in a dataset) was used to analyse the in-
terannual and decadal-scale variations in each parameter.
Subsequently, the relationship between each parameter and
the climate modes (SAM, SAO and ENSO) was assessed
through lag correlations.
e results suggested possible inuence of large-scale cli-
mate modes on SST, geostrophic current speed and WSC
around the PEIs. SST showed signicant interannual (at
periods of 0.8 and 2.8 years) and decadal-scale (at a period
of 7.5 years) variations (Fig. 2a). WSC showed signicant
interannual variation at 3–4 years and 7–8 years (Fig. 2b).
Geostrophic current speed showed signicant interannual
variability at periods of 1.3 and 4 years (Fig. 2c). In con-
trast, wind speed and Ekman current speed showed the
strongest variability at periods between 1 and 3 months,
with no signicant interannual or decadal-scale variations
(Figs. 2d, e).
Contrary to expectation, correlations between ENSO and
SST were weak (r < 0.3), at lags of 1–2 and 4 years. Correla-
tions between ENSO and geostrophic current speeds at lags
of 1, 2.5–3 and 4 years, and between ENSO and WSC at a
lag of 12 years, were also weak (r < 0.2). Correlations be-
tween the three local oceanographic parameters and both
the SAM and SAO were also negligible.
A possible reason for the lack of clear correlations is that
dierent climate modes interact with each other and es-
sentially moderate each other’s eects on the local oceano-
graphic parameters. For example, ENSO may be operating
to increase SST, while SAM or SAO is acting to decrease
SST at the same time. Other climate modes that are unac-
counted for here, such as the Indian Ocean Dipole or the
Antarctic Circumpolar Wave, could also be inuencing
oceanographic variability at the islands. Moreover, sub-
stantial shorter-term variability may in some cases mask
the longer-term climate-driven signals, which are typically
weaker (Fig. 2). Longer time series may thus be necessary
to identify signicant climate-driven signals, highlighting
the need for continued observations.
e Prince Edward Islands (PEIs) are situated in a region of the Southern Ocean that is frequently aected by the meander-
ing of fronts and passing mesoscale eddies. is creates a dynamic oceanic environment that strongly contributes to the
provision of critical feeding grounds for large populations of top predators that breed on the islands. Large-scale modes
of climate variability, such as the Southern Annular Mode (SAM), the Semi-Annual Oscillation (SAO) and the El-Niño
Southern Oscillation (ENSO) and their teleconnections, are known to inuence much of the Southern Ocean, causing
long-term, decadal and interannual changes in the oceanic environment. A good understanding of the interannual and
longer-term changes in oceanic properties, and the potential impact of climate modes around the PEIs, is essential to bet-
ter understand and predict any potential future ecosystem changes.
Figure 1. Bathymetry (m) around the Prince Edward Islands archipelago
in the Southern Ocean. Long-term mean positions of the southern branch
of the sub-Antarctic Front (S-SAF; dashed black line) and the northern
branch of the Antarctic Polar Front (N-APF; solid red line) are shown. e
black box indicates the 2° x 2° study area within which data were averaged,
and arrows indicate the average ow direction.
Figure 2. Wavelet analysis results for (a) SST (1982–2020), (b) WSC
(1979–2020), (c) Geostrophic current speed (1993–2020), (d) Wind speed
(1979–2020), and (e) Ekman current speed (1993–2020) around the PEIs,
with the 95% condence level indicated (dashed red lines). Interannual
and decadal signals are signicant when the power exceeds that at the
95% condence level. e grey shaded areas indicate where signals are
insignicant due to the limitations of the lengths of the respective time
series.
Authors: Toolsee T (UCT), Lamont T (OC Research)
15 Oceans and Coasts Annual Science Report, 2020
RESEARCH HIGHLIGHTS
12. NEW MOORED OBSERVATIONS REVEAL CONTRASTING OXYGEN SEASONALITIES
ALONG THE SOUTHERN BENGUELA COAST
Figure 1. Satellite chlorophyll a (mg m-3) for 18 February 2019, illustrating
the mooring site (black star).
Figure 2. Time series of (a) temperature, and (b) oxygen at the mooring
near Hondeklip Bay.
Figure 3. (a) Diurnal, and (b) seasonal variation of oxygen and tempera-
ture at the mooring near Hondeklip Bay.
As part of a German/South African collaborative research
project to investigate trophic transfer eciency in the
sBUS, a research cruise was conducted on the German
research vessel RV Meteor during February 2019. A
mooring equipped with a miniDOT sensor (MDO) was
deployed at the position 30.64°S and 17.02°E, southwest of
Hondeklip Bay (Fig. 1). e MDO measured temperature
and DO at a depth of 96 m (ca. 74 m above the sea oor) at
10-minute intervals, for a 20-month period from February
2019 to October 2020 (Fig. 2), when it was recovered dur-
ing a research cruise on the South African research vessel
RS Algoa.
Diurnal variability was evident, with DO reaching maxima
around 9 am, while minima occurred during the late aer-
noon. is implies enhanced downward mixing of warm,
well-ventilated surface waters, favoured by cooling and
reduced stratication in the surface layers during the late
aernoon and at night (Fig. 3). DO also showed rapid and
dramatic short-term decreases of ca. 100 µM over a few
days at a time (Fig. 2), due to oshore transport of oxygen-
depleted waters from the shallower nearshore area along
the coast.
Surprisingly, the seasonal cycle of DO near Hondeklip Bay
was marked by minima during winter and maxima in sum-
mer (Fig. 3). is seasonality is in contrast to DO seasonal-
ity further south in St Helena Bay, where DO minima are
usually observed at the end of the upwelling season, with
maxima occurring in winter. e opposing seasonal DO
cycle, that was observed southwest of Hondeklip Bay
(Fig. 3), was likely caused by the more frequent periods of
enhanced horizontal mixing that transported oxygen-
depleted water from the nearshore region to the mooring
location, and by the breakdown of the upwelling front, dur-
ing winter.
It is important to note that the mooring was deployed for
a relatively short period, and thus we are unable to make
inferences on the reproducibility of this DO seasonal cycle
from one year to the next. is stresses the crucial need
for continuous, long-term, high temporal resolution ob-
servations throughout the sBUS, to adequately capture the
variability of environmental conditions that impact the
surrounding ecosystem.
On the west coast of South Africa, the southern Benguela Upwelling System (sBUS) experiences seasonal wind-driven
upwelling that introduces nutrients to the surface layers, promoting enhanced phytoplankton production and sustaining
a diverse ecosystem (Fig. 1). One of the consequences of such enhanced productivity is the development of an oxygen
minimum zone (OMZ), where dissolved oxygen (DO) is consumed as organic matter decays. In the sBUS, the OMZ is
most pronounced in the bottom waters of St Helena Bay, but also develops elsewhere in nearshore regions along the coast
towards the end of the upwelling season.
Authors: Lamont T (OC Research), Rixen T (Leibniz Centre for Tropical
Marine Research), Lahajnar N (Universität Hamburg)
Oceans and Coasts Annual Science Report, 2020 16
RESEARCH HIGHLIGHTS
13. CURRENT REVERSALS OFF PORT EDWARD ON THE EAST COAST OF
SOUTH AFRICA
In September 2005, bottom mounted Acoustic Doppler
Current Prolers (ADCPs) were deployed at three loca-
tions across the shelf, just south of Port Edward (Fig. 1). e
mooring at P1 was located 3.2 km oshore at 36 m depth,
P2 was 6.3 km oshore at 64 m depth, and P3 was 10.2 km
oshore at a depth of 162 m. Southwestward ow domi-
nated at P1 (89% of the time) with current speeds between
0.33 and 163.38 cm s-1. Similarly, southwestward ow was
observed at P2 (88% of the time) with speeds varying from
0.32 to 188.99 cm s-1 (Fig. 2). Despite the shorter measure-
ment period at P3, there was good agreement with the cur-
rent variability observed further inshore at P1 and P2.
Previous studies demonstrated that the AC was located
around 10 km from the coast, but here we observed that
the AC extended as far inshore as 3.2 km from the coast.
Current speeds in the upper ocean layers at P2 and P3 ex-
ceeded 100 cm s-1 most of the time, indicating the presence
of the AC. At P1, such strong southwesterly ow was only
observed between December 2005 and May 2006, sug-
gesting possible seasonal variation in the intensity and/or
shoreward extent of the AC. However, substantially more
years of observations are required to conrm this.
A total of 11 northeasterly ow events, ranging from 1–15
days in length, were observed during the measurement period. Due to the shorter record at P3, only 4 of the 11
events were captured at the oshore location. During
September and November 2005, as well as January–Feb-
ruary and April 2006, northeasterly ow through the full
water column resulted from cyclonic eddies travelling
southwards along the inshore edge of the AC. e remain-
ing northeasterly ow events were associated with smaller
scale shoreward and oshore movements of the AC, which
resulted in much shorter periods of current reversals. Each
of these current reversals was associated with substantial
short-term variations in bottom temperature, with daily
uctuations of up to 4°C. Contrary to expectation, cor-
relations between bottom temperature and current rever-
sals were poor and not statistically signicant, since both
warming and cooling were observed during the more regu-
lar southwesterly ow, and during the northeasterly current
reversals. is reects the complex temperature response
to current variability in the region, and highlights the need
to extend high temporal resolution sampling to better
capture and understand this variability.
Along the east coast of South Africa, environmental conditions on the narrow continental shelf are strongly inuenced
by the dynamics and variability of the fast, southwestward-owing Agulhas Current (AC). Large solitary AC meanders
consist of cyclonic eddies travelling southwards along the inshore edge of the AC, and are commonly referred to as Natal
Pulses and Durban Eddies. ese eddies are typically associated with inshore current reversals (i.e. northeastward ows
are observed, in contrast to the more regular southwestward ow of the AC), as well as upwelling of nutrient-rich water
which has been shown to stimulate increases in phyto- and zooplankton biomass. Most previous environmental studies
in this region have been based on data that reect conditions averaged temporally (over several days) and spatially (over
larger geographic areas), as observed during research cruises. In this study, we examined historical, high temporal resolu-
tion moored observations to provide a baseline description of currents and bottom temperature variations across the shelf.
Figure 1. Bathymetry o Port Edward on the east coast of South Africa.
Red dots show mooring positions (P1, P2 and P3).
Figure 2. Meridional current speed (cm s-1) at mooring positions (a) P1,
(b) P2, and (c) P3, showing northward (positive) and southward (nega-
tive) ows o Port Edward.
Authors: Louw GS, van den Berg MA, Lamont T (OC Research)
17 Oceans and Coasts Annual Science Report, 2021
RESEARCH HIGHLIGHTS
14. OCEANOGRAPHIC TRIGGERING OF SOUTH AFRICA’S SARDINE RUN
Figure 1. Schematic illustration of sardine migrations culminating in the sardine run. Photo inset: common dolphins, Delphinus capen-
sis, feeding on sardines during the run (credit: Steve Benjamin; www.animalocean.co.za).
First reported in the mid-1800s, numerous explanations
have been proposed for the sardine run, including equa-
torward movement of juvenile sh, feeding migrations
facilitated by habitat expansion, herding of sardines by
predators, and relic spawning behaviour towards an area
that may have been an important nursery during the last
ice age. Given that marine species around South Africa
tend to be subdivided into regional populations associated
with distinct biogeographic provinces, it was previously
thought that east coast sardines represent a separate stock
component that mixes with sardines resident on the south
coast during summer, but then separates from them during
winter to travel toward their east coast spawning grounds.
Although the east coast sardines can be distinguished
from those elsewhere in southern Africa using phenotypic
(physical) characteristics, this may be due to the stress in-
volved in the migration, and traditional genetic analyses
have not supported the existence of separate populations.
Using genomic data of hundreds of South African sardines,
we found that east coast sardines were not unique, but clus-
tered with west coast sardines, and were distinct from those
on the south coast. is new nding suggests that sardines
participating in the run must be migrants that originate
from the cool-temperate Atlantic, and are not actually
well adapted to the warmer, subtropical conditions on the
east coast. Recently, we have identied a completely novel
explanation for the sardine run.
Mesoscale cyclonic eddies along the inshore edge of the
Agulhas Current drive upwelling, which transports cold,
nutrient-rich water from depth onto the shallower shelf
(Fig. 1). When these eddies occur along the southeast coast
during autumn and winter, shelf waters can become tem-
porarily cooler than those further west along the south
coast. is cooling creates conditions that favour west
coast sardines, triggering an aggregation of these migrants
at the northeastern limit of the south coast. Intermittent
upwelling, driven by the interaction of the Agulhas Cur-
rent with the continental slope, likely facilitates further
northward movement of the sardines in cooler water at
100–200 m below the surface. Eventually, the sardines nd
themselves in “hot water” - subtropical water that exceeds
their preferred thermal range, suggesting that the sardine
run does not benet South Africa’s sardine population.
e KwaZulu-Natal sardine run is a mass migration of South Africa’s commercially most important small pelagic sh, the
sardine Sardinops sagax, from warm-temperate waters on the south coast to subtropical waters on the east coast (Fig. 1).
It is one of Earths most spectacular marine migrations and is both ecologically and economically important. e tens to
hundreds of millions of sardines that participate in the run are accompanied by numerous other small pelagic sh, and
together, these swarms attract large numbers of marine predators. e spectacle further supports a temporary artisanal
shery and provides an important source of tourism revenue during the not so busy winter months.
Authors: Teske PR (University of Johannesburg), Lamont T (OC
Research), van der Lingen CD (Fisheries Management, Fisheries Research
and Development)
Oceans and Coasts Annual Science Report, 2021 18
RESEARCH HIGHLIGHTS
ree study areas were selected along the Cape Penin-
sula, at Camps Bay, Mouille Point and ree Anchor Bay.
Each area comprised an “impacted” site near a stormwa-
ter outlet and “control” site ± 200 m away from the outlet.
Sampling took place during two seasons, in summer 2020
and winter 2021. Fauna that were sampled comprised three
dierent feeding groups: lter feeders (mussels), grazers
(sea urchins) and carnivores (whelks). Seawater and sedi-
ment samples were also taken at each site.
A total of 1,362 samples were analysed, with 4,814 MP par-
ticles recorded during the two sampling periods. Samples
from impacted stormwater sites had consistently more
MPs in samples than control sites (Fig. 1). Against our
expectation, we recorded signicantly less MPs in winter
than in summer, indicating that the abundance and build-
up of MPs and other waste products may increase during
the summer period. is may be either because more MPs
are entering the coastal waters during summer, or because
they tend to be ushed out during winter, e.g. during
stormy weather.
Interestingly, the mean concentration of MPs found in
biotic samples from the intertidal zone was signicantly
higher than in the shallow subtidal zone (Fig. 2). is may
be because stormwater outlets discharge runo directly
into the intertidal zone, or because most plastics are buoy-
ant and get washed up on the shore. Mussels contained
more MPs in their so tissue than sea urchins and whelks,
during both seasons. is was not surprising because
mussels are non-selective feeders that digest any particles
depending on their size. Whelks are directly impacted by
15. STORMWATER CONTRIBUTION TO MICROPLASTICS IN COASTAL ZONES AROUND
CAPE TOWN
Figure 1. Mean seasonal abundance of MPs per unit in biotic (so
tissue weight- in grams) and abiotic samples (water per 100 L; sediment per
0.2 kg) across impact and control sites. Error bars represent the 95%
condence interval.
Figure 2. Mean seasonal abundance of MPs per unit biotic (so tissue weight- in grams) and abiotic samples (water per 100 L; sediment per
0.2 kg) found at Camps Bay, Mouille Point and ree Anchor Bay across impact and control sites in the subtidal and intertidal zones. Error bars
represent the 95% condence interval.
the abundance of MPs since they are a key predator of
mussels on rocky shores.
is study showed that stormwater outlets introduce
signicant quantities of MPs into the coastal region, which
are then ingested by coastal organisms. Further research is
needed on MP impacts on the coastal environment, and the
City of Cape Town discharge systems should be improved
to reduce MP pollution.
Due to inadequate waste management and poorly designed drainage systems, urban runo and associated pollutants
including plastics are commonly discharged into shallow coastal waters via stormwater outlets. In South Africa, between
0.09 and 0.25 million tons of marine plastic pollution originates from land-based sources, with Cape Town responsible for
a signicant quantity of this pollution. A pollutant of emerging concern is microplastics (MPs), with rivers and stormwater
outlets shown to be key transporters of these tiny (<5 mm) plastic particles into the coastal environment. is study inves-
tigated the distribution of MPs in the coastal environment, within sediments, seawater and marine organisms, near Cape
Town. Seasonal dierences were also assessed, in particular whether MPs are more abundant during the winter rainfall
season when stormwater outlets discharge more water, compared to the dry summer season.
Authors: Ariefdien R (OC Research), Sparks C (CPUT), Pfa M (UCT)
19 Oceans and Coasts Annual Science Report, 2021
RESEARCH HIGHLIGHTS
16. DNA METABARCODING OF MARINE ZOOPLANKTON IN SOUTH AFRICA – HOW
GOOD IS THE REFERENCE DATABASE?
Using traditional microscope analysis to identify zooplank-
ton specimens is labour-intensive and time-consuming
because of their small size, large numbers, high diversity
and community complexity. Over the past two decades,
DNA barcoding and online reference databases such as the
Barcode of Life Data Systems (BOLD) and GenBank have
revolutionised species identication and discovery. ese
DNA barcodes allow for distinction between visually simi-
lar species and identication of species irrespective of life
stage, and reduce researcher bias through the use of stand-
ard online reference databases.
Metabarcoding is the application of DNA barcoding to tax-
onomically complex samples that contain more than one
organism or species. It uses the same reference databases as
DNA barcoding but allows for the simultaneous identica-
tion of multiple taxa by using high-throughput sequencing
methods. We investigated the availability of DNA reference
barcodes for marine zooplankton expected to occur in the
coastal waters of South Africa (SA) as an initial step toward
developing a regional metabarcoding protocol.
Exploration of zooplankton DNA records on BOLD re-
vealed proportionally fewer representative barcode records
per taxon from SA compared to those available globally
(Fig. 1). Ray-nned sh (Actinopterygii) were the most
comprehensively sampled aquatic group globally (64% of
species barcoded) and in SA (48%), with an 89% success
rate in assigning immature specimens (including eggs and
larvae) to species level. For crustaceans, barcode records
were available for only 18% of known species globally and
6% in SA. is study has highlighted two clear trends in
SA. Firstly, for nearly all groups, there were proportion-
ally fewer records of known species compared to global
datasets. Secondly, barcode records were dominated by
meroplanktonic taxa with large benthic or pelagic adult
stages that are of value to commercial or recreational sh-
eries. Despite comprising the bulk of zooplankton diversity,
including most potential indicator species, holoplanktonic
taxa were underrepresented.
ere is thus an urgent need to expand and validate bar-
code reference databases of all key zooplankton classes
to improve the resolution and representivity of metabar-
coding outputs. Metabarcoding of marine zooplankton
has now been successfully applied in SA as a pilot pro-
ject, and the methodology is poised to shi research em-
phasis from individual species to assemblages, to facilitate
high-resolution monitoring of zooplankton biodiversity in
pelagic ecosystems, and to accelerate discovery of new
species.
Zooplankton play an essential role in marine pelagic food webs, through biogeochemical cycling and energy transfer to
higher trophic levels. Zooplankton include “full-time” plankton (holoplankton) that spend their whole lives as part of
the plankton, and “part-time” plankton (meroplankton) consisting of the early life stages of crustacean, sh and mollusc
species. e rapid response of zooplankton to environmental changes makes them ideal indicators of ecosystem health
and biodiversity.
Figure 1. e relative percentages of DNA barcode records available for selected marine zooplankton taxa, globally (pale bars) and for South Africa (dark
bars). e numbers of species known locally/globally are indicated next to the bars. Adapted from Singh et al. 2021. African Journal of Marine Science
43: 147–159.
Authors: Singh S, Groeneveld JC (ORI), Huggett J (OC Research),
Naidoo D (ORI), Cedras R (UWC), Willows-Munro S (UKZN)
Cephalochordata
Ostracoda 120/ 1300
4/ 173
3/ 12
35/ 859 58/ 1106
80/ 81 176/ 2300
1/ 33
457/ 3500
119/ 2139
2/ 400 16/ 350 2262/ 60000
650/ 9200
29/ 650
270/ 5700
11/ 166
38/ 1045
11/ 108
31/ 550
346/ 6000 28/ 121
760/ 8432 2/ 14
91/ 1859
19/ 635
59/ 940 122/ 1430
174/ 7500 2/ 36
10/ 175
101/ 1340
750/ 18000
454/ 7000 69/ 1220
429/ 3220
3/ 8
49/ 90 5/ 620
98/ 1593
277/ 11000
4/ 41
Arguloida
Cumacea
Leptostraca
Isopoda
Mystacocarida
Stomatopoda
Mysida
Pycnogonida
Decapoda
Amphipoda
Cirripedia
Copepoda
Euphausiacea
Cladocera
Thaliacea
Ascidiacea
Actinopterygii
Sarcopterygii
Hydrozoa
Anthozoa
Cubozoa
Scyphozoa
Ophiuroidea
Asteroidea
Crinoidea
Echinoidea
Holothuroidea
Aplacophora
Scaphopoda
Gastropoda
Bivalvia
Polyplacophora
Cephalopoda
Bryozoa
Ctenophora
Tardigrada
Hemichordata
Brachiopoda
Porifera
Chaetognatha
Polychaeta
Phoronida
Percentage of Species with a DNA Barcode Record
025 50 75 100
Arthropoda
Chordata
Cnidaria
Echinodermata
Mollusca
Other Phyla
195/ 800
2200/ 30383
Oceans and Coasts Annual Science Report, 2021 20
RESEARCH HIGHLIGHTS
17. MACROFAUNAL COMMUNITY OF A LARGE TEMPORARILY CLOSED ESTUARY
DURING PROLONGED MOUTH CLOSURE
e concentration of salts, referred to as salinity and
typically measured in parts per thousand (ppt) of salts in
water, is an important physico-chemical variable respon-
sible for structuring spatial patterns of physical/biogeo-
chemical properties and biota. Strong horizontal salinity
gradients may be present in the estuary when the mouth
is open, but mesohaline (salinity = 5.0–17.9 ppt) or oligo-
haline (0.5–4.9 ppt) conditions oen replace these gradi-
ents during prolonged mouth closure. Temporarily closed
systems are usually dominated by both marine and estua-
rine biota, whereas during oligohaline conditions, fresh-
water and estuarine biota dominate. Previous studies in
temporarily closed estuaries (e.g. uMdloti in KwaZulu-
Natal) have shown that polychaetes tend to dominate when
the system is closed, while oligochaetes dominate aer
prolonged closure and diluted salinity.
e Groot Brak estuary in the Western Cape is a large
temporarily closed estuary that has undergone substan-
tial modications due to human impacts and remains
closed for 80% of the time. Under natural conditions, the
estuary mouth would alternate between being open or
closed throughout the year, and would undergo various
physico-chemical states with corresponding shis in es-
tuarine conditions and biota. e likelihood of mouth
closure under natural conditions would peak in January–
February and June. However, under present-day condi-
tions, the mouth of the estuary shis between a primarily
marine dominated state (typically October–March), and a
closed mouth state (typically April–September). To deter-
mine the response of benthic macrofauna to the present-
day estuarine regime, 11 locations were sampled in May
2019 (autumn) and January 2020 (summer) (Fig. 1).
e mouth of the estuary was closed during both sam-
pling periods. Polyhaline conditions (18–29 ppt) prevailed
throughout the estuary during autumn (mean salinity =
22.60 ± 1.86 ppt), with mesohaline conditions in summer
(15.90 ± 3.31 ppt). ere was no salinity gradient observed
in the estuary, which is typical under prolonged mouth
closure conditions.
A total of 24 taxa were recorded during the study
period, with the polychaetes Capitella capitata, Aquilaspio
sexoculata, Ceratonereis erythraeensis, and the amphipod
Grandidierella lutosa dominating. Previous studies found
that the macrofaunal community was dominated by the
mudprawn Upogebia africana, the sandprawn, Kraussilli-
chirus kraussi and the bivalve, Loripes clausus. U. africana
and K. kraussi were not recorded in the present study, while
L. clausus was present in low numbers. Lower numbers of
U. africana may be due to prolonged mouth closure since
their recruitment is dependent on a marine phase. Lower
numbers (or observations) of K. kraussi can be attributed
to the cessation of reproduction at salinities <16 ppt, with
deeper burrowing to their preferred salinity.
e results of the study have shown the changes that have
occurred in the macrofaunal benthic community due to
prolonged closure of the estuary mouth. Additional sam-
pling during an open phase will determine if community
structure and composition returns to the state it was in
prior to the prolonged mouth closure, with mud- and sand-
prawns dominating.
South Africa is a water scarce country with an increasing population and a growing demand for freshwater resources. is
high demand for water negatively impacts ecosystems such as estuaries, where a reduction in ow may result in more fre-
quent and prolonged closure of the estuary mouth. Prolonged closure may alter water chemistry and inhibit recruitment
and migration of sh and invertebrates between the estuary and the sea.
Figure 1. Map showing the locations of the 11 sampling sites.
Author: Nhleko J (OC Research)
Contributors: Lamberth S, Erasmus C, Williamson C (Fisheries
Research and Development), Bebe L, Mushanganyisi K, Chavalala T (OC
Research)
Figure 2. Dominant benthic macrofauna of the Groot Brak estuary during
autumn (A) and summer (B).
Capitella capitata Ceratonereis erythraeensis Aquilaspio sexoculata
Grandidierella lutosa Brachidontes virgiliae Assiminea sp
Corophium acherusicum Nassarius kraussianus Other
Capitella capitata Aquilaspio sexoculata Ceratonereis erythraeensis
Grandidierella lutosa Desdemona ornata Nassarius kraussianus
Chironomidae Corophium acherusicum Other
(a)
(b)
Aquilaspio sexoculata Grandidierella lutosa
Brachidontes virgiliae Desdemona ornata
Capitella capitata Assiminea sp
Ceratonereis erythraeensis Chironomidae
Corophium acherusicum Other
Nassarius krassianus
A B
11
10
9
8
7
6
5
4
2
1
Estuary mouth
21 Oceans and Coasts Annual Science Report, 2021
RESEARCH HIGHLIGHTS
18. A COMPROMISED IMMUNE SYSTEM: THE CAPE URCHIN IN A RAPIDLY
ACIDIFYING WORLD
e traditional approach to this is short-term exposure
experiments. ese generally reveal that echinoderms,
particularly the adult life stages, are relatively robust to
OA. However, longer term, or chronic exposure to low pH
conditions (i.e. hypercapnia) is required to reveal the true
eco-physiological impacts.
e Cape urchin, Parechinus angulosus, is a widely dis-
tributed, keystone species in South Africa. In sea urchins,
cellular immune defences are shared amongst the coelomo-
cytes, a population of cells occurring in the coelomic uid
(blood equivalent). Amongst these are red spherule cells
which have antiseptic properties used to ght o harmful
bacteria. To examine if chronic exposure to hypercapnia
aects the immune response of P. an g u l o s u s , we exposed
adult urchins to hypercapnic (pH = 7.4) and normocap-
nic (pH = 8) conditions for ca. six months (Fig. 1). ere-
aer, counts of the total coelomocytes, and in particu-
lar, red spherule cells, were signicantly lower per ml of
coleomic uid in hypercapnic versus normocapnic urchins
(Fig. 2). ese results indicate that, in increasingly acidify-
ing waters, the immune system of the Cape urchin may be
compromised.
e signicant decline in red spherule cell numbers for
hypercapnic urchins further indicates that these organisms
may not easily recover from bacterial infections, or inju-
ries including spine loss (Fig. 3). us our study provides
evidence that adult life stages of echinoderms are not
exempt from stressors, and chronic exposure is crucial in
revealing the real, long-term, eco-physiological eects of
OA.
At the end of this century, a two-fold increase in acidity is
expected in surface ocean waters. With such unprecedent-
ed rates of change, coupled with the cumulative impacts of
other anthropogenic stressors, marine taxa such as echino-
derms will likely face considerable challenges. A signicant
decline in P. a n g u l o s u s populations will most likely have
a knock-on eect on benthic communities, including for
species that are dependent on urchins such as rock lob-
sters (as prey) and juvenile abalone (for refuge). Kelp bed
ecosystems are also likely to be altered if these important
grazers are reduced. Experimental studies such as this one
that enhance our understanding of how organisms respond
to environmental change can inform adaptive management
and conservation eorts.
High CO2 emissions continue to lower the pH of the world’s oceans. e result is Ocean Acidication (OA), a phenomenon
which negatively impacts marine life and associated ecosystems. An example of marine organisms that are potentially
negatively impacted by OA are echinoderms, such as sea urchins, sea stars and sea cucumbers. ese benthic organ-
isms are calciers that rely on calcium carbonate (CaCO3) to construct their shells and skeletons. Because OA is causing
many parts of the ocean to become under-saturated in CaCO3, these organisms are particularly relevant to studying the
physiological impacts of OA.
Figure 1. An illustration of the long-term experimental set-up. Seawater
owed from storage tanks into mixing tanks where pH was manipulated,
before owing into culture tanks housing urchins in hypercapnic (grey)
and normocapnic (white) conditions.
Figure 2. Total red spherule cell counts (per/ml of coelomic uid) aer
exposure to normocapnic and hypercapnic conditions (inserted
image adapted from Coates et al. 2018. Journal of Innate Immunity 10:
119–130).
Figure 3. Hardy normocapnic (N) urchins (top), versus spine loss and
bacterial infections in hypercapnic (H) urchins (bottom).
Author: Haupt T (OC Research)
Contributors: Auerswald L, Macy B (Fisheries Research & Development),
Reismann T (University of Rostock, Germany), Forgus J (SANBI), Fester
M (UWC)
Normocapnic Hypercapnic
Treatment
Red spherule cells (per/ml)
70000
60000
50000
40000
30000
20000
10000
0
Oceans and Coasts Annual Science Report, 2021 22
RESEARCH HIGHLIGHTS
Site
During
0,0
0,3
0,6
0,9
1,2
Impact Control Impact Control Impact Control
Seal response (mean %)
Site
Nourishing Interacting Alert
Before After
During
34
696
4
712
0
40
23
745
400
43
586
51
14
497
1
362
15
0
C
Experimental group
Site
During
0,0
0,3
0,6
0,9
1,2
Impact Control Impact Control Impact Control
Seal response (mean %)
Site
Nourishing Interacting Alert
Before After
During
34
696
4
712
0
40
23
745
400
43
586
51
14
497
1
362
15
0
C
Experimental group
Site
During AfterBefore
0
0,2
0,4
0,6
0,8
Impact Control Impact Control Impact Control
Seal response (mean %)
Site
Hauling out Entering water
During After
92
179
106
438
263
344
199
233
121
256
110
Before
288
B
1,2
Seal response (mean %)
Nourishing Interacting Alert
696
712
745
586
497
362
C
Site
During AfterBefore
0
0,2
0,4
0,6
0,8
Impact Control Impact Control Impact Control
Seal response (mean %)
Site
Hauling out Entering water
During After
92
179
106
438
263
344
199
233
121
256
110
Before
288
B
1,2
Seal response (mean %)
Nourishing Interacting Alert
696
712
745
586
497
362
C
0,0
20,0
40,0
60,0
80,0
Impact Control Impact Control Impact Control
Seal response (mean %)
Site
Lying down Sitting Moving Grooming
During AfterBefore
54,019
17,795
2,248
4,072
50,236
16,364
2,243
3,959
53,485
18,608
2,406
4,422
51,566
16,522
2,117
3,819
38,368
13,022
1,725
2,761
38,841
12,051
1,370
2,516
0,6
0,8
Seal response (mean %)
Hauling out Entering water
438
344
233
256
B
A
0,0
20,0
40,0
60,0
80,0
Impact Control Impact Control Impact Control
Seal response (mean %)
Site
Lying down Sitting Moving Grooming
During AfterBefore
54,019
17,795
2,248
4,072
50,236
16,364
2,243
3,959
53,485
18,608
2,406
4,422
51,566
16,522
2,117
3,819
38,368
13,022
1,725
2,761
38,841
12,051
1,370
2,516
0,6
0,8
Seal response (mean %)
Hauling out Entering water
438
344
233
256
B
A
19. BEHAVIOURAL RESPONSE OF CAPE FUR SEALS TO SWIM-WITH-SEAL TOURISM
ACTIVITIES IN THE ROBBERG MPA
Swim-with-seal (SWS) ventures have become a popular tourist activity in the Western Cape. e activity commenced in
2009 at the seal breeding colony in the Robberg Marine Protected Area (MPA) in Plettenberg Bay (Fig. 1). Currently, little
is known about the potential impacts of SWS, especially the associated disturbance on the well-being of the seal colony.
is is cause for concern given the growth in popularity of the activity and increasing numbers of operators, including
those in the Cape Town area. We therefore initiated a behavioural study at the Robberg MPA to assess impacts of SWS
tourism on the seal colony, with a goal of providing advice for the management of this activity.
Behavioural observations were conducted from the cli
tops above the seal colony. A vantage point was chosen that
allowed for observations of both an impact site (a part of the
colony adjacent to SWS activity) and a control site (separate
from SWS activities). In terms of study design, sequences
of photographs of the seal colony were taken before the ar-
rival of the tourist boat, during the period of SWS activity,
and aer the departure of the boat. is was done for both
the impact and control sites. e images, which provided a
digital record of seal activity in the colony, were augmented
by video recordings and visual observations using binocu-
lars. Data were also collected on the numbers of swimmers
in the water, their behaviour, and the environmental condi-
tions (air and seawater temperatures, wind speed/ direction
and visibility). Data collection was hampered by Covid-19
restrictions, but it was possible to collect data for 55 tour-
ist visits during 39 days from November 2020 to October
2021.
Seals in the images were counted and classied according
to four behavioural activities (lying down, sitting, moving
and grooming). No change was apparent in the proportions
of animals for each of these behavioural categories in the
impact area relative to the control area (Fig. 2A). Similarly,
no dierence was apparent between the impact and con-
trol sites with respect to the relative numbers of animals
entering versus exiting (hauling out of) the water (Fig. 2B).
e only apparent dierence between the impact and con-
trol sites was in the relative numbers of animals showing
varying levels of alertness (Fig. 2C). e results therefore
indicate that while there is awareness by seals in the colony
of the SWS activities occurring nearby, overall behaviour
patterns do not seem to be impacted. ese results are
preliminary and a more detailed analysis shall be conduct-
ed, taking into account the environmental data collected
during sessions. Repeated-measures analyses will be per-
formed to ensure an adequate comparison of impact and
control sites within each observation session.
Figure 1. Swim-with-seal activities as viewed from the observation point
above the Robberg seal colony.
Authors: Basson R, Kirkman SP (OC Research), Findlay KP (CPUT)
Contributors: Malwela N, Ariefdien R, Mushanganyisi K, Swart L (OC
Research), Abrahams C, Vos S (CapeNature), Penry G (NVT)
Figure 2. Comparison of seal responses for dierent behavioural
categories between impact and control areas, before, during and aer SWS
activities. Data labels show the total number of counted seals represented
by each data point with standard error bars.
Seal response (mean %)
Experimental group
80.0
54.019
17.795
16.364
16.522
13.022
12.051
18.608
2.248
2.243
2.406
2.117
1.725
1.370
2.516
2.761
3.819
4.072
3.959
4.422
50.236
53.485
51.566
38.368
38.841
A
B
C
60.0
40.0
20.0
0.0
0.8
0.6
0.4
0.2
1.2
0.9
0.6
0.3
0.0
0
23 Oceans and Coasts Annual Science Report, 2021
RESEARCH HIGHLIGHTS
20. ECOLOGICAL EFFECTIVENESS OF SOUTH AFRICAS MARINE PROTECTED AREAS
Figure 1. Numbers of published studies per MPA assessing their ecologi-
cal eectiveness for ve categories of taxa or life forms, namely sh, inver-
tebrates, larval phases, ora and iconic species. e bars are indicative of
the number of relevant studies per site. Only MPAs declared before 2019
are included, some of which were modied in 2019 (*expansion of for-
mer Bird Island MPA; ** expansion of previous Aliwal Shoal MPA; *** the
expansion and combination of former St Lucia and Maputaland MPAs).
Figure 2. Dierent types of eects that are applicable to assessing ecologi-
cal eectiveness of MPAs, and the numbers of studies in which they have
been evaluated.
Figure 3. Literature review results showing published MPA eects at the
level of individual species. Eects are categorised as positive (✓), weak,
neutral, insignicant or inconclusive (O), or negative (x). e number
of cases represents the sum of species per MPA for which eects were
tested. Abundance is inclusive of density, catch rate, biomass and percent-
age cover; Size represents either length or mass; Fec. = fecundity; Repr. =
reproductive output.
A review of relevant scientic literature was conducted to
assess the ecological eectiveness of South Africa’s MPAs.
Most studies were found to focus on the protection and
recovery of shery-targeted linesh species, or harvested
intertidal resources, highlighting a shortage of studies
on non-targeted species, including what are regarded as
“iconic” taxa, such as turtles. Specically, most of the studies
focused on sh or shark species (46%) followed by inver-
tebrates (26%), with the rest covering ora, iconic spe-
cies, or larval phases. Also evident was that most research
has been limited to a subset of the larger coastal MPAs
(Fig. 1), such as Tsitsikamma (mainly linesh), Dwesa-
Cwebe (mainly invertebrates), De Hoop, Goukamma and
iSimangaliso (mainly sh).
Most of the studies (ca. 60%) focused on changes in
population parameters in species or communities, associ-
ated with protection. Fewer studies looked at ecosystem
eects, the spillover of adults or larval export from MPAs
to adjacent areas, or benets to sheries in adjacent areas
(Fig. 2).
e majority of studies focusing on population parameters
of species recorded benecial ecological eects. ese were
detectable as increases in parameters (abundance, bio-
mass, size or reproductive output of species), inside versus
outside of MPAs, or before and aer MPA proclamation
(Fig. 3). Results for comparison of community-level
parameters were more ambiguous, but most studies again
showed positive ecological eects (not shown here).
e results denitively addressed the main objective of
this study, namely to assess whether South Africa’s MPAs
provide eective protection for its marine species and
species assemblages. Further research and monitoring to
achieve evaluations of MPA ecological eectiveness are
recommended, with enhanced focus on neglected MPAs
and species (Fig. 2), and on lesser studied eects (Fig. 3).
ere has been considerable work in recent years on improving the extent and representivity of marine protection in South
Africa, but less focus on evaluating the eectiveness of Marine Protected Areas (MPAs). Given the growing demand for
marine resources from various sectors, for which activities may be restricted or prohibited in MPAs, there is an increasing
need to justify the existence of MPAs by providing evidence of their eectiveness.
Author: Kirkman SP (OC Research)
Contributors: Pfa M, Samaai T, Williams L (OC Research), Mann BQ,
Mann-Lang VB (SAAMBR), Sink KJ, van der Bank MG (SANBI), Adams
R (WWF-SA), Livingstone T-C (EKZNW), Branch GM (UCT)
0
2
4
6
8
10
12
14
16
18
20
Rocherpan
Malgas Island
Marcus Island
Jutten Island
Langebaan Lagoon
Sixteen Mile Beach
Table Mountain
Helderberg
Betty's Bay
Walker Bay
De Hoop
Stilbaai
Goukamma
Robberg
Tsitsikamma
Sardinia Bay
*Addo Elephant
Amatole
Dwesa-Cwebe
Hluleka
Pondoland
Trafalgar
**Aliwal Shoal
***iSimangaliso
No. of published studies
Site
Iconic species
Flora
Larval phases
Invertebrates
Fish
No. of published studies
20
Site
18
16
14
12
10
8
6
4
2
0
0
10
20
30
40
50
Differences
in population
paramaters
Ecosystem
effects
Spillover of
adults
Larval
export
Benefits for
adjacent
fisheries
No. of studies
Type of effect
No. of studies
Type of eect
50
40
30
20
10
0
0
10
20
30
40
50
Abundance
Size
Age
Growth rate
Condition
Fec. / Repr.
Survivorship
Foraging effort
No. of cases
Parameter
...
0
10
20
30
40
50
Abundance
Size
Age
Growth rate
Condition
Fec. / Repr.
Survivorship
Foraging effort
No. of cases
Parameter
...
No. of cases
Parameter
50
40
30
20
10
0
Oceans and Coasts Annual Science Report, 2021 24
RESEARCH HIGHLIGHTS
21. THE POLYCENTRIC GOVERNANCE APPROACH OF THE BENGUELA CURRENT
CONVENTION
Figure 1. Simplied schematic illustration of governance architecture around the BCC (adapted from Naidoo et al. 2021). Note that not all global and
regional forums and their linkages are shown. WG = working group. EBSAs = Ecologically and Biologically Signicant Areas; MSP = Marine Spatial
Planning; SEAFO = South East Atlantic Fisheries Organisation; CCAMLR = Convention for the Conservation of Antarctic Marine Living Resources
CBD = Convention on Biological Diversity; UNFCCC = United Nations Framework Convention on Climate Change; IMO = International Maritime
Organisation.
e governance approach of the Benguela Current
Commission (BCC), an intergovernmental organisation
that implements the Benguela Current Convention (mem-
ber states include Angola, Namibia and South Africa), was
investigated to evaluate its support to eective transbound-
ary management. e working arrangements of the BCC
are considered to be polycentric to some extent (i.e. to
have multiple centres of authority, with some coordination;
Fig. 1).
Higher levels of polycentric governance are favoured for
regional ocean governance mechanisms because they
promote multi-directional integration across authorities,
towards a common goal. In the case of the BCC, this com-
mon goal is the implementation of the ecosystem-based
approach to ocean governance, nationally, regionally and
globally. Specialist Working Groups (WGs), where each
member state is represented through national departments
and agencies across ocean sectors, are key to the success of
the polycentric approach. e function of WGs is to bring
together experts from the member states to contribute
towards regional information sharing and planning.
e BCC would be able to achieve more integrated and
eective polycentric governance through dening and
implementing coherent transboundary governance objec-
tives and interventions. Actions must be cross-sectoral and
co-designed across the active user sectors and member
states. Water quality standards, for example, can be dened
as common precautionary guidelines for active economic
sectors and management authorities, in each of the mem-
ber states.
e concept of Large Marine Ecosystems (LMEs) was developed to provide a uniform method to sub-divide the global
ocean into functional units. e sixty-six recognised LMEs across the global ocean have been dened according to their
physical and biological characteristics, and many of them span the jurisdictions of multiple countries. South Africa shares
and participates in the governance of three LMEs, namely the Benguela Current LME on the west coast (BCLME), the
Agulhas Current LME on the east coast, and the Antarctic LME in the Southern Ocean. To achieve eective ocean govern-
ance of these LMEs, participating governments and their various ocean sector departments are required to integrate their
work, both at national and regional levels.
Author: Naidoo AD (OC Research)
Contributors: Hamukuaya H (Karneol Management Services), Hara M
(Institute for Poverty, Land and Agrarian Studies, University of the West-
ern Cape), Mngxe Y (OC Specialist Monitoring), Raakjær J (Centre for
Blue Governance, Department of Planning, Aalborg University)
Further reading: Naidoo et al. 2021. Frontiers in Marine Science 8:703451
WG
Small
Pelagic
Fisheries
WG
Demersal
Fisheries
WG
Top
Predators
WG
Environmental
monitoring
WG
Science
Logistics
WG
EBSAs
Global Forums
CBD, UNFCCC, IMO, etc
Regional Structures
BCC; SEAFO; CCAMLR, etc
National Member States
National Ministries
Representatives to the various BCC Working Groups & Structu res
WG
MSP
WG
Climate
Change
WG
Data &
Information
WG
Training
Capacity
Development
25 Oceans and Coasts Annual Science Report, 2020
RESEARCH HIGHLIGHTS
22. INTERACTIONS BETWEEN CAPE FUR SEALS AND CAPE GANNETS AT MALGAS
ISLAND - A NEED FOR URGENT MANAGEMENT INTERVENTION
In the last two decades, a shortage of forage sh o the west
coast of South Africa has been evident in the annual diet
samples of Cape gannets at Malgas Island. Seals, which
have considerable dietary overlap with the gannets, are
also likely to have been aected by the same prey shortages.
An increase in predation of gannets earlier this century,
particularly of edglings in the waters around the island,
was therefore considered to be a likely consequence of the
reduced availability of the seals’ normal prey. As a man-
agement intervention to address the predation, culling of
“rogue” seals commenced, successfully reducing seal-gan-
net predation. Unfortunately, the eort was not sustained,
and predation by Cape fur seals on Cape gannets has be-
come a regular occurrence in the waters around the island.
More recently (since 2015), a new phenomenon has been
detected on Malgas Island, whereby seals are observed pre-
dating on Cape gannets on their nests within the colony
(Fig. 1). As a result, the extent of the colony has contracted
considerably since 2016 (Fig 2; Table 1). Birds nesting on
the periphery of the colony are more exposed to predation,
thus breeding success will be higher in the interior of the
colony. e change to the colony shape is therefore most
likely a behavioral adaptation, with experienced birds se-
lecting nest sites that minimise exposure to predation.
Interestingly, the average distance between neighbour-
ing nests has remained relatively constant. is is because
aggressive interactions with neighbours also inuences
breeding success, therefore nests are spaced to minimise
these interactions.
With predation now occurring both on land and in the sur-
rounding waters at Malgas Island, the impact of Cape fur
seals on this Endangered species is greater than ever. Only
three breeding colonies of gannets exist in South Africa and
urgent management intervention is therefore vital.
e Cape gannet Morus capensis is an iconic seabird species, endemic to southern Africa and classied as Endangered in
terms of IUCN Red List criteria. All the extant breeding colonies of this species face various threats, and at Malgas Island
in the Western Cape, predation by Cape fur seals is one of the major threats currently. It is likely that historically these
species bred in close proximity to each other on the same coastal islands, before severe modication of the islands due to
unregulated seal harvesting, and exploitation of seabird eggs and guano. Seals were prevented by humans from hauling
and settling on the islands while they were being managed for exploitation of the seabird products. However, since these
activities ceased in the 1980s, seals are now able to haul out at such locations without being disturbed.
Figure 1. A collage of dead Cape gannets with various kill marks as a
result of seal predation at Malgas Island, December 2021. Authors: Masotla MJ, Dyer BM, Visagie L, Crawford RJM, Makhado AB
(OC Research)
Year Colony Area Area lost to seal
(ha) predation
2016 1.07 0.01
2017 0.82 0.25
2018 0.82 0.0
2019 0.7 0.012
Table 1. Details of area reductions at the Malgas Island Cape gannet
colony, 2016–2019. Note that in each year there is recruitment of breed-
ing birds.
Recommendations
e islands should be manned, at least during the Cape
gannet breeding season, if not for the whole year. Seals
must be discouraged from hauling out or settling on the
immediate islands, and the culling of “rogue” seals both on
the island and in the surrounding waters is recommended.
e use of streamers or scarecrows erected at the seal entry
point to the gannet colony should be assessed as a possible
mechanism to deter seals from approaching nesting birds.
Figure 2. A visual presentation from above of how the Cape gannet breed-
ing colony at Malgas Island has changed in shape and extent between 2016
(le) and 2019 (right).
Oceans and Coasts Annual Science Report, 2020 26
RESEARCH HIGHLIGHTS
Figure 2. Distribution of long-nned pilot whales. e number of sight-
ings is indicated at each location.
Until recently, information on G. melas’ distribution was
based only on stranding records. However, dedicated eort
by DFFE scientists and valuable reporting by water users
(e.g. divers) has generated new records of live encounters
oshore of south and west South Africa. Observed group
sizes ranged from 8–100 individuals (Fig. 2).
In August 2021, two adult G. melas individuals were tagged
opportunistically with Spot 5 range tags in an Impact
Minimally Percutaneous External-electronics Transmitter
(LIMPET) conguration. e tags (767 and 772) were con-
gured to transmit daily and were deployed remotely, using
a crossbow, into the dorsal n and the base of the dorsal n.
ey attached to the n by means of titanium darts with
backward-facing petals. Tags 767 and 772 transmitted for
12 and 15 days respectively, resulting in a total of 62 and
74 transmitted positions (Fig. 3). During the study, 767’s
movements were within latitudes 33–36°S, from oshore
of Cape Columbine to the Agulhas Bank. e move-
ment patterns of 772 were within latitudes 32.8–34.8°S,
corresponding with areas oshore of Cape Columbine
and southwest of Cape Point. Movements were conned
to the edge of the continental shelf, with 767 travelling
1,460 km in total, and 772 travelling 941 km.
A time-varying move persistence (mp) model was used to
objectively identify changes in movement patterns of the
two animals along a continuum (index 0–1). Higher index
values infer travelling, while lower values indicate area-
restricted-search behaviour, which is associated with feed-
ing (Fig. 3). Model output indicated that G. melas conduct-
ed area restricted searches mostly along the continental
shelf edge, with concentrations in canyon areas, especially
around the Cape Point Valley, and to a lesser degree around
the Cape Canyon.
e waters above underwater canyons are known to be
highly productive and are targeted by commercial sheries
in South Africa. Although no squid shery operates along
this portion of the continental shelf, the association of these
two G. melas individuals with the underwater canyons
regions suggests that there could be target prey in these
areas. Surveys should be undertaken in these areas to
assess the abundance and distribution of squid and other
potential prey species of these oceanic cetaceans.
Two species of pilot whale, the short-nned Globicephala macrorhynchus and long-nned pilot whale G. melas (Fig. 1),
occur in South African waters. e former species is thought to be conned to the southwest Indian Ocean whereas G.
melas inhabits the southeast Atlantic Ocean, where it prefers deep waters along the continental shelf, and tends to follow
prey inshore. Historical data from stranded animals indicate that the diet of G. melas is dominated by the oceanic squid
Todarodes angolensis. Stomach contents of recently stranded animals were almost exclusively full of yet to be identied
squid beaks, which tend to be retained in the gut such that other consumed prey (e.g. sh) are underrepresented in the
stomach contents of stranded animals.
Figure 1. Long-nned pilot whale with characteristic light streaks
behind the eye and dorsal n.
Figure 3. Estimated move persistence by both G. melas individuals along
the satellite tracks.
Authors: Seakamela SM, Kotze PHG, McCue SA (OC Research),
Benjamin S (Animal Ocean)
Acknowledgements: We thank Barry Stringer of Hout Bay Charters for
providing a boat for the tagging trip. Tegan Carpenter-Kling is acknowl-
edged for troubleshooting the model in R-Studio
23. THE FIRST SATELLITE TRACKING OF MOVEMENTS OF LONG-FINNED PILOT
WHALES IN SOUTH AFRICA
27 Oceans and Coasts Annual Science Report, 2021
RESEARCH HIGHLIGHTS
In the Benguela ecosystem, the African penguin Spheniscus
demersus, is one of four endemic seabird species that is cur-
rently classied as Endangered in terms of IUCN criteria.
Each of these species is known to compete with sheries for
prey, and prey depletion is recognised as a major driver of
recent declines in their populations, although other threats
have been noted. Seabirds are central-place feeders during
breeding, meaning that they must regularly return to their
breeding colony, and as such they have limited foraging
ranges. e foraging range of the African penguin during
breeding is especially limited because it must swim to nd
food, unlike volant (ying) seabirds. ey are therefore
particularly vulnerable to impacts of localised exploitation
on prey availability around their breeding colonies.
African penguins depend mainly on energy-rich small pe-
lagic sh, especially sardine Sardinops sagax and anchovy
Engraulis encrasicolus, for food. Both abundance and qual-
ity of prey are important in inuencing the penguins’ pop-
ulation dynamics. However, small pelagic prey has under-
gone both distributional shis and declines in biomass in
the last two decades. e size of the African penguin breed-
ing population appears to be highly related to the chang-
ing biomass of small pelagic prey, and has declined corre-
spondingly (Fig. 1). As the penguin population declines,
Figure 1. Comparison of time series of standardised estimates (maximum
= 1) of the combined sardine and anchovy biomass versus the breeding
population size of African penguins, for (A) the west and southwest coasts
of South Africa (1989–2016) and (B) Algoa Bay colonies (1999–2016).
Figure 2. Numbers of African penguin colonies in dierent size categories
in 2004 and 2019 in South Africa and Namibia (bars, le y-axis). Also
shown is their probability of extinction over a 40 y period (black line,
right y-axis), based on empirical information in Crawford et al. 2001
(South African Journal of Marine Science 23: 435–447). e 0-category for
colony size includes colonies that existed in 1956, but were extinct by 2004
or 2019.
colony sizes also dwindle, rendering them susceptible to
Allee eects (reduced growth rate due to overcrowding)
and higher probabilities of extinction. Based on empirical
evidence, for a colony to have 0% likelihood of extinction
in the medium term (40 y), it requires >5,000 breeding
pairs. In 2004, three colonies exceeded this size, but none
do currently (Fig. 2). Most colonies are now <100 breeding
pairs with a very high likelihood of extinction, while sev-
eral others are extinct.
Conservation eorts should focus particularly on remain-
ing colonies with 1,000–5,000 breeding pairs, which are es-
timated to have only a 33% probability of extinction over
40 y, as opposed to colonies with <250 breeding pairs and
a 100% likelihood of extinction. With food shortage likely
driving the decline, it is vital to ensure adequate food avail-
ability within range of breeding colonies. Interventions to
achieve this may include closing important seabird forag-
ing areas (oen adjacent to key colonies) to relevant sh-
ing, implementing ecosystem thresholds below which such
shing is restricted, or attempt to articially establish new
colonies in the vicinity of prey resources.
e status of many predator populations is dependent upon availability of prey, which can inuence foraging eort, body
condition and various demographic parameters, including survival, breeding participation and success. Insucient food
resources may negatively aect demographic parameters, leading to population decreases. Given current and projected
future change in South Africas marine environment, there is growing concern regarding potential eects that climate-
mediated shis on the availability of prey will have on the long-term viability of marine top predator populations, includ-
ing seabirds.
Authors: Makhado AB, Crawford RJM, Dyer BM, Visagie L, Masotla MJ
(OC Research)
24. EFFECTS OF LIMITED FORAGE FISH AVAILABILITY ON AFRICAN PENGUINS
Colony size
Proportion of maximum
Year
0.8
1
0.6
0.4
0.2
0
0.8
1
0.6
0.4
0.2
0
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
1999
2001
2003
2005
2007
2009
2011
2013
2015
sardine and anchovy African penguin
sardine and anchovy African penguin
A
B
0
20
40
60
80
100
0
2
4
6
8
10
Percentage
Number
Colony size
2004 2019 % Probability of extinction
0
20
40
60
80
100
0
2
4
6
8
10
Percentage
Number
Colony size
2004 2019 % Probability of extinction
Oceans and Coasts Annual Science Report, 2021 28
RESEARCH HIGHLIGHTS
25. MORTALITY EVENT OF CAPE FUR SEALS IN SOUTH AFRICA DURING 2021
Mass mortalities in Cape fur seals have previously only
been reported for Namibia. e 1994/95 mass mortality
event in Namibia was the largest known for any seal spe-
cies. In South Africa, an unprecedented mortality event
was recorded between September and December 2021.
Mostly pups and juveniles were aected, at colonies around
the West Coast Peninsula (Fig. 1) and to the north (i.e.
Lambert’s Bay and Elands Bay). Approximately 1,660 dead
seals were found on the beaches, and subsequently buried.
Sporadic reports of emaciated seals were also received from
the south coast and as far east as Gqeberha.
Post-mortem investigations of carcasses (Fig. 2) revealed
that seals died in a poor condition with reduced blubber
reserves. Protein energy malnutrition was detected for
aborted foetuses, for juveniles and subadults, and for
one adult. A series of tests for bacteria (e.g. brucella and
salmonella), viruses (e.g. distemper and avian inuenza)
and biotoxins (e.g. domoic acid and dinophysistoxin) were
conducted to identify possible underlying factors.
All test results were negative or insignicant, although the
negative results for domoic acid may have been aected
by delayed testing due to limited laboratory availability
and funding. Some dying seals were observed to undergo
convulsions, which commonly occur due to domoic acid
poisoning. However, these signs can also be attributed to
hypoglycaemia (abnormally low blood sugar), which can
be caused by long-term starvation.
Figure 1. Distribution of Cape fur seal colonies around South Africa.
Figure 2. Dead seal pup found lying on a rock in St Helena Bay during the
2021 mortality event.
Cape fur seals Arctocephalus pusillus pusillus breed o the coasts of Angola, Namibia and South Africa. e South
African population, which includes colonies at 25 localities, mostly on the west coast (Fig. 1), was estimated at ca. 725,000
individuals in 2020. is is about 40% of the total population, which has previously been estimated at up to 2 million.
Known threats to the conservation of the species include entanglement in shing gear, bycatch during trawling, oversh-
ing and environmental perturbations aecting prey availability, increased impacts of extreme weather (storms drowning
pups, heat stroke) and epidemic diseases.
Authors: Seakamela SM, Kotze PHG, McCue SA, De Goede J, Lamont T
(OC Research), Pieterse J (Cape of Good Hope SPCA), Smith M (CSIR),
Anthony T (Western Cape Department of Agriculture)
Acknowledgements: We thank organisations that participated in moni-
toring and responding to this event; including West Coast Seal Project,
SeaSearch and Hout Bay Seal Rescue
No unusual environmental conditions that may have
triggered the die-o, or caused it indirectly, could be iden-
tied. For example, there were no harmful algal blooms
(HABs), which typically occur at a dierent time of year
(February-March). e 2021 wind forcing, circulation
patterns, and water column structures (vertical variations
in temperature and salinity) near the aected colonies
showed no obvious dierences compared to environmental
conditions observed during 2020. Nevertheless, sheries
research surveys showed 2021 to be a year of below average
recruitment of anchovy and sardine, which are favoured
prey of seals. Fisheries catches were also poor, with only
44.4% of the Total Allowable Catch (TAC) being landed.
ese results support the low availability of at least some of
the seals’ prey resources.
However, the underlying causes of the 2021 mortality event
remain uncertain. Various samples have been preserved for
future testing using alternative methods in collaboration
with the international community. On the positive side,
the event provided a learning experience for veterinary
pathologists and marine biologists alike. In a process led
by the state veterinarian and DFFE, a monitoring protocol,
including sampling and testing guidelines, will be
developed prior to the next breeding season. It is envis-
aged that these new protocols will enable early detection of
diseases and poor health in the seal populations.
29 Oceans and Coasts Annual Science Report, 2021
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
26. OCIMS DATA MANAGEMENT RESPONSE EFFORTS TO THE AVIAN INFLUENZA
OUTBREAK
Initial reporting mechanisms were uncoordinated, with in-
formation received in various formats such as text messag-
es, spreadsheets, social media or telephonic reports. ere
was thus a need to formalise a coherent reporting structure
that could incorporate locational information to monitor
the spread of the outbreak. e DFFE Oceans and Coastal
Management Information System (OCIMS) team proposed
the use of Geo Apps for in-eld data collection that can be
deployed on any Windows, Android or IOS device.
us, with input from the WCG, a data capture form us-
ing an online platform (ESRI Survey123) was developed
to provide information to better manage the current event
and serve as a technology pilot for improved management
of future events. e platform creates and stores data in a
cloud-hosted geodatabase comprised of spatially enabled
homogenous groupings of features with standardised at-
tributes. e primary attributes include the bird species,
time, date, location, condition of the aected bird(s), inter-
ventions and images where available. To facilitate the data
capture process, the forms are accessible through a simple
graphical user interface on the user’s device (Fig. 1).
Geodatabases are spatially enabled, meaning data can be
seamlessly integrated into various visualisation applica-
tions. us, an interactive webmap was developed and
hosted in DFFE’s ArcGIS Online cloud. e map updates
in real time, and allows universal access to a single source
of data at any given time. On conclusion of the declared
incident, the data will be made available for general access
via the Marine Information Management System (MIMS).
Between 1 October 2021 and 31 January 2022, the app
was used by delegated trained ocials to report multiple
aected bird species found during 297 individual coastal
monitoring patrols. Data were submitted at the time of
observation and were immediately available for interroga-
tion, visualisation and/or download via the webmap. Sub-
sequently, the time-enabled data provided an animated
visual depiction of the spread of the outbreak, based on the
date, time and location of observations (Fig. 2).
rough this data management approach, coordina-
tion across sectors was improved as data were received in
real time and data formats were standardised and readily
available for analysis and/or reporting. e time-enabled
function aorded authorities the ability to visually moni-
tor the spread of the outbreak, consider hotspots, and
determine the deployment of resources. rough combined
local knowledge, regular stakeholder engagements and
embracing technological solutions, the overall response
eorts to monitor and curb the progression of the out-
break were improved, with lessons learnt through this
experience that can be applied to future extreme events.
Between October and November 2021, the Western Cape Government (WCG) veterinary services reported that approxi-
mately 18,388 birds had died as a result of an avian inuenza (AI) outbreak. Of these mortalities, 17,926 were Cape
cormorants Phalacrocorax capensis, an Endangered species that is endemic to Southern Africa. While it was not
declared a disaster, the Provincial Disaster Management Centre (PDMC) was activated in September 2021 to coordinate
the response, in the interest of public heath and safety.
Figure 1. An example of data capture through the interface on a
smartphone. Author: Williams L (OC Research)
Figure 2. Online webmap interface with red dots indicating where
aected birds were found, and blue dots showing where patrols occurred,
but no birds were found. e hotspot intensity radius (red margins with
yellow ll) depicts the total number of birds found per site, relative to
other sites (a greater radius indicates a higher number of aected birds).
Oceans and Coasts Annual Science Report, 2021 30
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
27. NEW MIMS WEB PORTAL IMPROVES DATA SHARING AND DISCOVERY
is portal (Fig. 1) provides access to marine observations
in both (near) real-time and delayed modes, and supports
a full range of processes such as data discovery, extraction,
access and citation in a timely manner. Access to data is
facilitated through a single-entry point. Some of the key
developments that have improved data access and sharing
since 2019 are the new MIMS catalogue and the new MIMS
ticket system. e catalogue was developed to improve
data discovery and accessibility and provide users with
dynamic ways to search for datasets. e MIMS handles
a large number of datasets, and the catalogue now allows
keyword-driven searches to focus results, based on, for
example, sensor type, project, or vessel name specications.
e ticket system allows users to request any dataset
archived on the MIMS, and also to track progress on their
requests, ensuring that users receive their requested data-
sets in a timely manner. e ticket system also enables
data providers to submit their data in a systematic way. It
implements ‘end to end’ data management by ensuring that
all requirements such as metadata records are completed
prior to data submission.
In 2014, OC Research, in partnership with the South African Environmental Observation Network (SAEON), embarked
on a journey to preserve marine and coastal datasets collected by the South African marine science community through
data hosting on the DFFE Marine Information Management System (MIMS) platform. e motivation, objectives and a
brief description of the development of the MIMS were highlighted in the 2019 Annual Science Report. Here we present on
recent progress, including increased functionality and user experience following the launch of the new MIMS web portal.
Figure 1. e MIMS web portal available via https://data.ocean.gov.za.
e South African marine science community generates
a substantial variety of data, including additional data
products generated from in situ, satellite, and model out-
put. e aim of the MIMS is to ingest and provide long-
term archiving of all of these datasets, regardless of format.
For example, the MIMS has accessioned a total of 712
recent datasets and has completed the ingestion of histori-
cal data collections from the now-defunct South African
Data Centre for Oceanography (SADCO). To keep up with
the growing needs of these datasets and products, newly
added computing infrastructure has increased the storage
and computational capability of the MIMS by 150%.
e MIMS therefore provides the gateway to South Africa’s
marine and coastal data collected in multiple disciplines
(ocean physics, chemistry, biogeochemistry and biology),
with users spending less time searching for data due to
the user-friendly nature of the portal. e MIMS will also
be working closely with the Benguela Current Commis-
sion (BCC) to develop a regional geoportal to expand data
discovery regionally and internationally.
Authors: Rasehlomi T (OC Research), Chiloane L (SAEON), Krug M (OC
Research)
31 Oceans and Coasts Annual Science Report, 2021
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
28. UNDERWAY TEMPERATURE MEASUREMENTS FROM THE THERMOSALINOGRAPH
ON THE SA AGULHAS II
Seawater is pumped continuously from just below the sea
surface (5 to 7 m depth) through a plumbing system to the
TSG (Fig. 1). Prior to the water reaching the TSG, it passes
through a de-bubbler that reduces air bubbles to allow for
more accurate measurements. e SBE45 measures tem-
perature over a range of -5 to 35°C, with a resolution of
0.0001°C. e conductivity cell measures over a range of
0–70 mS cm-1, with a resolution of 0.0001 mS cm-1.
During the annual SA Agulhas II voyage in April 2020
(Fig. 2), we collected TSG data between Cape Town and
the Prince Edward Islands (PEIs) and compared it to the
Operational Sea Surface Temperature and Sea Ice Analy-
sis (OSTIA) Sea Surface Temperature (SST) data. OSTIA
SST and the SBE38 temperature showed good agreement,
with only minor dierences, particularly at sharp gradients
associated with oceanographic fronts and features. TSG
data provides very high spatial resolution observations
along the ship’s track, while OSTIA SST has a much coarser
spatial resolution. us, it is expected that strong gradients
will be smoothed out in the lower resolution SST data.
e TSG temperature is on average 2°C warmer than the
SBE38 (Fig. 2). Such consistent warming of seawater from
the hull intake to the TSG is abnormal and illustrates a
technical design aw for the SA Agulhas II. In compari-
son, dierences between intake and TSG temperatures on
another DFFE research vessel, the RS Algoa, are on average
well below 0.5°C. During the 2021 SA Agulhas II voyage
to the PEIs, diagnostic tests revealed that the large tem-
perature dierences were likely due to the long plumbing
lengths required to transport seawater from the hull intake
to the TSG in the scientic laboratory. is plumbing is
situated directly below heater-warmed decks and passes
through areas of the vessel that are kept constantly warm
by internal air-conditioning, further exacerbating the
warming of seawater en route to the TSG.
TSG data provides high-resolution characterisations of
surface temperature and salinity variations that are use-
ful for long-term studies to identify climate-related sur-
face oceanographic changes. is data is also required
to improve knowledge on air-sea exchange, and can play
an important role in validation of satellite SST, as well as
other modelled datasets. us, continued acquisition of
accurate in situ TSG data is crucial, and eorts to eliminate
the abnormal warming of the TSG seawater supply on the
SA Agulhas II are currently underway.
A ermosalinograph (TSG) is an instrument that continuously measures in situ temperature and conductivity. All DFFE
research vessels are equipped with a SBE38 sensor, which measures temperature at the intake on the vessel’s hull, and a
SBE45 TSG that measures temperature and conductivity, every 6 or 10 seconds (Fig. 1).
Figure 1. Simplied schematic diagram of a TSG installation:
1 = hull intake, 2 = SBE38 sensor, 3 = de-bubbler, 4 = SBE45 TSG 5 = inter-
face box, and 6 = computer acquiring and storing TSG data.
Figure 2. Le: OSTIA SST map on 22 April 2020 showing the ship’s track (black line) between Cape Town and the PEIs. Right: SBE38, SBE45
and OSTIA temperature extracted along the ship’s track. Labels indicate the positions of fronts (in black) and features (in purple), with
a = warm water plume inshore of the Agulhas Current (AC), b = inshore edge of AC, c = northern edge of Agulhas Return Current (ARC), d = southern
edge of ARC, e = anticyclonic eddy, f = cyclonic eddy, g = warm water plume advected around the cyclonic eddy, h = northern branch of the sub-Antarctic
Front (N-SAF) and i = middle branch of sub-Antarctic Front (M-SAF).
Authors: Jacobs L (OC Research), Toolsee T (UCT), van den Berg MA,
Lamont T (OC Research)
Contributor: Tutt GCO (OC Research)
65
3
4
2
1
Oceans and Coasts Annual Science Report, 2021 32
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
29. GLORYS OCEAN MODEL CAPTURES EVENT-SCALE MESOSCALE EDDIES ON THE
SOUTHEAST COAST OF SOUTH AFRICA
GLORYS assimilates satellite-derived along-track sea level
anomalies, sea surface temperature (SST), as well as in situ
temperature and salinity from proling Argo oats. We
expanded previous evaluations (see Report 27 from 2020)
by demonstrating the ability of GLORYS to adequately
simulate event-scale mesoscale dynamics. Although mes-
oscale eddies observed by altimetry (Fig. 1a, b) are all
captured by GLORYS (Fig. 1c, d), there are some notable
dierences. On 19 January 2017, the size and shape of the
cyclonic eddy (CE1) is more distinguishable in GLORYS
than in the altimetry. e anticyclonic eddy (AE1) ob-
served in the model cannot be clearly identied in the
altimetry (Fig. 1a, c). Similarly, on 29 July 2017 (Fig. 1b,
d), altimetry shows a single cyclonic eddy (CE2), while
GLORYS shows two distinct features (CE1 and CE2), in
agreement with higher spatial resolution SST data (not
shown). ese discrepancies are due to dierences in spa-
tial resolution between satellite altimetry (ca. 25 km) and
GLORYS (ca. 8 km), which means that the model is able to
resolve much smaller features than altimetry.
GLORYS simulated the water column structure over 50
dierent vertical levels, which allows for more detailed in-
vestigations of vertical variations. Figure 2 highlights east-
ward ow (in red) along the northern side of the cyclonic
eddy, with westward ow (in blue) along the southern side.
is image clearly demonstrates that the eddy is intensi-
ed in the upper 1000 m but extends throughout the water
column. Upwelling associated with the eddy is discernible
from upward-doming temperature isotherms, highlighting
the movement of cold, nutrient-rich waters from deeper
layers to the surface (Fig. 2). ese observations illustrate
that GLORYS eectively simulates mesoscale ocean vari-
ability, providing daily three-dimensional views that sub-
stantially expand our contemporary in situ and satellite
observational capability in the region.
Due to limited societal drivers, such as large-scale sheries and marine mining, shelf regions along the southeast coast
of South Africa have remained largely under-sampled. In order to improve the understanding of hydrographic condi-
tions and the inuence of the Agulhas Current on the southeast coast shelf, DFFE, together with the African Coelacanth
Ecosystem Programme (ACEP) Phakisa Ocean Cruises initiative, conducted multi-disciplinary surveys during summer
and winter of 2017. ese cruises provided the only high spatial resolution in situ hydrographic observations in this region
to date. Despite providing an excellent baseline description of the shelf hydrography, they essentially remain “snapshots”
of conditions during the cruise periods. Given the limited in situ sampling opportunities, we examined output from the
GLORYS (Global Ocean Reanalysis and Simulation) model to determine whether the model can be used to investigate the
inuence of mesoscale eddies on shelf regions in more detail.
Figure 1. Sea level anomaly (SLA) maps from (a, b) satellite altimetry and
(c, d) GLORYS model output, illustrating mesoscale eddies observed on
19 January and 29 July 2017. Anticyclonic eddies are labelled with AE,
while CE indicates cyclonic eddies. e black line in panel (c) indicates the
transect along which data for Figure 2 were extracted.
Figure 2. Vertical section of zonal (west-east) current speed (m s-1)
anomalies throughout the water column, along a transect (black line on
Figure 1c) across the cyclonic eddy (CE). Positive values denote eastward
ow, while negative values denote westward ow. Black contours indicate
conservative temperature (°C).
Authors: Halo I, Lamont T (OC Research)
33 Oceans and Coasts Annual Science Report, 2021
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
30. HYDROGRAPHY OF THE SOUTHEAST COAST OF SOUTH AFRICA AS DETERMINED
FROM THE GLORYS OCEAN MODEL
During January–February 2017 (summer), in situ bottom
temperature showed cold (<10°C) water extending onto
the shelf and toward the coast in the southwestern part of
the region (Fig. 1a). is was a direct result of upwelling
and advection associated with a cyclonic eddy along the
inshore edge of the AC (see Report 29 for eddy location).
In contrast, during July–August 2017 (winter), there was
no eddy present along the southern part of the south-
east shelf, and bottom temperatures were much warmer
(>15°C), with the 10°C isotherm conned to the continen-
tal slope further oshore (Fig. 1b). Although a cyclonic
eddy was observed during the July–August cruise period,
this eddy was located further oshore (see Report 29 for
eddy location) and had minimal inuence on shelf condi-
tions (Fig. 2b). In addition, both cruise periods showed a
clear latitudinal temperature gradient, with warmer water
in the north, and generally cooler water toward the south
(Fig. 1).
While the in situ data were collected over periods of sev-
eral weeks (Fig. 1), the GLORYS model simulates daily
representations of oceanographic conditions through-
out the region (Fig. 2). Bottom temperature maps from
the GLORYS model showed excellent agreement with the
in situ data. e latitudinal temperature gradient was
also apparent in GLORYS output (Fig. 2). Similarly, shelf
cooling induced by the cyclonic eddy was evident on
19 January (Fig. 2a), with relatively warmer conditions
on 29 July (Fig. 2b). is provides further evidence of
the ability of the GLORYS model to adequately simulate
event-scale responses to mesoscale variability, suggest-
ing that the model output provides a reasonable means of
monitoring oceanographic variability in this region, in the
absence of in situ observations. e accuracy of GLORYS
is likely due to the routine assimilation of satellite data, as
well as in situ observations from proling Argo oats.
It is, however, important to note that model simula-
tions are highly dependent on the choice of numerical
scheme, the criteria used to parameterise it, as well as the
quality and quantity of the forcing elds and assimilated
observations. Analysis of water masses present on the shelf
(Fig. 3) showed that GLORYS correctly captures the large-
scale distribution of water masses, but fails to identify the
warm, low salinity waters associated with river outow in
the surface layers during summer. is is primarily due to
the absence of river discharge data in the simulation, and
the lack of adequate and sucient river discharge observa-
tions to assimilate into the model. is highlights the need
to continue and expand in situ observations in the region.
Environmental conditions on South Africa’s southeast coast are strongly inuenced by the dynamics and variability of
the Agulhas Current (AC), which is the strongest Western Boundary Current in the Southern Hemisphere. Previous
in situ studies have provided clear demonstrations of upwelling that is driven by small-scale movements of the AC, as
well as by mesoscale eddies along the inshore edge of the AC. is upwelling introduces nutrient-rich water from deeper
depths onto the shelf, which enhances biological productivity on the shelf by stimulating increases in plankton biomass.
Following on from Report 29, which demonstrated that the GLORYS (Global Ocean Reanalysis and Simulation) model
captures mesoscale variability with good accuracy, we further investigated how well the model output compared with
in situ data, collected during the summer and winter seasons of 2017. We also examined the distribution of water masses
on the southeast shelf.
Figure 1. Maps of in situ bottom Conservative Temperature (°C) during
(a) January–February 2017, and (b) July–August 2017. e solid black
contours indicate the 15°C and 10°C isotherms. Black dots indicate sam-
pled station positions.
Figure 2. Maps of GLORYS bottom Conservative Temperature (°C) on (a)
19 January 2017, and (b) 29 July 2017. e solid black contours indicate
the 15°C and 10°C isotherms.
Figure 3. Conservative Temperature versus Absolute Salinity relation-
ships for (a) in situ data from cruises during January–February and July–
August 2017, and (b) GLORYS model output during the cruise periods.
TSW - Tropical Surface Water, STSW - Subtropical Surface Water, SICW
- South Indian Central Water, AAIW - Antarctic Intermediate Water, and
RSW - Red Sea Water.
Authors: Lamont T, Halo I (OC Research)
Oceans and Coasts Annual Science Report, 2021 34
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
31. THE SOUTH AFRICAN CONTINOUS PLANKTON RECORDER SURVEY – MAPPING
PLANKTON COMMUNITIES AT THE BASIN SCALE
CPRs are robust and can operate successfully at speeds of
up to 25 knots and in rough seas. Although initially de-
signed to support sheries, CPRs have contributed to
mapping seasonal, annual and long-term changes in plank-
ton abundance and diversity since 1931 and have been
towed in most of the world’s oceans, including the South-
ern Ocean.
Plankton are sensitive to environmental change and are
ideal indicators of ocean ecological health. Plankton data
are therefore relevant for a range of management and
conservation-related matters, including impacts of cli-
mate variability and change. Common applications include
monitoring responses to ocean warming, acidication,
eutrophication, pollution and sheries, and monitoring of
marine biodiversity, invasive species, harmful algal blooms,
microplastics and pathogens aecting human health (e.g.
cholera bacteria). Plankton data from CPRs can also con-
tribute to biogeographic classications, which are useful for
planning of MPA networks.
A ‘Proof of Concept’ tow between Luanda and Port
Elizabeth (Gqeberha, Fig. 1B), in the Benguela Current
Large Marine Ecosystem (BCLME) was conducted suc-
cessfully in Oct/Nov 2005. Selected data from this survey
(Fig. 1C), includes the Phytoplankton Colour Index (PCI),
which is an indicator of phytoplankton biomass, and the
relative abundance of various key zooplankton taxa. Since
then, 32 surveys have been conducted as part of the South
African CPR survey (SA-CPR), covering over 43,000 nm
(Fig. 1D). Most eort has focused on the Southern Ocean,
since it is poorly sampled relative to other oceans and
particularly vulnerable to climate change. In 2021, quar-
terly tows across the South Atlantic from Brazil to Angola
were initiated as part of the AtlantECO programme.
Recent technological advances are enabling the addition
of sensors to CPR bodies to measure a range of physical
variables, uorescence and even molecular information.
Along with DNA extraction from historical formalin-pre-
served samples, these additions promise to yield valuable
data on basin-scale changes in plankton biodiversity in
response to climate change.
e Continuous Plankton Recorder (CPR; Fig. 1A) is a mechanical device designed for towing from merchant ships on
their normal trading routes. It is towed behind vessels at a constant depth of ca. 10 m, allowing underway collection of both
phytoplankton and zooplankton. As the CPR is towed, water enters through the front aperture, and plankton is trapped
between two layers of 270-µm silk mesh. Both silks are spooled together into a formalin-preservation tank, preserving the
‘plankton sandwich’ until subsequent laboratory analyses.
Figure 1. (A) A CPR; (B) the BCLME proof of concept tow route in 2005, with red dots indicating locations of analysed samples; (C) data from the
BCLME survey in 2005, where D/N indicates day or night sampling (white or black squares respectively), PCI is the phytoplankton colour index, NOC
is the red-tide species Noctiluca scintillans, OID and MET are upwelling-associated copepods Calanoides natalis and Metridia lucens, CAL is Calanus
agulhensis, the dominant copepod on the Agulhas Bank, COP is total copepod abundance, EUP is late stage/adult euphausiids (krill), and HYP is
hyperiid amphipods. Paler/darker shades indicate lower/higher concentrations; (D) SA-CPR tows completed to date around Southern Africa. Colour
key: green = spring, magenta = summer, orange = autumn, blue = winter. Inset: the SA Agulhas II from which most SA-CPR tows in the Southern Ocean
have been conducted.
Authors: Huggett J, Worship M (OC Research)
Contributors: van der Poel J (OC Research)
35 Oceans and Coasts Annual Science Report, 2021
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
32. THE 2022 WESTERN INDIAN OCEAN REGIONAL BENTHIC IMAGERY WORKSHOP
Underwater instruments such as drop or towed cameras,
baited remote underwater videos (BRUVs) and remotely
operated vehicles (ROVs) (Fig. 1), have increased our un-
derstanding of benthic biodiversity by enabling the collec-
tion of data from shallow and deep habitats alike. Imagery
evaluations are an important research tool used by both
scientists and industry, and the non-destructive nature of
these techniques allows for use in protected areas.
e workshop, hosted by DFFE and sponsored by the
Marine Science for Management (MASMA) programme
of the WIO Marine Science Association (WIOMSA),
provided training on how to use underwater imagery to
better understand benthic invertebrate communities and
associated sh assemblages. Initially planned as a physi-
cal workshop, it was eventually held via a Zoom webinar
platform, due to the global Covid-19 pandemic. e event,
which was conducted in partnership with Scifest Africa
and specialists from the region and beyond, drew 266 par-
ticipants (researchers, lecturers, students, technicians and
interns) from within the WIO and neighbouring countries
(Fig. 2).
Sessions covered: (i) An overview of benthos, showcas-
ing ndings of the IIOE-2 expeditions; (ii) Survey design
using visual techniques with suitable monitoring strategies
for Marine Protected Areas and the use of GIS to stand-
ardise long-term monitoring; (iii) Data collection using
camera systems along with step-by-step training videos
demonstrating the set-up and operation of a drop camera,
towed camera and BRUVs; (iv) Annotation techniques
and imagery analyses; (v) BRUV research, and (vi) Good
data management practices to enhance access and utili-
sation of data for inputs to spatial conservation planning
and management. A discussion session covered the best
practices for regional-scale habitat classication, data
challenges to overcoming these barriers, and developing
of collaborations and infrastructure.
Workshop outcomes included: (i) International attend-
ance with a far greater audience than anticipated prior
to Covid-19 (266 vs 20!); (ii) Participants trained in the
steps required to collect underwater imagery; (iii) A free
online training resource (https://www.youtube.com/
channel/UC3FT8SyYif6X_b1s4zchgog/videos) - a digital
legacy accessible to researchers both regionally and glob-
ally; (iv) Step-by-step training videos for three dierent
camera systems, providing both a teaching tool and a rst
step in standardising techniques across institutes; and (v) A
well-established network for benthic-image-based research
within the WIO.
ere is a clear need for continued development of sam-
pling guidelines, best practice protocols, locally relevant
species identication resources and image databases in the
region. us it is important to build on the momentum
established through this workshop, while continuing to
form and strengthen partnerships. In this regard, the lack
of resources and infrastructure required to utilise under-
water imagery may be seen as an opportunity for promot-
ing collaboration and innovation within the region.
e Second International Indian Ocean Expedition (IIOE-2) is a globally renowned program, and one of its objectives is
to enhance collaboration and capacity development within the Indian Ocean community. As part of this initiative, DFFE
hosted two IIOE-2 expeditions in Mozambique, Tanzania and the Comoros during 2017 and 2018, to train individuals
from the Western Indian Ocean (WIO) in a variety of marine disciplines. Participants in the benthic biodiversity team
were introduced to both traditional and innovative methodologies, including underwater camera systems.
Figure 3. Workshop success, as gauged from the feedback captured in an
online evaluation form.
Figure 2. Countries represented at the workshop. Source: L Williams (OC
Research).
Author: Haupt T (OC Research)
Contributors: DFFE; Iziko South African Museum; SAIAB; Nekton
Foundation; the University of Oxford; SAEON; UCT; UWC; SANParks;
UKZN; ORI; the University of Dar es Salaam and the University of
Comoros
Figure 1. Examples of visual sampling equipment. Photographic credit:
DFFE, SAEON, SAIAB.
Drop camera Towed camera
BRUV ROV
Good
Fair
Poor
Overall opinion Quality of content Lesson format Organisation & structure
Q & A/Discussion
sessions useful?
Step-by-step
training videos useful?
Download the
recordings?
Inspired to undertake
such research?
Excellent
Yes
No
Undecided
Oceans and Coasts Annual Science Report, 2021 36
TOOLS AND TECHNOLOGIES / TECHNOLOGICAL INNOVATION AND TRAINING
33. TRAINING WORKSHOP ON BIOLOGICAL OBSERVATIONS IN THE INDIAN OCEAN
In support of this challenge, the Indian Ocean Global
Ocean Observing System (IOGOOS) partnered with the
Sustained Indian Ocean Biogeochemistry and Ecosys-
tem Research (SIBER) team, the Indian National Centre
for Ocean Information Services (INCOIS), the Malaysian
Centre for Marine and Coastal Studies (CEMACS), and
the South African DFFE, to host a virtual training work-
shop on “Biological Observations in the Indian Ocean
(from Microbes to Megafauna), from 8–12 November
2021. Financial and logistical support was provided by the
International Training Centre for Operational Oceanogra-
phy (ITCOocean) at INCOIS and the Intergovernmental
Oceanographic Commission (IOC) of UNESCO.
e main objectives of the workshop were to (i) promote
the need for sustained biological observations in the Indian
Ocean, (ii) promote best practices for biological observa-
tions in the Indian Ocean, and (iii) present methods which
are practical and aordable, as well as those considered
as state of the art. is vision was guided by the Essential
Ocean Variable (EOV) framework developed by GOOS.
Biological EOVs have been dened for six Functional
Groups and four Habitat States (Fig. 1).
e workshop was attended by 70 students and young
researchers from 22 countries (mainly Indian Ocean Rim,
Fig. 2), with training provided by experts from Australia,
Kenya, Malaysia, Mozambique, Philippines, Réunion,
South Africa, the United Kingdom and the USA. Tech-
niques that were presented include microbial imaging and
sequencing; applications of Bio-Argo data, remote sensing
applications for harmful algal blooms, fronts and sheries,
seagrass and mangrove cover, and megafauna; zooplankton
image analysis and metabarcoding; eDNA; benthic image-
ry (various camera and video platforms including ROVs,
AUVs and BRUVs); and acoustic telemetry and satellite-
tagging for sh, sharks, turtles, seabirds and mammals.
A keynote presentation on ‘e use of biological indica-
tors in ecosystem assessments’ addressed the importance
of developing ecological indicators that are t-for-purpose
and regionally meaningful. ese include indicators for
locally relevant organisms (e.g. Noctiluca), and for physical
features inuencing local biology (e.g. boundary currents).
A desired outcome for this workshop was enhanced re-
gional networking, collaboration and mentorship opportu-
nities, and to create a cohort of regional ambassadors for
biological observations (Fig. 2). Adoption of standardised
methodology and data collection protocols will enable
comparisons and integrated data analysis across the region.
is approach will be promoted in future regional work-
shops, including hands-on training sessions.
Countries around the Indian Ocean rim rely on a healthy ocean to provide food security. However, climate change has
resulted in numerous challenges, including ocean warming, acidication and deoxygenation, which are already having
extreme impacts on ocean biology. ese impacts are predicted to increase over the coming decades. Accounting for such
changes in the conservation and sustainable management of the ocean requires baseline data provided by observations;
however, these data are still insucient for most of the world’s oceans. Hence, one of the ten key challenges of the UN
Ocean Decade is to ‘Expand the global ocean observing system’.
Figure 1. GOOS Biological Essential Ocean Variables (EOVs).
Figure 2. Screenshot from a feedback session with young Indian Ocean
researchers.
Authors: Huggett JA (OC Research, SIBER, IOGOOS), Kumar MN
(INCOIS), Tan Shau Hwai A (CEMACS)
37 Oceans and Coasts Annual Science Report, 2021
Peer-reviewed publications
Adekola OE, Crawford RJM, Dyer BM, Makhado AB, Upfold L,
Ryan PG. 2021. Timing, duration and symmetry of moult in
Cape Gannets. Ostrich 92: 295–306.
Beal M, Dias MP, Phillips RA, Oppel S, Hazin C, Pearmain EJ,
Adams J, Anderson DJ, Antolos M, Arata JA, Arcos, JM,
Arnould JPY, Awkerman J, Bell E, Bell M, Carey M, Carle R,
Clay TA, Cleeland J, Colodro V, Conners M, Cruz-Flores M,
Cuthbert R, Delord K, Deppe L, Dilley BJ, Dinis H, Elliott
G, De Felipe F, Felis J, Forero MG, Freeman A, Fukuda A,
González-Solís J, Granadeiro JP, Hedd A, Hodum P, Igual
JM, Jaeger A, Landers TJ, Le Corre M, Makhado A, Metzger
B, Militão T, Montevecchi WA, Morera-Pujol V, Navarro-
Herrero L, Nel D, Nicholls D, Oro D, Ouni R, Ozaki K,
Quintana F, Ramos R, Reid T, Reyes-González JM, Robert-
son C, Robertson G, Romdhane MS, Ryan PG, Sagar P, Sato
F, Schoombie S, Scoeld RP, Shaer SA, Shah NJ, Stevens
KL, Surman C, Suryan RM, Takahashi A, Tatayah V, Taylor
G, ompson DR, Torres L, Walker K, Wanless R, Waugh
SM, Weimerskirch H, Yamamoto T, Zajkova Z, Zango L,
Catry P. 2021. Global political responsibility for the conser-
vation of albatrosses and large petrels. Science Advances 7:
eabd7225.
Button RE, Parker D, Coetzee V, Samaai T, Palmer RM, Sink K,
Kerwath SE. 2021. ROV assessment of mesophotic sh and
associated habitats across the continental shelf of the Amat-
hole region. Scientic Reports 11: 18171.
Cade DE, Fahlbusch JA, Oestreich WK, Ryan J, Calambok-
idis J, Findlay KP, Friedlaender AS, Hazen EL, Seakamela
SM, Goldbogen JA. 2021. Social exploitation of extensive,
ephemeral, environmentally controlled prey patches by
supergroups of rorqual whales. Animal Behaviour 182: 251–
266.
Cade DE, Seakamela SM, Findlay KP, Fukunaga J, Kahane-Rap-
port SR, Warren JD, Calambokidis J, Fahlbusch JA, Fried-
laender AS, Hazen EL, Kotze D, McCue S, Meÿer M, Oestre-
ich WK, Oudejans MG, Wilke C, Goldbogen JA. 2021.
Predator-scale analysis of intra-patch prey distribution
reveals the energetic drivers of rorqual whale super-group
formation. Functional Ecology 35: 894–908.
Campbell MD, Schoeman DS, Venables W, Abu-Alhaija R, Bat-
ten SD, Chiba S, Coman F, Davies CH, Edwards M, Eriksen
RS, Everett JD, Fukai Y, Fukuchi M, Garrote OE, Hosie G,
Huggett JA, Johns DG, Kitchener JA, Koubbi P, McEn-
nulty FR, Muxagata E, Ostle C, Robinson KV, Slotwinski
A, Swadling KM, Takahashi KT, Tonks M, Uribe-Palomino
J, Verheye HM, Wilson WH, Worship MM, Yamaguchi A,
Zhang W, Richardson AJ. 2021. Testing Bergmann’s Rule in
marinecopepods. Ecography 44:1283–1295.
Carr M, Lamont T, Krug M. 2021. Satellite Sea Surface Tempera-
ture product comparison for the Southern African marine
region. Remote Sensing 13: 1244.
Dakwa FE, Ryan PG, Dyer BM, Crawford RJM, Pistorius PA,
Makhado AB. 2021. Long-term variation in the breeding di-
ets of macaroni and eastern rockhopper penguins at Marion
Island (1994–2018). African Journal of Marine Science 43:
187–199.
Engelbrecht GD, Masotla MJ. 2021. Breeding ecology of the
Quailnch (Ortygospiza atricollis) in the Limpopo Prov-
ince, South Africa. e Wilson Journal of Ornithology 132:
548–559.
Fassbender N, Stefanoudis PV, Filander ZN, Gendron G, Mah CL,
Mattio L, Mortimer JA, Moura CJ, Samaai T, Samimi-Namin
K, Wagner D, Walton R, Woodall LC. 2021. Reef benthos
of Seychelles - A eld guide. Biodiversity Data Journal 9:
e65970.
Filander ZN, Kitahara MV, Cairns SD, Sink KJ, Lombard AT.
2021. Azooxanthellate Scleractinia (Cnidaria, Anthozoa)
from South Africa. ZooKeys 1066: 1–198.
Ford D, Tilstone GH, Shutler JD, Kitidis V, Lobanova P, Schwarz
J, Poulton AJ, Serret P, Lamont T, Chuqui M, Barlow R, Lo-
zano J, Kampel M, Brandini F. 2021. Wind speed and mes-
oscale features drive net autotrophy in the South Atlantic
Ocean. Remote Sensing of Environment 260: 112435.
Gardner BR, Spolander B, Seakamela SM, McCue SA, Kotze PGH,
Musson M. 2021. Disentanglement of Cape fur seals (Arc-
tocephalus pusillis pusillus) with reversible medetomidine-
midazolam-butorphanol. Journal of the South African Vet-
erinary Association 92: e1–e5.
Kersalé M, Meinen CS, Perez RC, Piola AR, Speich S, Campos
EJD, Garzoli SL, Ansorge I, Volkov DL, Le Héna M, Dong
S, Lamont T, Sato OT, van den Berg M. 2021. Multi-year es-
timates of the daily heat transport by the Atlantic Meridion-
al Overturning Circulation at 34.5°S. Journal of Geophysical
Research Oceans 126: e2020JC016947.
Kerwath S, Roodt-Wilding R, Samaai T, Winker H, West W, Sura-
jnarayan S, Swart B, Bester-van der Merwe A, Götz A, Lam-
berth S, Wilke C. 2021. Shallow seamounts represent spe-
ciation islands for circumglobal yellowtail Seriola lalandi.
Scientic Reports 11: 3559.
Kirkman SP, Mann BQ, Sink KJ, Adams R, Livingstone T-C,
Mann-Lang JB, Pfa MC, Samaai T, van der Bank MG,
Williams L, Branch GM. 2021. Evaluating the evidence for
ecological eectiveness of South Africa’s marine protected
areas. African Journal of Marine Science 43: 389–412.
Lamont T, Louw GS, Russo CS, van den Berg MA. 2021. Obser-
vations of northeastward ow on a narrow shelf dominated
by the Agulhas Current. Estuarine, Coastal and Shelf Science
251: 107197.
Lamont T, van den Berg MA. 2021. Mesoscale eddies inuencing
the sub-Antarctic Prince Edward Islands archipelago: Tem-
poral variability and impact. Continental Shelf Research 212:
104309.
Makhado AB, Braby R, Dyer BM, Kemper J, McInnes AM, Tom
D, Crawford RJM. 2021. Seabirds of the Benguela Ecosys-
tem: Utilisation, Long-Term Changes and Challenges. In:
Mikkola H (ed.), Birds - Challenges and Opportunities for
Business, Conservation and Research. London, United King-
dom: IntechOpen. pp. 51–70.
Mann-Lang JB, Branch GM, Mann BQ, Sink KJ, Kirkman SP, Ad-
ams R. 2021. Social and economic eects of marine protect-
ed areas in South Africa, with recommendations for future
assessments. African Journal of Marine Science 43: 367–387.
Manta G, Speich S, Karstensen J, Hummels R, Kersalé M, Lax-
enaire R, Piola A, Chidichimo MP, Sato OT, Cotrim da Cun-
ha L, Ansorge I, Lamont T, van den Berg MA, Schuster U,
Tanhua T, Kerr R, Guerrero R, Campos E, Meinen CS. 2021.
e South Atlantic Meridional Overturning Circulation and
Mesoscale Eddies in the First GO-SHIP Section at 34.5°S.
Journal of Geophysical Research Oceans 126: e2020JC016962.
Masiko OB, Ryan PG, van der Lingen CD, Upfold L, Somhlaba S,
Masotla M, Geja Y, Dyer BM, Crawford RJM, Makhado AB.
2021. Are Cape Cormorants Phalacrocorax capensis losing
the competition? Dietary overlap with commercial sheries.
Ostrich 92: 280–294.
Naidoo A, Hamukuaya H, Hara M, Mngxe Y, Raakjær J. 2021.
Polycentric Regional Ocean Governance Opportunity in the
Benguela Current Convention. Frontiers in Marine Science
8: 703451.
Ngwakum BB, Payne RP, Teske PR, Janson L, Kerwath SE, Samaai
T. 2021. Hundreds of new DNA barcodes for South African
sponges. Systematics and Biodiversity 19: 747–769.
OUTPUTS FOR 2021
Oceans and Coasts Annual Science Report, 2021 38
OUTPUTS FOR 2021
Preston-Whyte F, Silburn B, Meakins B, Bakir A, Pillay K, Wor-
ship M, Paruk S, Mdazuka Y, Mooi G, Harmer R, Doran D,
Tooley F, Maes T. 2021. Meso- and microplastics monitoring
in harbour environments: A case study for the Port of Dur-
ban, South Africa. Marine Pollution Bulletin 163: 111948.
Reisinger RR, Corney S, Raymond B, Lombard AT, Bester MN,
Crawford RJM, Davies D, de Bruyn PJN, Dilley BJ, Kirkman
SP, Makhado AB, Ryan PG, Schoombie S, Stevens KL, Tosh
CA, Wege M, Whitehead TO, Sumner MD, Wotherspoon S,
Friedlaender AS, Cotté C, Hindell MA, Ropert-Coudert Y,
Pistorius PA. 2021. Habitat model forecasts suggest potential
redistribution of marine predators in the southern Indian
Ocean. Diversity and Distributions 28: 142–159.
Reisinger RR, Friedlaender AS, Zerbini AN, Palacios DM, An-
drews-Go V, Dalla Rosa L, Double M, Findlay K, Garrigue
C, How J, Jenner C, Jenner M-N, Mate B, Rosenbaum HC,
Seakamela SM, Constantine R. 2021. Combining regional
habitat selection models for large-scale prediction: Cir-
cumpolar habitat selection of Southern Ocean Humpback
Whales. Remote Sensing 13: 2074.
Rixen T, Lahajnar N, Lamont T, Koppelmann R, Martin B, van
Beusekom JEE, Siddiqui C, Pillay K, Meiritz L. 2021. Ox-
ygen and nutrient trapping in the southern Benguela Up-
welling System. Frontiers in Marine Science 8: 730591.
Russo CS, Lamont T, Krug M. 2021. Spatial and temporal variabil-
ity of the Agulhas Retroection: Observations from a new
objective detection method. Remote Sensing of Environment
253: 112239.
Samaai T, Sink K, Kirkman S, Atkinson L, Florence W, Kerwath
S, Parker D, Yemane D. 2021. e Marine Animal Forests of
South Africa: Importance for Bioregionalization and Marine
Spatial Planning. In: Rossi S, Bramanti L (eds.), Perspectives
on the Marine Animal Forests of the World. Springer, Cham.
pp. 17-61.
Singh SP, Groeneveld JC, Huggett J, Naidoo D, Cedras R, Willows-
Munro S. 2021. Metabarcoding of marine zooplankton in
South Africa. African Journal of Marine Science 43: 143–159.
Stirnimann L, Bornman TG, Verheye HM, Bachèlery M-L, van
der Poel J, Fawcett SE. 2021. Plankton community composi-
tion and productivity near the Subantarctic Prince Edward
Islands archipelago in autumn. Limnology and Oceanogra-
phy 66: 4140–4158.
Teske PR, Emami-Khoyi A, Golla TR, Sandoval-Castillo J, La-
mont T, Chiazzari B, McQuaid CD, Beheregaray LB, van der
Lingen CD. 2021. e sardine run in southeastern Africa is
a mass migration into an ecological trap. Science Advances
7: eabf4514.
Toolsee T, Lamont T, Rouault M, Ansorge I. 2021. Characteris-
ing the seasonal cycle of wind forcing, surface circulation
and temperature around the sub-Antarctic Prince Edward
Islands. African Journal of Marine Science 43: 61–76.
Vargas-Fonseca OA, Yates P, Kirkman SP, Pistorius PA, Moore
DM, Natoli A, Cockcro V, Hoelzel AR. 2021. Population
structure associated with bioregion and seasonal prey distri-
bution for Indo-Pacic bottlenose dolphins (Tursiops adun-
cus) in South Africa. Molecular Ecology 30: 4642–4659.
Vinayachandran PNM, Masumoto Y, Roberts MJ, Huggett
JA, Halo I, Chatterjee A, Amol P, Gupta GVM, Singh A,
Mukherjee A, Prakash S, Beckley LE, Raes EJ, Hood R.
2021. Reviews and syntheses: Physical and biogeochemical
processes associated with upwelling in the Indian Ocean.
Biogeosciences 18: 5967–6029.
Popular articles
Haupt T. 2021. e WIO regional benthic imagery work-
shop. Western Indian Ocean Marine Science Association
(WIOMSA) Newsbrief 29: 9–11.
Presentations at symposia, conferences and workshops
Carr M, Lamont T, Krug M. 2021. Comparing Sea Surface Tem-
perature products for implementation within South Africa’s
operational system. 2nd International Operational Satellite
Oceanography Symposium (OSOS-2), Online, 25–27 May
2021.
Chioze C, Noyon M, Huggett J, Isari S, Malauene B, Roberts M.
2021. e abundance and distribution of zooplankton com-
munities on the Mozambican Shelf. Recruitment success in
Mozambique sheries in a highly Turbulent shelf edge ocean
ecosystem – ReMoTurb, Opening Address and Workshop
Event, Maputo, Mozambique and Online, 29 July 2021.
Coelho P, Tchipalanga P, Macuéria M, van der Plas A, N’Guessan
B, Desiré K, Makaoui A, Bessa I, Idrissi M, Ettahiri O, Hilmi
K, Halo I, Schmidtko S, Dengler M, Brandt P, Cervantes D,
Lødemol H, Chierici M, Ostrowski M, Bâ ML, WG TOM-
SWA – Ad-Hoc Working Group on Transboundary Oxygen
Monitoring Status in West Africa. 2021. An inventory of
dissolved oxygen conditions along the eastern boundary of
tropical and subtropical Atlantic: building oxygen monitor-
ing capacity in West African countries, 2013-2019. European
Geophysical Union (EGU) General Assembly, Online, 19–30
April 2021.
deYoung B, Visbeck M, Speich S, Le Traon PY, Snowden J, Hol-
liday NP, Lamont T, Cotrim L, Pinto IS, Chidichimo MP,
Ketelhake S, Zinkann A. 2021. AtlantOS: An international
program for basin-scale observing of the Atlantic Ocean.
e Global Ocean Observing System (GOOS) Steering Com-
mittee Meeting, Online Webinar, 19–23 April 2021.
Filander Z. 2021. e role of deep-water coral taxonomy in spatial
planning. Invited Presentation. Western Indian Ocean Gov-
ernance Exchange Network (WIOGEN) Conference on Ocean
Governance, Online, 27–29 October 2021.
Halo I. 2021. An Introduction to Ocean Modelling. Workshop to
Orient Final Year Undergraduate Students Pursuing Aquacul-
ture/Marine Programmes to the Coastal and Regional Ocean
Community model (CROCO). Invited Online Presentation.
University of Dodomo, Tanzania, 9 November, 2021.
Harris L, Holness S, Kirkman S, Sink K, Majiedt P, Driver M. Na-
tional Coastal and Marine Spatial Biodiversity Plan V1.1.
17th National Biodiversity Planning Forum, South African
National Biodiversity Institute, Online, 3–5 August 2021.
Haupt T. 2021. Exploring and monitoring benthic biodiversity –
local lessons and a note on sample design. Western Indian
Ocean (WIO) Regional benthic imagery workshop, Online,
30 August–3 September 2021.
Haupt T. 2021. e WIO Regional benthic imagery workshop:
a capacity development initiative. Invited Presentation.
Western Indian Ocean Governance Exchange Network
(WIOGEN) Conference on Ocean Governance, Online,
27–29 October 2021.
Huggett J. 2021. Biological observations in the context of Climate
Change and the UN Decade of Ocean Science - the EOV
approach. Indian Ocean Global Ocean Observing System
(IOGOOS) Workshop on Biological Observations in the
39 Oceans and Coasts Annual Science Report, 2021
Indian Ocean (from microbes to megafauna), Online, 8–12
November 2021.
Huggett J, Noyon M. 2021. Recent zooplankton research in the
Southwest Indian Ocean. Second International Indian Ocean
Expedition Steering Committee Meeting No. 4 (IIOE-2 SC4),
Online, 12–15 April 2021.
Kirkman SP. Marine Protected Areas in South Africa: are they
eective and where do we draw the line? Invited Online
Presentation. Biology Graduate Programme, UNISINOS
University, Brazil, 7 July 2021.
Kirkman SP. South Africa’s Marine Protected Areas: status, tar-
gets and trends. Online Presentation. WildOceans Marine
Stewardship Programme Science Session, Durban, 20–22 July
2021.
Kirkman SP, Holness S, Harris L, Majiedt P, Sink KJ. 2021. South
Africa’s revised Ecologically or Biologically Signicant Ma-
rine Areas and how they are being integrated in marine
area-based management in South Africa. Benguela Current
Convention (BCC) Science and Governance Forum 2021, On-
line, 3–5 November 2021.
Kirkman SP, Mann BQ, Sink KJ, Adams R, Livingstone T-C,
Mann-Lang JB, Pfa MC, Samaai T, van der Bank MG,
Williams L, Branch GM. 2021. Evaluating the evidence for
ecological eectiveness of South Africa’s marine protected
areas. e Conservation Symposium 2021, Online, 1–5
November 2021.
Lamont T. 2021. Status of the SAMBA-East observing sys-
tem. South Atlantic Meridional Overturning Circulation
(SAMOC) VIIII Workshop, Online, 6th and 15th April 2021.
Maletzky E, Kirkman SP, Barreto TM. 2021. Ecologically or Bio-
logically Signicant Marine Areas in the Benguela Cur-
rent Large Marine Ecosystem - transboundary alignment
of EBSA zoning and management recommendations to-
wards achieving a regional MSP. Benguela Current Conven-
tion (BCC) Science and Governance Forum 2021, Online,
3–5 November 2021.
Noyon M, Adams J, Deyzel S, Huggett J, van der Lingen C, Andre
M, Ruby C. 2021. Synopsis of Southern African Plankton
Imaging Platforms & Research. I/ITAPINA: Imagine/Imag-
ing e Atlantic – A Pelagic Imaging Network Approach, First
I/ITAPINA workshop, Online, 28–29 June 2021.
Pfa MC, Krug M, Kirkman SP. 2021. Coastal ecosystem moni-
toring initiatives and decision-making tools in South Africa.
Invited Online Presentation. Ocean Visions Summit: Towards
a global ecosystem for ocean solutions, Scripps Institution of
Oceanography, USA, 18–21 May 2021.
Puccinelli E, Filander Z, Lamont T, van den Berg M, Tutt G, Sny-
ders L, Jacobs L, Lombi M. 2021. e eect of hydrography
on the benthos of a submarine canyon: the case study of
the South African Cape Canyon. Online Presentation. 16th
Deep-Sea Biology Symposium (DSBS), Brest, France, 12–17
September 2021.
Russo CS, Lamont T, Krug M. 2021. Location of the Agulhas Cur-
rent Core and Edges (LACCE): A new tool for monitoring
variability. 2nd International Operational Satellite Oceanogra-
phy Symposium (OSOS-2), Online, 25–27 May 2021.
Toolsee T, Lamont T, Rouault M. 2021. e seasonal cycle of
wind forcing, surface circulation and temperature around
the sub-Antarctic Prince Edward Islands. 2nd International
Operational Satellite Oceanography Symposium (OSOS-2),
Online, 25–27 May 2021.
von der Meden C, Pattrick P, van der Heever G, Wozniak D,
Porri F, Atkinson L, Filander Z, Madjit P, Levin L, Sink K.
2021. Prime Primnoid real estate: Quantifying complex sur-
face area of biogenic habitat and incidence of use. Online
Presentation. 16th Deep-Sea Biology Symposium (DSBS),
Brest, France, 12–17 September 2021.
Published datasets
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from Algoa Voy-
age 268, February - March 2020. DFFE. doi: 10.15493/DEA.
MIMS.25010001.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from Algoa Voyage
268, February - March 2020. DFFE. doi: 10.15493/DEA.
MIMS.25010002.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from SA Agulhas
II Voyage 029, December 2017 - February 2018. DFFE. doi:
10.15493/DEA.MIMS.25010004.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from the SA Agulhas II
Voyage 029, December 2017 - February 2018. DFFE. doi:
10.15493/DEA.MIMS.25010005.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from the SA Agul-
has II Voyage 035, November 2018 - March 2019. DFFE. doi:
10.15493/dea.mims.25010007.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from the SA Agulhas
II Voyage 035, November 2018 - March 2019. DFFE. doi:
10.15493/DEA.MIMS.25010008.
Jacobs L, van den Berg M, Lamont T. 2021. Processed under-
way ermosalinograph (TSG) observations from the SA
Agulhas II Voyage 036, April 2019 - May 2019. DFFE. doi:
10.15493/DEA.MIMS.25010009.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from the SA Agulhas II
Voyage 036, April 2019 - May 2019. DFFE. doi: 10.15493/
DEA.MIMS.25010010.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from the SA Agul-
has II Voyage 041, December 2019 - February 2020. DFFE.
doi: 10.15493/DEA.MIMS.25010012.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from the SA Agulhas II
Voyage 041, December 2019 - February 2020. DFFE. doi:
10.15493/DEA.MIMS.25010013.
Jacobs L, van den Berg M, Lamont T. 2021. Processed under-
way ermosalinograph (TSG) observations from the SA
Agulhas II Voyage 042, April 2020 - May 2020. DFFE. doi:
10.15493/DEA.MIMS.25010014.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from the SA Agulhas II
Voyage 042, April 2020 - May 2020. DFFE. doi: 10.15493/
DEA.MIMS.25010015.
Jacobs L, van den Berg M, Lamont T. 2021. Processed under-
way ermosalinograph (TSG) observations from the Al-
goa Voyage 269, October 2020. DFFE. doi: 10.15493/DEA.
MIMS.20210409.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway
ermosalinograph (TSG) observations from the Algoa
Voyage 269, October 2020. DFFE. doi: 10.15493/DEA.
MIMS.20210410.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from SA Agulhas II
Voyage 015, April 2015 - May 2015. DFFE. doi: 10.15493/
DEA.MIMS.20210411.
OUTPUTS FOR 2021
Oceans and Coasts Annual Science Report, 2021 40
OUTPUTS FOR 2021
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from SA Agulhas II
Voyage 015, April 2015 - May 2015. DFFE. doi: 10.15493/
DEA.MIMS.20210412.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from SA Agulhas II
Voyage 019, April 2016 - May 2016. DFFE. doi: 10.15493/
DEA.MIMS.20210413.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from SA Agulhas II
Voyage 019, April 2016 - May 2016. DFFE. doi: 10.15493/
DEA.MIMS.20210414.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from SA Agulhas II
Voyage 024, April 2017 - May 2017. DFFE. doi: 10.15493/
DEA.MIMS.20210415.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from SA Agulhas II
Voyage 024, April 2017 - May 2017. DFFE. doi: 10.15493/
DEA.MIMS.20210416.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from SA Agulhas II
Voyage 030, April 2018 - May 2018. DFFE. doi: 10.15493/
DEA.MIMS.20210417.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway er-
mosalinograph (TSG) observations from SA Agulhas II
Voyage 030, April 2018 - May 2018. DFFE. doi: 10.15493/
DEA.MIMS.20210418.
Jacobs L, van den Berg M, Lamont T. 2021. Processed underway
ermosalinograph (TSG) observations from SA Agul-
has II Voyage 037, July 2019. DFFE. doi: 10.15493/DEA.
MIMS.20210420.
Jacobs L, van den Berg M, Lamont T. 2021. Raw underway
ermosalinograph (TSG) observations from SA Agul-
has II Voyage 037, July 2019. DFFE. doi: 10.15493/DEA.
MIMS.20210421.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 3, July - December 2014. DFFE. doi: 10.15493/
DEA.MIMS.20210317.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 4, July - December 2014. DFFE. doi: 10.15493/
DEA.MIMS.20210318.
Lamont T, van den Berg MA. 2021. Long-term observations
of hourly currents along the SAMBA transect at SAM-
BA Mooring 4, December 2015 - April 2017. DFFE. doi:
10.15493/DEA.MIMS.20210319.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 4, April 2017 - October 2018. DFFE. doi: 10.15493/
DEA.MIMS.20210320.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 7, September 2014 - December 2015. DFFE. doi:
10.15493/DEA.MIMS.20210321.
Lamont T, van den Berg MA. 2021. Long-term observations
of hourly currents along the SAMBA transect at SAM-
BA Mooring 7, December 2015 - April 2017. DFFE. doi:
10.15493/DEA.MIMS.20210322.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 7, April 2017 - October 2018. DFFE. doi: 10.15493/
DEA.MIMS.20210323.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 8, September 2014 - December 2015. DFFE. doi:
10.15493/DEA.MIMS.20210324.
Lamont T, van den Berg MA. 2021. Long-term observations
of hourly currents along the SAMBA transect at SAM-
BA Mooring 8, December 2015 - April 2017. DFFE. doi:
10.15493/DEA.MIMS.20210325.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 8, April 2017 - October 2018. DFFE. doi: 10.15493/
DEA.MIMS.20210326.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 9, September 2014 - December 2015. DFFE. doi:
10.15493/DEA.MIMS.20210329.
Lamont T, van den Berg MA. 2021. Long-term observations
of hourly currents along the SAMBA transect at SAM-
BA Mooring 9, December 2015 - April 2017. DFFE. doi:
10.15493/DEA.MIMS.20210330.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 9, April 2017 - May 2018. DFFE. doi: 10.15493/
DEA.MIMS.20210331.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 10, September 2014 - December 2015. DFFE. doi:
10.15493/DEA.MIMS.20210334.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents along the SAMBA transect at SAMBA
Mooring 10, December 2015 - April 2017. DFFE. doi:
10.15493/DEA.MIMS.20210335.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents measured by DVS at SAMBA Mooring 8,
December 2015 - April 2017. DFFE. doi: 10.15493/DEA.
MIMS.20210327.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents measured by DVS at SAMBA Mooring 8,
April 2017 - October 2018. DFFE. doi: 10.15493/DEA.
MIMS.20210328.
Lamont T, van den Berg MA. 2021. Long-term observations of
hourly currents measured by DVS at SAMBA Mooring 9,
December 2015 - April 2017. DFFE. doi: 10.15493/DEA.
MIMS.20210332.
Lamont T, van den Berg MA. 2021. Long-term observations
of hourly currents measured by DVS at SAMBA Moor-
ing 9, April 2017 - May 2018. DFFE. doi: 10.15493/DEA.
MIMS.20210333.
Lamont T, van den Berg MA. 2021. Raw ADCP data from
long-term observations of currents at SAMBA Mooring
3, July 2014 - December 2014. DFFE. doi: 10.15493/DEA.
MIMS.25010017.
Lamont T, van den Berg MA. 2021. Raw ADCP data from
long-term observations of currents at SAMBA Mooring
4, July 2014 - December 2014. DFFE. doi: 10.15493/DEA.
MIMS.25010018.
Lamont T, van den Berg MA. 2021. Raw ADCP data from long-
term observations of currents at SAMBA Mooring 4, No-
vember 2015 - April 2017. DFFE. doi: 10.15493/DEA.
MIMS.25010019.
Lamont T, van den Berg MA. 2021. Raw ADCP data from long-
term observations of currents at SAMBA Mooring 7, Sep-
tember 2014 - December 2015. DFFE. doi: 10.15493/DEA.
MIMS.25010020.