Basking shark population
assessment
Final Project Report
for
Global Wildlife Division of the
Department for Environment, Food and Rural Affairs
Basking shark population assessment
Final report for Global Wildlife Division of Defra
Tender CR 0247
D.W. Sims1, E.J. Southall1, J.D. Metcalfe2, M.G. Pawson2
1
Marine Biological Association of the UK, The Laboratory, Citadel Hill, Plymouth, Devon, UK
PL1 2PB.
Contact: dws@mba.ac.uk
2
Centre for Environment, Fisheries and Aquaculture Science, The Lowestoft Laboratory,
Pakefield Road, Lowestoft, Suffolk UK NR33 0HT.
Contact: j.d.metcalfe@cefas.co.uk
Acknowledgements: The authors wish to thank the following for their help and assistance at
various stages during the fieldwork, data analysis and report writing: David Uren, Peter Harris,
Gareth Fraser, Steve Moszolics, Victoria Wearmouth, Alec and Donald MacKenzie, John and
Nick Boyle, the late Alan Russell (MBA tagging team), Peter Miller (NERC RSDAS), Anthony
Richardson (SAHFOS), Matt Witt (University of Exeter), Vin Fleming (JNCC), Sarah Fowler
(Nature Bureau), Phil Lewis and Sally Davis (Defra, GWD), Jeremy Stafford-Deitsch and the
Shark Trust for the use of the basking shark photograph (front cover), and the Marine
Conservation Society for the use of the map of historic UK basking shark fishing grounds (p.
24).
Copies of this report can be downloaded as an Adobe Acrobat file from the Defra web site at:
http://defraweb/wildlifecountryside/resprog/findings/index.htm
Department for Environment, Food and Rural Affairs
Nobel House
17 Smith Square
London SW1P 3JR
Telephone 020 7238 6000
Website: www.defra.gov.uk
© Crown copyright 2005
Copyright in the typographical arrangement and design rests with the Crown.
This publication (excluding the logo) may be reproduced free of charge in any format or
medium provided that it is reproduced accurately and not used in a misleading context. The
material must be acknowledged as Crown copyright with the title and source of the publication
specified.
EXECUTIVE SUMMARY
Introduction
Populations of many large marine vertebrates are threatened by high levels of
fisheries exploitation (both targeted and as bycatch). This applies particularly to
sharks, skates and rays (elasmobranch fishes) that have life-history traits that make
them especially vulnerable to levels of harvest mortality that are above that of natural
mortality. In particular, many elasmobranchs have a late age at maturity and low
fecundity leading to low rates of reproduction. This results in little scope for the
compensatory mechanisms that enable many “bony” fish species like cod or mackerel
to withstand unnaturally high levels of mortality. As a consequence, elasmobranch
fisheries not only exhibit rapid declines in catch rates as exploitation increases, but
there is a greater potential for the fishery to collapse.
The basking shark is the world’s second largest fish and is widely distributed in
coastal waters on the continental shelves of temperate zones in both northern and
southern hemispheres. Individuals take 12-20 years to reach maturity, females have
long gestation periods (1-3 years) and give birth to a few, large young. This inherent
vulnerability to exploitation, together with concern over the strong possibility that
populations are depleted as a result of exploitation by fisheries and the lack of
scientific knowledge of the species, has led to the basking shark being listed as
Vulnerable worldwide, and Endangered in the north-east Atlantic, in the 2004 IUCN
Red List (IUCN, 2004). In 2002, the species was listed on Appendix II of the
Convention on International Trade in Endangered Species. They are also protected in
British territorial waters under Schedule 5 of the Wildlife and Countryside Act (1981).
The current approach to conservation of basking sharks relies heavily on the
precautionary principle, which states that insufficient scientific knowledge about
biology and stock status is no defense for a lack of action. In particular, very little is
known about migration routes, whether there are discrete local populations of basking
sharks, or the relationship between regional population abundance and global trends.
There is also a need to distinguish between the effects on population status of climate
change on the sharks’ environments and the legacy of the impact of fisheries, and to
ascertain whether stocks are depleted or recovering.
The latter concern is recognised in the UK basking shark Biodiversity Action Plan,
which states that improved long-term (many decades) monitoring of the UK
population is necessary to enable population trends to be identified. Though some
monitoring data for this species are available, most are based on sightings of sharks
feeding on plankton near the sea surface during spring and summer and no analyses
done to date has provided information on population trends.
Aims of the study
This study set out to use modern satellite telemetry to determine movements and
behaviour of basking sharks in the north-east Atlantic, especially that part of the
population that occurs at some time within UK territorial waters. This information has
improved our understanding of the status of basking shark stocks that will inform
decisions in relation to conservation measures and help ensure recovery and
sustainability of basking shark populations.
Methods
The movements and behaviour of 20 basking sharks was monitored by tagging them
with pop-up archival transmitting (PAT) tags. The tags record swimming depth, water
temperature and light level while being towed by a shark. At a pre-specified time, the
tag detaches from the shark, floats to the sea surface and transmits summarised
i
information via the Argos system aboard NOAA polar-orbiting satellites. In addition to
using PAT tags, 3 sharks were tagged with pop-up archival tags that could transmit
data via the terrestrial cellular-telephone network once they had drifted near to shore.
Geographical movements of the sharks were calculated using retrieved data on daily
light intensity to estimate the local times of midnight or midday for longitude
calculations. Latitude was then determined using night time Advanced Very High
Resolution Radiometer (AVHRR) remote-sensing images of sea surface temperature
along the longitude and temperature data recorded by the tag. Retrieved data also
provided summary information about time-at-depth and the time-at-temperature.
A number of tags were also returned by members of the public who found them in the
sea or on beaches. These tags provided detailed minute-by-minute data on the
vertical movements of the sharks.
Results and conclusions
Data sufficient for reconstructing geographical movements were retrieved from 8 of
the 23 basking sharks tagged. The results show that the sharks spent most of their
time on the European continental shelf, moving between centres of high zooplankton
productivity characterised by tidal fronts and fronts associated with the shelf break.
Individuals foraging along fronts off the south-west peninsula of England moved to
three main areas: the Celtic Sea front, the Goban Spur and the north Biscay regions
of the shelf edge. Two individuals moved northward from these areas along the shelf
edge into rich feeding areas in the Hebridean Sea that are also characterised by
strong tidal fronts that aggregate zooplankton. None of the sharks tagged off south
west England in spring moved northwards through the Irish Sea during summer,
whereas those sharks tagged in the Clyde Sea off western Scotland in summer
travelled relatively rapidly south through the Irish Sea in late summer and early
autumn to areas off south-west England. In winter, some individuals remained or
moved into shallow coastal waters, generally in the southern region of the shelf.
These results suggest that there are no separate populations of basking sharks
inhabiting northern or southern UK waters. Individuals move freely between these
areas and probably represent a single population.
The depth data show that there is generally a preference for deeper depths in the
water column (or deeper water) on the shelf during winter compared with summer, a
shift that at least in part explains why basking sharks are rarely observed at this time.
Patterns of movement show that, contrary to earlier belief, basking sharks do not
hibernate on the sea bed in the winter.
The existence of differing patterns of diel (daily) vertical movement in individual
sharks in different ocean habitats in UK waters resulted in very different surfacing
frequencies. The daytime-surfacing frequency of a tracked individual feeding in an
inner-shelf area near a front was over 100 times higher than another shark feeding in
well-stratified water. This large difference in ‘basking’ behaviour between regions was
reflected in our shark survey data with 11.5 times more shark sightings per unit effort
in frontal areas than in stratified water. This suggests sightings per unit effort may not
necessarily reflect real differences in geographic (horizontal) abundance between
areas, because the probability of sighting a basking shark may be about 60 times
higher in frontal sea areas than in sea areas that are well-stratified. This has profound
implications for the use of sighting data both in defining population distribution and
estimating abundance trends. These results suggest that bias-reduction according to
habitat type (and zooplankton behaviour) should be incorporated into analyses of
survey data when attempting to estimate abundance.
ii
The distribution of tag geolocations for individual sharks has been compared with
sightings of basking sharks compiled as part of the “Conserving endangered basking
sharks” project and sightings made during scientific surveys by the MBA, Hebridean
Whale and Dolphin Trust, The Shark Trust, UK Wildlife Trusts and the International
Fund for Animal Welfare. Whilst the broad distribution patterns revealed by these
different methods are similar, there are considerable differences in density
distributions, with a strong emphasis in the sightings data in the Hebridean Sea,
Clyde Sea, Irish Sea and close inshore around Devon and Cornwall, areas that
clearly represent important habitat for basking sharks, most probably in relation to
feeding opportunities. Tag geolocations, however identified two areas where
individuals spent considerable time outside the distributions indicated by sightings:
the Celtic Sea and Western Approaches. In particular, tag geolocations show sharks
undertaking persistent ranging movements near the Celtic Sea front, whereas surface
sightings would indicate relatively few sharks in this area. The most likely reason for
this discrepancy is that ‘basking’ behaviour in this area is reduced, resulting in few
surface sightings.
Recommendations and further work
The principal finding of the present study is that basking sharks use a much greater
geographical distribution range than sightings information alone would indicate.
However, whilst it is apparent that sharks using feeding areas off southwest England
and in the Clyde Sea occupy a similar distribution range over a period of a year or so,
we do not know how representative these fish are of the population living more widely
across the north-east Atlantic, nor whether there are longer term population
movements that might indicate a much wider stock range. This information is crucial
for underpinning robust scientific advice on the conservation needs of basking sharks
and management of their fisheries (if any).
To enable us to be confident that we know what part of the north-east Atlantic
population the basking sharks present around Britain represent, genetic studies are
being carried out under a linked Defra-funded project aimed at describing stock
characteristics in the north-east Atlantic in comparison with basking shark populations
elsewhere in the world. However, we do not yet know if the results will enable us to
distinguish a “British” population. To achieve this, additional archival tagging of sharks
is required along different parts of the western seaboard, such as south Brittany,
northwest Scotland and Norway to elucidate the extent of movement and mixing
between local groups of basking sharks. Further archival tagging of mature (>7 m)
sharks is also required. The results will allow us to know whether exploitation outside
the “normal” range of British basking sharks will affect them, and whether
conservation measures around the UK will have any effect, either locally or on the
north-east Atlantic population as a whole.
Such studies should enhance and verify the existing behavioural information on
basking shark movements; seasonally, horizontally and vertically, and in relation to
oceanic conditions. This will provide weightings to sightings data in order to help
determine changes in local abundance and stock trends, on the assumption that
sightings surveys should be encouraged to raise the basking shark’s profile and as a
long-term monitoring tool. These data can also be used as inputs to bio-energetic
models of movement and migration that could be used to test climate change and
exploitation-pattern type scenarios, validated against existing information (including
historic fisheries data) and to make up for the inevitable data deficiencies. This may,
for example, help to identify the potential impact of environmental change on basking
shark populations, and how this compares with past or current directed fishing
activities. Future consideration of conservation measures will need to take such
factors into account.
iii
Basking shark population assessment
Contents
Executive summary
i
1. Policy and scientific background
1.1 Scope of the Proposed Study
1.2 Objectives
1
2
3
2. General methodology
2.1 Previous methods
2.1.1 Population monitoring
2.1.2 Behaviour
2.1.3 New technologies
4
4
4
4
4
2.2 Methodologies used
2.2.1 Satellite transmitters
2.2.2 Track reconstruction
2.2.3 GSM telephone tags
2.2.4 Plankton analysis
2.2.5 Deployment of electronic tags
2.2.6 Summary of field surveys
2.2.7 Sex and maturity
2.2.8 Animal welfare
3. Results
3.1 Geographical Area
3.2 Sightings
3.3 Length-frequency
3.4 Data recovery
3.4.1 Argos
3.4.2 Mobile telephone network
3.4.3 Tag recovery
5
5
7
8
9
9
11
13
13
14
14
14
14
15
15
16
16
3.5 Horizontal movements
3.5.1 Northward movements
3.5.2 Southward movements
3.5.3 Celtic sea – Biscay movements
3.5.4 Eastward movements
3.5.5 General movement patterns
3.5.6 Occurrences of sharks in different sea areas
17
18
19
20
21
22
24
3.6 Vertical Movements
3.6.1 Summer behaviour
3.6.2 Autumn/Winter behaviour
3.6.3 Daily dive patterns
3.6.4 Surfacing frequency
25
25
27
29
31
4. Conclusions
4.1 The technical approach
31
31
4.2 Summary of results
33
4.3 Population monitoring
4.4 Stock identification
4.5 Recommendations and further work
33
34
35
iv
5. References
37
6. Outputs
40
Appendix I:
Review of the biology and ecology of the
basking shark (Cetorhinus maximus)
Appendix 2:
Seasonal movements and behaviour of
basking sharks from archival tagging: no
evidence of winter hibernation. D.W. Sims,
E.J. Southall, A.J. Richardson, P.C. Reid and
J.D. Metcalfe, 2003, Marine Ecology Progress
Series, 248: 187-196
v
1. Policy and Scientific Background
The populations of many large marine vertebrates are threatened by too high a level
of fisheries exploitation, both by targeted fisheries and as bycatch. This applies
particularly to elasmobranchs (sharks, skates and rays), which have a number of lifehistory traits that make them particularly vulnerable to levels of harvest mortality that
are well above that of natural mortality. Recent estimates suggest that populations of
some shark species have declined by over 75% in the past 15 years (Baum et al.,
2003). In particular, many elasmobranch species have a late age at maturity and low
fecundity, leading to low rates of reproduction and low potential rates of population
growth compared to marine teleosts (Pratt & Casey, 1990). They also have relatively
little scope for the compensatory mechanisms that enable teleosts to withstand
unnaturally high levels of mortality. As a consequence, elasmobranch fisheries not
only exhibit rapid declines in catch rates as exploitation increases, but there is a
greater potential for the fishery to collapse (Holden 1973, 1977; Casey & Myers,
1998; Dulvy et al., 2000). For example, the fishery for basking shark (Cetorhinus
maximus) near Achill Island off the west coast of Ireland appeared to collapse in the
early 1960’s after only ten years of peak catches (Kunzlik, 1988). Despite this
vulnerability, the only control on fishing for basking sharks in European waters is a
total allowable catch (TAC, currently set at zero) for Norwegian vessels fishing in
Community waters. However, Norway and all other countries have now ceased to
fish for basking sharks in EC waters.
The basking shark is the world’s second largest fish and is widely distributed in
coastal waters on the continental shelves of temperate zones in both the northern
and southern hemispheres. Individuals take 12-20 years to reach maturity, females
have long gestation periods (1-3 years) at the end of which they give birth to a few,
large young (the only recorded litter is of six 2-m long pups) (Kunzlik, 1988). This
inherent vulnerability to high levels of exploitation, together with concern over the
strong possibility that populations are depleted as a result of exploitation by fisheries
and the lack of scientific knowledge of the species, has led to the basking shark
being listed as Vulnerable (A1a,d, + 2d) worldwide, and Endangered (EN A1a,d) in
the north-east Atlantic in the IUCN Red List (first listed in 1996, most recent listing
2004, IUCN, 2004). In 2000, the species was listed in Appendix III of the Convention
on International Trade in Endangered Species (CITES). In 2002, on the basis of a UK
proposal, the species was listed in Appendix II of CITES which requires that
International trade in these species is monitored through a licensing system to
ensure that trade can be sustained without detriment to wild populations. Trade in
wild, captive bred and artificially propagated specimens is allowed, subject to permit.
Though no longer exploited there, they are also protected in British (but not Northern
Irish) territorial waters under Schedule 5 of the Wildlife and Countryside Act (1981),
and it is a priority species under the UK Biodiversity Action Plan (English Nature,
1999). They are also protected within the territorial waters of the Isle of Man and
Guernsey, in the Mediterranean under the Bern Convention (with EU reservation)
and Barcelona Convention (unratified), and in US Atlantic waters. The basking shark
has also been included in the 2004 Initial OSPAR list of threatened and/or declining
species and habitats (OSPAR, 2004).
This approach to conservation of basking sharks relies heavily on the precautionary
principle, which states that insufficient scientific knowledge about biology and stock
status is no defense for a lack of action. In particular, very little is known about
annual migration routes, whether there are discrete local populations of basking
sharks, or the relationship between regional population abundance and global trends.
There is also a need to distinguish between the effects on population status of
1
climate change on the sharks’ environments and the legacy of the impact of fisheries,
and to ascertain whether stocks are depleted or recovering.
The latter concern is recognised in the UK basking shark BAP (English Nature,
1999), which states that improved long-term (many decades) monitoring of the UK
population is necessary to enable population trends to be identified. Though some
monitoring data for this species are available, most are based on sightings of sharks
feeding on plankton near the sea surface during spring and summer (April to
September) (Sims & Quayle, 1998; Marine Conservation Society, 2003), and no
analyses have been done to date that provide information on population trends.
Historic catches of basking shark have also been recorded by some fisheries
departments, including Norway, New Zealand, Ireland and Scotland. In particular,
catch data have been reported from “boom and bust” fisheries in Norway, Scotland
and Ireland, but it is unlikely that the resulting catch data reflect the status of the
regional stock as a whole. Currently, the fishery for basking sharks in the north
Atlantic is at such a low level that fishery-derived data are inadequate as a means of
monitoring population changes.
Three public sightings recording schemes for the basking shark are presently
underway in the UK: The Marine Conservation Society’s Basking Shark Watch
(Marine Conservation Society, 2003), Seaquest South-West and the Solway Shark
Watch and Sea Mammal Survey. All shark sightings are heavily dependent on
weather conditions because sharks even a short distance below the surface will
usually not be recorded. In these surveys, observer effort is concentrated in a few
surface-feeding “hotspots” but not usually recorded, neither is it standardised or
stratified with respect to population sampling requirements. Variation in numbers
sighted between areas and years cannot, therefore, reliably be attributed to changes
in population size.
In order to assess the status of basking shark stocks in the north-east Atlantic, and to
provide advice on appropriate conservation and management measures (should they
be considered necessary), more reliable population data are required, including
information on relative numbers of juveniles and adults, sex ratios, reproductive
biology, and distribution and seasonal migrations in relation to feeding grounds and
pupping areas. This knowledge, and the developed means of obtaining it, is also
relevant to basking shark populations that are exploited elsewhere, given the global
nature of the market for shark products.
1.1 Scope of the Proposed Study
In accordance with Defra’s requirements, this study aimed to elucidate elements of
the life history of the basking shark in the north-east Atlantic, especially that part of
the population that occurs at some time within UK territorial waters.
Modern satellite telemetry has been used to determine movements and behaviour of
individual sharks. The results will contribute towards our understanding of the
structure of basking shark stocks and inform decisions in relation to conservation
measures to ensure recovery (if necessary) and sustainability of basking shark
populations.
The work contributed to the 2002 UK Convention on International Trade in
Endangered Species (CITES) listing proposal and will also contributed to the
implementation of the UK Biodiversity Action Plan, the European Community Wildlife
Trade Regulations and the management of fisheries under the Common Fisheries
Policy, the EC shark action plan, and to the implementation of the Food and
2
Agriculture Organisation (FAO) International Plan of Action (IPOA) for the
conservation and management of Sharks.
1.2 Objectives
1. To determine movement patterns (including any migration), over-wintering areas,
and ‘site’ fidelity of basking sharks in EU waters using methods such as satellite
telemetry, tagging or other methods of tracking individual sharks.
2. To identify any areas that may be of special significance to basking sharks
because they are used for mating, wintering or by pregnant females.
3. To determine, as far as possible, the population structure and dynamics of
basking sharks in EU waters, especially whether populations in some areas form
discrete stocks or whether mixing occurs between populations further afield in the
north Atlantic or Mediterranean. (NB. much of this objective would have been
dependent on the molecular genetics analysis originally included in the project
tender, but ultimately excluded because of cost. This molecular genetics analysis
of basking shark populations is now being undertaken as part of GWD project CR
0288)
3
2. GENERAL METHODOLOGY
2.1 Previous methods
2.1.1 Population monitoring
Basking sharks are often sighted feeding or cruise swimming at the sea surface
during summer months in UK waters. This ‘basking’ behaviour allows counts to be
made in sightings surveys with the aim to determine numbers of animals using
particular areas. Ship-borne and aerial surveys have been conducted (Kenney et al.
1985; Sims et al. 1997) in addition to several public sightings schemes that record
basking sharks observed by members of the public either from coastal locations or at
sea (e.g. Marine Conservation Society, 2003). Although this type of monitoring
provides useful information on where basking sharks occur at the surface, it is
possible that the observed trends do not reliably indicate changes in population size
and/or distribution. Various surveys have counted basking sharks in local sea areas,
but no study to date has undertaken broad-scale surveys or determined biases (e.g.
sea state, observer effort, diving behaviour) that would enable relative abundance
measurements to be corrected for use in estimating absolute abundance, and hence
population size. Estimations of population size also require validation of the
assumption that a single population is sampled during surveys. However, many of
the availability bias factors (shark surfacing frequency, distribution of “basking”
sharks in relation to that of the whole population) still remain unknown for this
species. To determine stock structure, or whether subpopulations exist, and to
quantify biases that will affect population surveys, movement patterns of basking
sharks in northern European waters need to be described.
2.1.2 Behaviour
Prior to Priede’s pioneering work in the late 1970s (Priede, 1984), the movement
patterns of individual basking sharks had not been studied using electronic tracking
devices. Existing observations of basking shark behaviour were purely anecdotal and
contributed very little to our knowledge of their biology. Priede designed and built an
electronic tag that could be attached to free-ranging animals during their normal
activities and that was also capable of signalling to over-passing satellites. This
consisted of a platform terminal transmitter (PTT) that was compatible with Argos
receivers operating on 401.65 MHz frequency aboard US National Oceanic and
Atmospheric Administration (NOAA) satellites. The PTT was housed in a large float
that was connected to a shark via a 10-m long tether. This arrangement was
positively buoyant above 5 m depth and allowed the tag to break the surface and
begin transmitting to Argos receivers when the shark swam in the uppermost layer.
Geoposition of the tag when at the sea surface was calculated from the Doppler shift
of the transmitter’s frequency. It was hoped that this technique of ultra-high frequency
radio tracking of a single individual swimming near surface would provide the first
long-term remote tracking of a marine animal via polar-orbiting satellites. After
numerous attempts to attach the large tag on a shark, a 17-day track of a basking
shark moving within the Clyde Sea was obtained (Priede, 1984). The tracked shark
undertook a large circular route during this summer period close to the location of
productive thermal fronts. Until the current project, this remained the only basking
shark tracked by satellite telemetry.
2.1.3 New technologies
Unlike air-breathing marine vertebrates, fish have no need to return to the surface to
breathe between dives, and this presents a number of technical difficulties when
trying to track submerged fish compared to tracking air-breathers such as reptiles
and mammals. The main problem is that a transmitter’s antenna needs to be in the
air over extended periods during which time frequent transmissions (uplinks) to over-
4
passing Argos satellites can be made. Four species of sharks have been tracked with
varying degrees of success using tags in buoyant floats or attached directly to the
first dorsal fin (basking shark, Priede 1984; blue shark, F.G. Carey, unpublished data;
whale shark, Eckert & Stewart 2001; salmon shark, Block et al. 2003). However, if a
shark does not re-surface after initial tagging its position cannot be determined using
those same tags. The problem is, how can the horizontal movements of submerged
fish be tracked by satellite in the absence of regular opportunities for uplinks?
In the late 1990s, satellite transmitters that released from individual fish at preprogrammed times were developed and trialled successfully with bluefin tuna
Thunnus thynnus (Block et al. 1998). These so-called ‘pop-up’ tags enabled the
release position to be determined because when the small, buoyant tags floated to
the surface they would begin transmitting to Argos receivers (above) as the satellite
over-passed each tag. Clearly tags that signalled a fish’s position after a specific time
period were an improvement, but they did not provide any data on the intervening
positions of an animal between initial tagging and transmitter release.
The concept of determining an animal’s position anywhere on the surface of the
Earth by recording changes in the ambient light intensity as a function of real time
was first suggested by Wilson et al. (1992). Because the light intensity impinging on
particular places on the earth’s surface depends primarily on the time of year, time of
day and the locality, it is possible to determine position (geolocation) by calculating
the time of local midnight or midday to estimate longitude and day-length for latitude
estimation. A light-sensitive, datalogging (termed an archival) tag was developed and
subsequently used to study movements of wandering albatrosses in the Antarctic
Ocean (Weimerskirch & Wilson, 2000). However, these tags must be physically
recovered by researchers because there was no way of remotely downloading light
data from each tag. This method worked for albatrosses because they are long-lived
and highly philopatric, returning to the same nest sites for many years. Fish, on the
other hand, may not return to specific locations where tags can be removed. Hence,
the light geolocation idea was further developed (Hill, 1994) with the design of new
electronic devices that combined a light sensor for recording light intensity levels with
a pop-up satellite tag. This arrangement allows light-level data for geolocation
estimation to be accessed for the duration of deployment. By including pressure (to
give depth) and temperature sensors in such devices it is also possible to record
behaviour (swimming depth) and environmental (water temperature) data. These
tags have been used successfully to monitor the geographical movements of several
species of pelagic fish including blue fin tuna (Block et al., 2001) and white sharks
(Boustany et al., 2002).
2.2 Methodologies used
2.2.1 Satellite transmitters
In this project, the movements and behaviour of basking sharks were tracked with
pop-up archival transmitting (PAT) tags (length: 175 mm; mass in air, 76 g) (Figure
1). These tags combine a data-logger that records swimming depth, water
temperature and light level with an Argos-certified transmitter with 0.5 W power
output (Wildlife Computers, Redmond, WA, USA). Depth is measured to 1000 m
(min. resolution, 0.5 m), temperature from –40 to +60 oC (min. resolution, 0.05 oC),
and light level is measured as irradiance (W cm-2) at 550 nm wavelength. The PAT
tag archives depth, temperature and light-level data while being towed by a shark. At
a user-specified date and time, the tag actively corrodes the pin to which the tether is
attached, thus releasing it from the shark. The PAT tag then floats to the surface and
transmits summarised information collected over set intervals of 4 or 6 h via the
Argos system. The summary data transmitted comprises time-at-depth and time-at-
5
temperature histograms, temperature-depth profiles, and profiles giving the maximal
change in light intensity recorded over specified time periods. Argos also uses the
transmitted messages to provide the position of the tag at the time of surfacing. All
electronic tags were labelled with a contact address and notification of a reward
payable on return of the tag to the MBA.
Figure 1. Pop-up archival transmitting (PAT) tag.
Summary data
As described above, PAT tags summarise data collected from their depth,
temperature and light-level sensors into three types of messages. These messages
are stored in non-volatile flash memory for later transmission, or for download if the
tag is physically recovered. Any memory unused for the collection of these messages
is available for recording the depth, temperature or light-level readings at prespecified intervals. This is called the archival data that, if the tag is physically
recovered, provides a minute-by-minute dataset of swimming depth, temperature and
light intensity for the duration of tag deployment.
Histogram messages
Histogram messages comprise two parts, the first containing time-at-depth and the
second containing the time-at-temperature for the same period. These histograms
are constructed by counting the number of depth readings (taken at 1 minute
intervals) that fall within each of 12 pre-specified depth and 12 temperature ranges.
The depth range classes are set by the user including the time over which these
histograms are constructed (from 1 per h to 1 per 24 h).
Depth-temperature profile messages
The profiles of temperature at depth provide information on the water-column type
through which the PAT tag is passing, contingent on the shark diving through the
water column at least once in each collection period. Minimum and maximum
temperatures are recorded at each depth over the same time period used for
accumulating the histograms. Each message transmitted to Argos comprises of the
shallowest and the deepest measured depth and their associated minimum and
maximum temperatures, and six further depths between the shallowest and deepest
depth and their associated minimum and maximum temperatures. These
intermediate depths are dynamically chosen to be roughly equally spaced from the
shallowest to the deepest depths, and also to be generated for the depths that have
the greatest number of readings. The PAT tag temperature sensor has a fast
response so there is no appreciable lag in the measurement of temperature at any
depth.
6
Location messages
The location messages are used to help generate the track of each shark from
release to pop-up. In principle, the idea is to use changes in light level over time to
identify the times of local dawn and dusk from the light-level record that can then be
used to geolocate the tag’s position through time. In practice, the PAT tag processes
the light-level data to correct for depth, and identifies the times of dawn and dusk.
Sections of the light-level readings from dawn and dusk and the depths over which
the light-level data were collected form the location message. Hence, the dawn and
dusk curves for a given day are included in a single location message. The newest
version of the PAT tag (the Mk 4) also includes in the location message a
measurement of sea-surface temperature (SST).
Status messages
These messages provide diagnostic information on the PAT tag’s performance, for
example, the number of messages transmitted and battery voltage.
Decoding the messages
Decoding and display of the Argos dispose files (containing transmitted messages)
and data downloaded from a recovered tag are achieved principally using analysis
programs such as HexDecode, WC-Argos Message Processor, WC-Global Position
Estimator, PatDecoder, WC-Time Series Processor, Instrument Helper and
SATPAK2003. Additional routines were also used to reconstruct shark tracks (below)
because of the inherent problems associated with light geolocation using summary
light-level data from animals that undertake frequent dives.
2.2.2 Track reconstruction
The estimated accuracy of pop-up locations of PAT tags determined by Argos
receivers was between 350 and 1000 m. Satellite-retrieved data of daily maximal rate
of change in light intensity were used to estimate the local time of midnight or midday
for longitude calculations (Hill, 1994). The accuracy of these estimates is between ±
0.15 and 1.20o of longitude (1o longitude = 71.7 km) (Welch & Eveson, 1999; Block et
al. 2001). Only consecutive longitude estimates < 3o apart were used for geolocation.
Longitude estimates were discarded for days when the maximal light intensity
change was dive-induced, i.e. measured at times inconsistent with the regionally
expected times of dawn or dusk. Latitude was fixed using summary data. Daily
temperature–depth (TD) profiles from sharks, or histogram summaries of swimming
depth and water temperature were used to determine minimum and maximum sea
surface temperature (SST) on each day for which longitude was estimated. We then
used night time Advanced Very High Resolution Radiometer (AVHRR) remotesensing images of SST to fix latitude along the longitude (images were obtained from
the Natural Environment Research Council’s Remote Sensing and Data Analysis
Service based in Plymouth). A Java Applet Viewer (JAV) was used to determine SST
values for individual pixels on each false-colour image. Colour intensity in each pixel
(representing a 1 x 1 km area) was coded by a number from 0 - 256. By scanning
each image using the JAV’s cursor, the latitude representing the best fit of
temperature-scaled colour criteria to measured SST values was made along each
longitude estimate. Colour-criteria number values were converted to oC using a
calibrated scale on each image. If cloud cover was high on a particular day, remote
sensing SST images on the next closest day (before or after) were used to fix
latitude. Only images to a maximum of 5 days before or after were used. The latitude
estimate fixed using SST was then filtered for water-mass type, depth and swimspeed anomalies. TD profiles were used to determine water-mass type (stratified,
frontal, mixed). If the water mass at the shark’s location from SST images was
inconsistent with that determined from TD profiles, then further JAV scans on the
next available image were conducted. Similarly, if the maximum daily depth attained
7
by each shark was greater than the known depth at the estimated location shown on
oceanographic charts, and/or the distance between consecutive geolocations was
greater than 100 km per day straight-line travel speed (1.4o longitude) (Sims, 2000),
then the geolocation was considered anomalous and further JAV scans were
initiated. If a shark’s position could not be resolved using other daily SST images,
then the longitude estimate was discarded from the dataset.
The accuracy of latitude-position estimates varied over the range of standard
deviation per location of 0.01–1.56o latitude (1.11 – 173.52 km; n=31 locations
examined), which were similar error fields to those associated with longitude
estimation (Welch & Eveson, 1999; Block et al., 2001).
2.2.3 GSM telephone tags
The Argos system used to recover information form PAT tags (above) has a very
limited data transfer capacity and it is not possible to recover the archived minuteby-minute depth and temperature record. To overcome this limitation, CEFAS
engineers have been developing (as part of a separate Defra-funded project) an
archival tag that has the capacity to transmit data via the terrestrial cellular
telephone network if within a suitable range. The tag (Figure 2) consists of three
basic elements. The data logging circuit is based on the successful CEFAS Mk 3
data storage tag. Data transmission is achieved using a Siemens TC35 cellular
GSM (Global System for Mobile communication) telephone engine and relies on the
tag drifting close to shore in order to access the terrestrial cellular telephone
network. The pop-up function is achieved using an “electric match” that acts as the
primary release mechanism. The electronics are housed in a plastic pressure casing
that is equipped with external flotation sufficient to keep the tag at the sea surface
once it has detached from the shark.
Figure 2. Pop-up GSM
telephone tag with tether
and T-bar arrowhead.
(with a £1 coin for scale)
These tags record depth and temperature data every minute and then, at a predetermined date and time pop off, float to the sea surface and then drift in the
westerly wind-driven surface water current. Every five days the tag attempts to
establish radio contact with the terrestrial cellular telephone system. When contact is
established the tag downloads its archived data to a host computer in CEFAS at
Lowestoft. As with PAT tags, the GSM tags were labelled with a contact address
and notification of a reward payable on return of the tag to CEFAS.
8
2.2.4 Plankton analysis
To help interpret basking shark movements obtained from satellite telemetry we
related some of the movements to changes in ocean productivity. Basking sharks
feed selectively on high-density zooplankton patches dominated by large calanoid
copepods (Sims & Quayle, 1998; Sims, 1999). Although there is currently no way of
remotely sensing zooplankton abundance using satellites, the abundance and
distribution of phytoplankton (upon which many copepods feed) can be reliably
detected using the Sea-viewing Wide Field-of-view Scanner (SeaWiFS) on NOAA
satellites. SeaWiFS measures and records the reflectance/absorbance of the plant
photosynthesising pigment chlorophyll ‘a’ contained in all phytoplankton. These
values provide a proxy measure of the amount of phytoplankton in surface seawater
in all the oceans on a more or less daily basis.
We used SeaWiFS images in conjunction with Continuous Plankton Recorder (CPR)
data (Batten et al., 2003) of phytoplankton colour and copepod abundance to identify
whether phytoplankton productivity trends identified on a daily basis from satellites
were linked to similar changes in phytoplankton and copepod abundance measured
directly by the CPR.
The pattern of surface chlorophyll-a (chl-a) pigment concentration was determined for
a six month period (1 April – 30 September 2001) using daily SeaWiFS
(nasa_chlor_a) remote sensing images (see above). As with SST determinations, a
JAV was used to obtain precise number values corresponding to map-colour criteria,
which were converted to chl-a concentrations using the formula,
C = 10((n * 0.015) + log10(0.01))
where C is chl-a concentration in mg m-3 and n is pixel colour-criteria number (Sims
et al. 2003). At each geolocation 13 determinations were made: one at the estimated
location of the shark and three others at 0.2o intervals in each of four compass
directions (north, east, south, west) originating at the shark’s estimated location.
To estimate the long-term monthly chlorophyll and copepod abundance for the
Goban Spur region (48.0 - 50.0o N, 10.5 - 12.5o W), an area used by about half the
tracked sharks, we used CPR data. This plankton sampler, which is towed behind
merchant ships on their normal routes at about 7 m depth, filters plankton from the
water on a constantly moving band of silk of 270 µm mesh. Samples equivalent to 10
nautical miles (18.5 km) of tow and approximately 3 m3 of water filtered are analysed
under a microscope using a standard procedure (Batten et al., 2003). To estimate
chlorophyll, we used phytoplankton colour, a visual index of the intensity of
greenness observed on the silk and for total copepod abundance, we summed all the
counts for all copepod species found (developmental stages: copepodites, CI-CV;
adults, CVI). Data were from 1955-2000 and comprised 1094 samples.
2.2.5 Deployment of electronic tags
Shark surveys
Basking sharks for tagging were located by visual monitoring survey from 8.5 or 10m-long inshore research vessels during daylight hours. Line transects within a study
area were used to structure the search pattern. At least two observers scanned the
sea surface either side of the boat for basking shark dorsal fins at any one time.
Surveys were conducted only on relatively calm days when the wind speed was < 30
km h-1 (Beaufort scale wind force 4 or less) and sea state 3 or less. Surveys for
9
basking sharks were conducted in two main study areas. An area in the English
Channel off Plymouth (50o 20′ - 50o 10′ N, 003o 57′ - 004o 20′ W) was searched
between May and August each year 2001-2004, and an area comprising Lower Loch
Fyne and the northern Clyde Sea, Scotland (56o 00′ - 55o 35′ N, 004o 57′ - 005o 28′
W) was searched in July and August in 2001 and 2002. Two short, opportunistic
surveys were conducted off the Cornish coast in 2002: an area off Land’s End was
searched on 26 and 27 June and off St. Agnes on 18 July.
Satellite transmitters
The PAT tags were fully tested following delivery from the manufacturer in April 2001,
April 2002 and February 2004 and all found to be in good working order. Specifically,
the light, depth and temperature sensors were tested and specimen data collected
from each. Satellite time was applied for from Service Argos (Toulouse, France) and
each tag was provided with an Argos ID number under Dr Sims’ existing programme
no. 02026 ‘Migration of basking sharks’. The satellite link component of each tag was
tested for uplink frequency and accuracy when ID numbers were allocated.
Shark tagging
PAT tags were fitted with a 1.8-m long monofilament tether connected to a 40-mm
long stainless steel T-bar arrowhead via a wire trace (Figure 2). The PAT tag was
connected to the distal end with the monofilament line looped around the tag’s
corrosive link. An RD1500 quick-release mechanism was also fitted along the nylon
line near the tag to sever the line and release the tag should it be taken deeper than
the tag’s depth rating (~1500–1800m). Each PAT tag was programmed for optimal
data collection and storage and configured according to one of four release (‘pop-up’)
schedules: 2, 6, 9 and 12 months. Using a research vessel, we approached surfacefeeding sharks slowly from behind and at a distance of about 1 m to the side. When
the vessel’s bow drew level with the shark’s first dorsal fin, the arrowhead was
inserted rapidly through the middle of the fin using a rubber-band spear-gun pulled at
approximately 40 kg pressure (Figure 3). Attached PAT tags trailed behind sharks
close to the body, and the known distance from tag to the first dorsal fin (approx. 1.8
m) was used to obtain an estimate of total body length. After attachment and at a
pre-programmed time, each tag released from its host shark and floated to the
surface where it was geolocated by Argos satellites (the ‘pop-up’ location) and data
summaries transmitted (above).
Figure 3. Tagging a 5.5-m long basking
shark with a PAT tag in May 2001.
10
2.2.6 Summary of field surveys
2001 field season
Plymouth, Devon:
Systematic searches for basking sharks were conducted on 34 different days
between 2 May and 8 August 2001 in the Plymouth study area (~500 km2) (Figure 4).
In total, 246.4 h were spent searching in good weather conditions, during which 34
different individuals were observed. This gives a sightings per unit effort (SPUE) of
0.14 sharks per h-1. Five of these sharks were fitted with PAT tags on 24-25 May.
Clyde Sea, western Scotland:
A similar search strategy was employed in the Clyde Sea area, principally between
Arran and Lower Loch Fyne (Figure 4). Surveys were conducted on 10 days between
27 July and 5 August 2001 and during a total of 85.6 h spent searching, 9 different
basking sharks were sighted (SPUE of 0.11 sharks h-1). Five sharks were fitted with
PAT tags between 28 and 31 August.
Figure 4. The two study areas and
the sub-areas within each that were
searched for basking sharks.
2002 field season
Plymouth, Devon
Systematic searches for basking sharks were conducted on 15 different days
between 7 May and 25 June 2002 in the Plymouth study area. In total, 105.8 h were
spent searching during which 8 different individuals were observed (SPUE of 0.07
sharks per h-1). Three PAT tags were deployed on two basking sharks (Table 1), with
two tags on one shark programmed to release at different times (in October and
December 2002) in order to provide an interim geolocation to test the accuracy of the
methodology for reconstructing tracks.
Land’s End, Cornwall
Searches for basking sharks were conducted on the 26 and 27 June 2002. Search
time totalled 13.3 h during which time 10 different sharks were sighted (SPUE, 0.75
sharks h-1). Two of these individuals were tagged with PAT tags (Table 1).
11
Firth of Clyde and the Sound of Jura, Scotland
Searches were conducted throughout the Firth of Clyde, Loch Fyne and the Sound of
Jura. Surveys were undertaken on 13 days between 4 and 16 July 2002 and during a
total of 130.4 h spent searching, no basking sharks were sighted (SPUE of 0).
However, porpoises, seals and some minke whales were frequently observed. The
sea-time in Scotland was curtailed by one day so that the research team could return
to SW England where there had been reports of large numbers of basking sharks in
the previous 7 days.
St. Agnes, Cornwall
A day search (5.2 h) off the north coast of Cornwall was undertaken on 18 July
immediately after returning from Scotland. Twenty-three different basking sharks
were observed feeding within 1-3 km of the shore (SPUE of 4.42) six of which were
tagged with PAT tags (Table 1).
Overall, forty-one different basking sharks were sighted in a total search time of
254.6 h during the 2002 field season. Basking sharks between 4.0 and 7.0 m total
body length (LT) were tagged with PATs. Three females and one male were positively
identified among the 10 individuals tagged.
2003 field season
Plymouth, Devon
Daily searches for basking sharks were undertaken on 18 different days between 13
May and 25 June 2002. In total, 136.5 h were spent searching during which 45
different individuals were observed (SPUE of 0.33 sharks per h-1) (Figure 5). No PAT
tags were available to deploy on sharks this season and the new GSM tags (above)
became available at a time when basking sharks were no longer present at the sea
surface off Plymouth.
Re-sightings of individuals
Of the 45 sharks observed during surveys in 2003, 4 were re-sightings. The intervals
between sightings were 4, 7, 16 and 329 days. The latter was a 6-m long male shark
that was satellite tagged by us on 18 July 2002 off St. Agnes, north Cornwall. The
satellite tag, monofilament tether and wire tracer were not attached, but a small white
circular mark (indicating new connective tissue growth) was visible in the base of
right side of the first dorsal fin where the tag would had been attached. Unfortunately,
no data was retrieved from the satellite tag attached to this shark, presumably due to
complete tag failure. Using an established method for determining the total length of
free-swimming basking sharks (Sims et al., 1997) we estimated the total length of the
shark to have increased by approximately 0.5 m.
2004 field season
Plymouth, Devon:
Surveys to locate basking sharks were completed on 21 days between 14 May to 5
August 2004, with a total search time of 162.3 hours covering a total distance of
2,005 km. The total number of basking sharks sighted at the surface was 8,
amounting to an SPUE for the period of 0.049 sharks h-1.
Five of the basking sharks sighted between 1 and 26 June 2004 were each fitted
either with a single PAT tag (2 sharks) or GSM tag (3 sharks) (Table 1). Of the three
other sharks seen, one spent < 30s on the surface and could not be tagged, while
two dived just before we were within the 3-m maximum range necessary for
successful tagging.
12
Table 1. Details of the location and time of basking sharks tagged between 2001 and
2004. PAT tags are numbered from 01-01 to 04-02 and mobile telephone (GSM) tags
from 04-03 to 04-05.
Location
(longitude,
W)
Shark
body
length
(m)
50.294
4.110
4.5
♀
31/07/01
55.863
55.904
5.353
5.396
6.0
4.5
?
30/09/01
15/02/01
11:20
50.299
4.132
5.5
24/05/01
10:20
50.293
4.108
5.5
Plymouth
24/05/01
12:45
50.301
4.127
7.5
13263
Clyde
28/07/01
09:10
55.870
5.389
7.0
01-08
01-09
01-10
02-01
02-02
02-03
13264
13265
13266
13264
14996
15081
Clyde
Plymouth
Clyde
Cornwall
Cornwall
Plymouth
31/07/01
25/05/01
31/07/01
18/07/02
26/06/02
18/06/02
11:05
08:35
09:00
11:47
15:15
11:37
55.902
50.316
55.898
50.283
50.097
50.181
5.390
4.129
5.399
5.385
5.706
4.244
6.5
6.0
3.0
6.0
7.0
4.0
02-04
02-05
02-06
02-07
15096
15100
15101
15108
Cornwall
Cornwall
Cornwall
Plymouth
18/07/02
18/07/02
26/06/02
18/06/02
09:25
10:19
14:05
10:40
50.313
50.305
50.115
50.182
5.281
5.297
5.718
4.267
5.0
6.0
4.0
6.0
02-08
02-09
02-10
15088
15111
15124
Plymouth
Cornwall
Cornwall
18/06/02
18/07/02
18/07/02
11:00
10:25
11:36
50.184
50.304
50.283
4.261
5.301
5.385
Same as above
6.5
?
5.0
♀
02-11
06630
Cornwall
18/07/02
11:20
50.283
5.385
6.0
♂
15/04/03
04-01
04-02
04-03
04-04
04-05
49020
49021
-
Plymouth
Plymouth
Plymouth
Plymouth
Plymouth
01/06/04
08/06/04
02/06/04
03/06/04
21/06/04
09:47
08:26
13.44
14:50
08:57
50.269
50.353
50.185
50.189
50.269
4.030
4.142
4.283
4.288
4.071
4.5
3.0
5.5
3.0
2.5
?
?
?
?
?
19/08/04
06/10/04
170/7/04
13/07/04
29/06/04
PAT
tag
code
Argos
PTT
ID
Area
deployed
Date
deployed
Time
deployed
01-01
13254
Plymouth
24/05/01
10:50
01-02
01-03
13255
13256
Clyde
Clyde
28/07/01
31/07/01
08:20
09:50
01-04
13258
Plymouth
24/05/01
01-05
13259
Plymouth
01-06
13260
01-07
Location
(latitude,
N)
o
o
Sex
♂: male;
♀: female;
? : unknown
Intended/progra
mmed tag
release date
♂
♀
♀
♀
♀
?
?
?
?
?
♀
?
?
?
♀
01/05/02
07/12/01
20/02/02
05/01/02
25/07/02
30/11/01
15/03/02
30/09/02
15/11/02
30/11/02
31/10/02
30/11/02
01/12/02
15/10/02
01/12/02
31/12/02
15/01/03
2.2.7 Sex and maturity
Of the 25 basking sharks tagged during the project, the sex was determined for 10
individuals (2 males and 8 females). Further, only 3 of the 25 sharks were 7 m or
more in length (i.e. mature) suggesting that most (~23) of the sharks tagged were
either juvenile or sub-adult.
2.2.8 Animal Welfare
Tagging procedures were carried out in accordance with English Nature licence
numbers 20010014, 20020988 and 200441090 and Scottish Natural Heritage licence
numbers 2852 and 3460 under the Wildlife and Countryside Act 1981, for activities in
English and Scottish waters respectively. Tagging procedures were also carried out
under the Scientific (Animal Procedures) Act 1986, through a project licence PPL
30/1819 and personal licence PIL 30/3442 granted by the UK Home Office.
13
3. Results
3.1 Geographical Area
Figure 5 shows a map of the European continental shelf area with locations given
that are mentioned in the text in relation to basking shark movements.
Figure 5. Map of the European
shelf with place names.
3.2 Sightings
The sightings per unit effort (SPUE) for basking sharks during the four field seasons
of this study ranged from 0.33 sharks h-1 (2003) to 0.049 (2004). In 2004 the SPUE
value was much lower than in previous years (Figure 6) indicating that fewer sharks
were in the survey area off Plymouth, or that sharks spent a smaller proportion of
their time at the surface than normal, or some combination of these two factors.
Since sightings surveys could only be undertaken on relatively calm days weather is
unlikely to have affected SPUE in this study.
Figure. 6 Effort-corrected
sightings data of basking sharks
recorded during MBA surveys
(1996-2004) in the western
English Channel off Plymouth,
Land’s End and north Cornwall.
3.3 Length-frequency
The majority of basking sharks fitted with PAT or GSM tags were between 4 and 6 m
total length (Figure 7). The preponderance of 4 – 6-m long sharks tagged in the study
was broadly representative of the modal length class observed during surveys off
Plymouth since 1995 (Figure 8), suggesting the sharks we tagged were
representative of the size (age) structure of the sharks present annually in coastal
14
areas. Very few individuals over 7 m (i.e. reproductively mature) were observed
compared to the relatively high number of juveniles that were seen and tagged.
M
10
Frequency
8
6
Figure 7. Total body lengthfrequency of tagged basking
sharks (n = 25). Labelled arrows
denote estimated length at sexual
maturity for males (M) and females
(F) (Compagno, 1984).
F
4
2
0
<2
2 -3 3 -4
4 -5 5 -6 6 -7 7 -8
>8
Frequency
L e ng th c la s s (m )
M
90
80
70
60
50
40
30
20
10
0
Figure 8. Total body lengthfrequency of basking sharks
observed in surveys for which
body length could be estimated
accurately (n = 300; years 1995
– 2004; data from MBA Fish
Group surveys). Arrows as
above.
F
<2
2-3
3-4
4-5
5-6
6-7
7-8
>8
Length class (m)
3.4 Data recovery
3.4.1 Argos
Of the 23 PAT tags deployed during the project, a total of 10 tags transmitted data to
Argos receivers, although only 8 of these transmitted enough messages per satellite
overpass to enable pop-up location to be determined. Two tags transmitted few
histogram, temperature-depth profiles or light curves, so these particular tracks were
of low spatial and temporal resolution. The duration of tag deployment in the eight
sharks for which tracks were reconstructed ranged from 74 to 198 days and
amounted to a total of 1,013 days deployment. A summary of PAT tag performance
is given in Table 2. Overall, the success rate for recovering data from PAT tags via
Argos was 43% of tags deployed, but pop-up location determination was lower at
35%. The number of useable histograms, temperature-depth profiles and light curves
totalled 1,459.
Table 2. Summary of data recovery rates from PAT tags (24/05/01 to 03/11/04)
Tag
ID
Argos
PTT
ID
Pop-up
date
Days
at
liberty
01-01
01-05
01-07
01-09
02-01
02-07
02-08
02-04
04-01
04-02
13254
13259
13263
13265
13264
15108
15088
15096
49020
49021
07/08/01
07/12/01
05/01/02
07/12/01
30/09/02
15/10/02
01/12/02
31/10/02
19/08/04
06/10/04
77
198
162
197
74
120
7
105
80
120
Popup
positio
n (9 or
x)
9
9
9
9
9
9
X
9
9
X
Data
message
s
transmitte
d (N)
425
48
526
342
909
412
30
817
49
62
Uncorrupted messages (n)
Histograms (n)
Data lines (n)
Depth
Temp.
T-D
Light
profile
level
82
0
81
45
101
36
5
66
4
0
77
1
67
32
85
23
5
45
4
0
27
1
41
20
69
47
6
96
9
0
46
2
70
33
42
45
4
79
4
0
Sum of
uncorrupt
ed data
(Σn)
232
4
259
130
327
162
20
304
21
0
15
Percent
uncorrupt
ed data
((Σn/N)*1
00)
54.6
8.3
49.2
38.0
35.9
39.3
66.6
37.2
42.8
0
3.4.2 Mobile telephone network
Of the 3 GSM tags deployed in 2004 (Table 3), one (tag 04-04) released prematurely
due to failure of the tether attaching the tag to the shark. Fortuitously, this tag was
retrieved from the sea by a fisherman some 8 miles west of the tagging location a
week later and returned to the MBA laboratory where it was submerged in a
seawater tank. The “pop-off” mechanism activated correctly as programmed and the
tag transmitted dive data (recorded while the tag was attached to the shark) via the
terrestrial cellular telephone network to the CEFAS Laboratory in Lowestoft. The
other two tags were physically recovered, one from Elmscott beach (tag 04-03), north
Devon on the 26 July, and the other (tag 04-05) from Den Oever in Holland on 2nd
October. The tag recovered from Devon appeared to have become disabled during
final assembly and had therefore not recorded data. The tag recovered from Holland
had recorded depth for almost 3 days prior to a malfunction of the pressure sensor.
Temperature data was recorded for the entire deployment. Neither of these tags
transmitted data via the terrestrial cellular telephone network.
Table 3. Summary of data recovery from GSM tags (29/06/04 to 03/11/04)
Tag ID
04-03
04-04
04-05
Date deployed
02/06/04
03/06/04
21/06/04
Pop-up date
17/07/04
13/07/04
29/06/04
Date recovered
26/07/04
10/06/04
02/10/04
Data recovered
depth & temperature every min. for 7.9 h
depth & temperature every min. for 67.2 h
3.4.3 Tag recovery
The principal mode of data recovery from the electronic tags used in the project was
planned to be by radio transmission via satellites or via the terrestrial cellular
telephone network. However five PAT tags and all three GSM tags (n = 8; 31%) were
physically recovered by members of the public after tags were found floating (2
occasions) or washed up on beaches. The shortest time from tag deployment to
recovery was 73 days (tag 01-08) and the longest period was just under 3 years (tag
01-09, containing 197 d of data). Table 3 gives the summary information for all
recovered tags. Full archived datasets containing depth, temperature and light levels
sampled each minute for the entire duration of deployment were downloaded from all
PAT tags. This totalled approximately 2.4 million data points throughout 550 tracking
days. Due to technical problems, the level of data collection by GSM tags was limited
by comparison, though in two cases useful depth and temperature data were
downloaded successfully.
Table 4. Summary of electronic tags recovered.
Tag ID
01-01
01-08
01-09
01-10
02-08
04-03
04-04
04-05
Tag
type
PAT
PAT
PAT
PAT
PAT
GSM
GSM
GSM
Deployed
Date
24/05/01
31/07/01
25/05/01
31/07/01
18/06/02
02/06/04
03/06/04
21/06/04
Recovered
Location
Plymouth
Clyde
Plymouth
Clyde
Plymouth
Plymouth
Plymouth
Plymouth
Date
01/03/03
12/10/01
15/04/04
03/05/02
19/07/02
26/07/04
10/06/04
02/10/04
Data
collection
days
Location
Stornoway, Outer Hebrides
Broddick, Isle of Arran
Brighstone Bay, Isle of Wight
Lleyn Peninsula, North Wales
Perran Sands, Cornwall
Elmscott beach, North Devon
Off Looe, Cornwall
Den Oever, Holland
Full
archived
dataset
9
9
9
9
9
9
9
77
52
197
217
7
0.33
2.8
16
3.5 Horizontal movements
Pop-up locations of eight sharks (Figure 9; Table 4) showed that two tagged in May
and July moved from areas off south-west England to the Outer Hebrides whereas
one tagged in July off western Scotland moved to waters off the south-west of
England during tag deployments. Pop-up tags on three sharks tagged off Plymouth
and Cornwall apparently remained off south-west England in the Celtic Sea, whereas
tags on two sharks also tagged off Plymouth were located further east in the English
Channel and southern North Sea respectively. Horizontal trajectories were
successfully reconstructed for these 8 individuals. Summary data for the tracks
shows that minimum total distances covered ranged between 581 and 3,421 km for
tag deployment times of 74 and 198 days respectively (Table 4). Rates of movement
(ROM) ranged from 2.9 to 28.9 km per day (mean, 17.6 km d-1 ± 3.2 s.e.) although
ROM was positively correlated with the number of locations used to reconstruct the
track (log transformed ROM vs N locations; r = 0.46, P = 0.011). Comparison of ROM
for shark tracks having between 26 and 32 locations over periods ranging from 74 to
162 days, showed similar movement rates (range, 21.1 – 28.9 km d-1; mean, 25.4 km
d-1 ± 2.3 s.e.). Tag recovery locations obtained from reconstructed tracks were
broadly similar to the areas where tags had released according to satellite fixes
(Figure 9) and a track was reconstructed for one of these sharks.
(b)
(a)
(c)
(d)
Figure 9. (a) Number of electronic tags deployed in each area. (b) The PAT tag popup locations on the European continental shelf and (c) their dispersal directions and
distances from the point of deployment. (d) Locations where electronic tags were
17
physically recovered. Positions in b labelled with shark ID number and month of tag
release; d, positions labelled with shark ID number.
Table 4. Summary data for basking shark track reconstructions.
Tag
ID
01-01
01-05
01-07
01-09
01-10
02-01
02-04
02-07
Shark
total
length
(m)
4.5
5.5
7.0
6.0
3.0
6.0
5.0
6.0
Date
tagged
Tagging
location
Pop-up location
Track
days
N
locations
Minimum
distance
travelled
Rate of
movement
(km/d)
24/05/01
24/05/01
28/07/01
25/05/01
31/07/01
18/07/02
18/07/02
18/06/02
Plymouth
Plymouth
Clyde
Plymouth
Clyde
Cornwall
Cornwall
Plymouth
Outer Hebrides
off Suffolk
off north Cornwall
English Channel
Celtic Sea front
Celtic Sea front
Outer Hebrides
Celtic Sea front
77
198
162
197
217
74
105
120
10
3
26
8
59
32
28
24
1,878
581
3,421
1,616
3,201
1,937
3,034
1.667
24.4
2.9
21.1
8.2
14.8
26.2
28.9
13.9
3.5.1 Northward movements
Shark 01-01
Shark 01-01, a 4.5-m long female, travelled a minimum distance of 1,878 km in 77
days (24 May – 8 August) (Table 4). From the tagging area off Plymouth it moved
west and then northward before the PAT tag released and was geolocated by
satellites just south of the Hebrides in Scotland (Figure 10). The track reconstruction
for this individual showed that in early June, within 7-10 days of being tagged, this
shark was near the productive Ushant tidal front off Brittany, France. Daily vertical
temperature profiles confirmed the frontal habitat. Some 11 days later this shark was
geolocated 460 km west on the shelf edge (Goban Spur) where it remained for at
least 8 days before moving relatively rapidly north around the west coast of Ireland.
In making this apparent directed movement between 22 June and 16 July, this shark
covered a minimum distance of 780 km at a rate of 32.5 km d-1.
(a)
(b)
31 Oct
7 Aug
15 Aug
11 Jul
11 Aug
24 May
1 Aug
22 Jun
18 Jul
1 Jun
Figure 10. Reconstructed movements of (a) shark 01-01 and (b) 02-04 between
tagging locations (red circle) and tag pop-up positions (red cross).
18
Shark 02-04
This 5.0-m long individual was tagged off north Cornwall and travelled 3,201 km in
105 days between July and October 2002. In a similar manner to shark 01-01, this
individual also undertook a west then northward movement characterised by
relatively long time periods spent in the Celtic Sea and Hebridean Sea, between
which there was a rapid northward movement along the continental shelf edge
(Figure 9). After tagging this individual moved up into the Celtic Sea front area, then
between 19 July and 1 August followed a coastal route along southern Ireland before
being geolocated near the Goban Spur on the shelf edge. Only 15 days later the
shark was geolocated near the Outer Hebrides having moved a minimum distance of
834 km (ROM, 55.6 km d-1). The next 1.5 months were spent in an area south of the
Outer Hebrides. Shark 01-01 spent time in this area after the northward movement,
which suggests that this is an important region for feeding.
3.5.2 Southward movements
Shark 01-07
This 7.0-m long female was tagged in late July 2001 in Lower Loch Fyne, which lies
to the north of the Isle of Arran. It remained in the Clyde Sea during August before
moving south in mid-September (temperature range, 11.0-14.0o C) through the Irish
Sea and arriving in the southern Celtic Sea near the shelf edge in mid-October
(temperature range, 14.0-16.0o C). From mid-October 2001 to early January 2002
this shark remained off south-west England undertaking wide-ranging movements to
the shelf edge and into the western English Channel (Figure 11).
Shark 01-10
Shark 01-10 was one of the smallest sharks tagged in the study at 3.0 m total length.
Soon after tagging on 31 July 2001 it moved out of the Clyde Sea and spent August
near the Mull of Kintyre (Figure 11). Following a relatively rapid southward move
through the Irish Sea at the end of August, this juvenile spent about one month south
of the Celtic Sea front. From mid-October until early February 2002, the shark
occupied first the eastern and then the western sectors of the Celtic Sea front
between Pembroke, southwest Wales and Rosslare in Ireland. The movements
shown by sharks 01-07 and 01-10 demonstrate that some basking sharks occupy
shallow waters of the continental shelf and remain active in autumn and winter.
(a)
(b)
31 Jul
28 Jul
28 Aug
8 Feb
5 Jan
2 Oct
10 Oct
24 Nov
Figure 11. Reconstructed movements of (a) shark 01-07 and (b) 01-10 between
tagging locations (red circle) and tag pop-up positions (red cross).
19
3.5.3 Celtic Sea – Biscay movements
Shark 01-09
As with shark 01-01, shark 01-09 moved into the middle part of the English Channel
within 4 days of being tagged on 25 May 2001 (Figure 12). It remained close to the
Ushant front for the next month, and possibly for some time after that, because it was
not until mid-October that it was geolocated over very deep water on the shelf edge
south of Brittany in the Bay of Biscay. This 6.0-m long shark made ranging
movements in this area for the next 28 days until mid-November. Three weeks later
the PAT tag released and was located in the north-eastern end of the Hurd Deep, a
narrow, deep trough in the western English Channel. Shark 01-09 therefore appears
to have moved back into the English Channel during winter having spent the major
part of the summer, autumn and winter on the shelf edge in the Bay of Biscay. The
shark was undertaking vertical dives through the water column prior to tag release in
December in the eastern part of the Hurd Deep.
Shark 02-01
Shark 02-01 was estimated to be 6.0 m total length when tagged off north Cornwall in
mid July 2002. Following tagging this shark was geolocated making a relatively rapid
movement south around the tip of Land’s End between 21 and 23 July (Figure 12).
All of August was spent undertaking ranging movements within a fairly localised sea
area to the south of the Scilly Isles in the west to Whitsand Bay in the east. A more
rapid movement north into the Bristol Channel at the beginning of September was
once again followed by ranging behaviour, this time south of the Celtic Sea front, until
the PAT tag released off south-west Wales on 30 September.
Shark 02-07
This 6.0-m long female shark was tagged off Plymouth on 18 June 2002. Over the
next month it ranged south of the south-west peninsula of England before moving
around Land’s End and remaining in a restricted sea area off the northwest Cornish
coast (Figure 12). Towards the end of August this shark moved north into the eastern
sector of the Celtic Sea front off south-west Wales, remaining there until the
beginning of October when she commenced a westwards movement along the
southern Irish coast just prior to PAT tag release.
(a)
(b)
30 Sep
25 May
7 Dec
4 Sep
18 Jul
19 Aug
7 Nov
27 Oct
20
(c)
Figure 12. Reconstructed movements of
(a) shark 01-09 and (b) 02-01 and (c) 0207 between tagging locations (red circle)
and tag pop-up positions (red cross).
23 Sep
15 Oct
18 Jun
2 Jul
The movements of sharks 01-09, 02-01 and 02-07 demonstrate that over seasonal
time scales basking sharks may remain within relatively localised areas of the UK
continental shelf.
3.5.4 Eastwards movement
Shark 01-05
Shark 01-05, a 5.5 m female, was the only shark that was tracked moving east up the
English Channel after tagging on 24 May 2001 (Figure 13). This shark’s PAT tag
released and was geolocated off the Suffolk coast on 7 December. However, due to
the few messages transmitted via satellite it is not possible to identify whether the tag
was carried into the North Sea by the shark or whether it released prematurely and
was subsequently transported by prevailing wind-driven currents and tides through
the Dover Strait. The only light geolocation obtained showed the tag to be still
attached to the shark on 18 July when it was in mid-Channel (Figure 13), although
whether it released prematurely after this time cannot be determined. One of the
GSM tags attached to a shark off Plymouth in June 2004 released prematurely and
was found on a beach in the Wadden See, Netherlands, some few months later. This
demonstrates how floating tags can move long distances eastwards in quite short
timescales and so eastwards movements of basking sharks into the southern North
Sea may be rarer than the tracking data might suggest.
Figure 13. Reconstructed movements of
shark 01-05 between tagging locations (red
circle) and tag pop-up positions (red cross).
7 Dec
24 May
18 Jul
21
3.5.5 General movement patterns
The reconstructed tracks of basking sharks have suggested patterns of movement on
the European continental shelf between centres of high productivity. Sharks spent
most of the time in shelf sea areas characterised by tidal fronts and fronts associated
with the shelf break. It was noticeable that individuals foraging along fronts off the
south-west peninsula of England moved to three main areas: the Celtic Sea front, the
Goban Spur and the Biscay regions of the shelf edge (Figure 14a,b). Two individuals
moved northward from these areas along the shelf edge into rich feeding areas in the
Hebridean Sea that are also characterised by strong tidal fronts that aggregate
zooplankton (Le Fevre 1986). Interestingly, no sharks tagged off south west England
in spring moved northwards through the Irish Sea during summer, whereas those
sharks tagged in the fjord-like Clyde Sea in summer travelled relatively rapidly south
through the Irish Sea in late summer and early autumn to areas off south-west
England. Overall, it appears that during summer basking sharks were moving
between centres of high zooplankton abundance in fronts, tending towards a
northward movement of variable distance (from Plymouth to the Celtic Sea or
Hebridean Sea). In winter, some of these individuals remained or moved into
shallow coastal waters, generally in the southern region of the shelf (Figure 14b).
These results suggest that there are no separate populations of basking sharks
inhabiting northern or southern UK waters, but, rather, individuals move freely
between these areas and probably form a single population.
Fronts
(a)
(b)
Fig 14. (a) Tidal and shelf break fronts (red lines) on the European shelf and (b) the
generalised movement patterns of “tracked” basking sharks between these
productive regions. Remote sensing image in a is a monthly composite of sea
surface temperature during August 2002 from AVHRR on NOAA satellites.
Shelf-edge movements
To investigate the hypothesis that basking sharks not only utilise inshore tidal fronts
for summer foraging but also shelf-edge areas (locations with >200 m water depth)
the relationship between broad-scale plankton abundance and foraging movements
was determined. Trends in surface primary production (chlorophyll a pigment
concentration) were calculated at each position of shark 01-01 over a six-month
period (1 April – 30 September 2001) using SeaWiFS remote-sensing images (e.g.
Figure 15). The shelf-edge positions occupied by shark 01-01 on 14 and 17 June
22
coincided with the highest levels of surface primary production seen at these
locations for 6 months (between 2 and 67 mg m-3 compared to a background level of
<1 mg m-3), and higher than those encountered previously by the shark off north-east
France (Ushant front) (Figure 15). A SeaWiFS chl-a image from 19 June 2001
indicates that in mid June shark 01-01 was associated with a discrete productivity
‘hotspot’ over the shelf edge (Figure 14). By contrast, prior to this shark moving north
on or after 22 June, the level of primary production around its position had declined
to about 0.5 mg m-3 (Figure 16).
11
10
9
8
6o W
7
50
Western English
Channel
49
Figure 15. Reconstructed movements of
shark 01-01 between an area of high
surface plankton abundance off Brittany to
a primary production hotspot located near
the Goban Spur on the shelf-edge.
48
Shelf edge
47oN
(a)
20
(b)
1.
Shark
present
15
3 June
5
3
4
0
20
2
14 June
2000
5
0
20
3.
15
17 June
10
5
0
20
4.
15
22 June
10
5
2.0
1800
Copepod abundance
1.8
1600
Phytoplankton colour
1.6
1400
1.4
1200
1.2
1000
1.0
800
0.8
600
0.6
400
0.4
200
0.2
0
00
1
10
/2
00
1
09
/2
0.0
Jan
Feb Mar Apr May Jun
Jul
Aug Sep
Oct Nov Dec
16
/
09
/2
00
1
26
/
06
/
07
/2
00
1
08
/2
00
1
17
/
07
/2
00
1
28
/
06
/2
00
1
08
/
18
/
05
/2
00
1
05
/2
00
1
29
/
04
/2
00
1
09
/
19
/
03
/2
00
1
0
Phytoplankton colour
10
30
/
1
2.
15
Copepod abundance (no. per sample)
Chlorophyll-a concentration (mg m-3)
10
Figure 16. Trends in chlorophyll-a pigment concentration at (a) four positions occupied by shark 01-01
between 3 and 22 June (numbered panels 1-4) determined from SeaWiFS satellite images. Six-month
period in each panel extends from 1 April to 30 September 2001. Grey vertical lines denote the time
when shark 01-01 was present at each location during the six-month period. Two high chl-a values (41
-3
and 67 mg m ) that occurred immediately prior to the shark’s presence at that location were omitted
from panel 3 for clarity. (b) Mean monthly phytoplankton colour and total copepods measured by the
CPR survey in the Goban Spur region (shown on inset map) between 1955 and 2000. Numbered
locations on inset map (1-4) refer to positions examined for phytoplankton concentrations in panels 1-4
in (a). Error bars on mean monthly values denote one S.E.M. Arrow denotes when shark 01-01 was
present in the Goban Spur region.
23
Patterns of phytoplankton and total copepod abundances measured by the
Continuous Plankton Recorder (CPR) survey over the same area show high
congruence (Figure 16), indicating that peaks in primary and secondary production in
this region are tightly coupled temporally (Sims et al., 2003). Basking shark 01-01
arrived in the Goban Spur region at a time when total copepod abundance peaked
(Figure 16). Decreased numbers of copepods in July at the Goban Spur coincided
with the relatively fast northward movement of shark 01-01 into areas where copepod
abundance peaked in August.
3.5.6 Occurrences of sharks in different sea areas
Four tagged sharks moved through and/or occupied areas within historical basking
shark fishing grounds (Figure 17). Sharks 01-07 and 01-10 spent short periods posttagging in the Firth of Clyde and Hebrides and Minch hunting grounds prior to making
sustained southerly movements. In contrast, shark 01-01 headed north from the
western English Channel via the Goban Spur through the Achill and Sunfish Bank
area, before remaining near or within the Hebrides and Minch fishing grounds for
some weeks. Similarly, shark 02-04 spent an extended period during autumn 2002
within the Hebrides and Minch grounds having first been located and tagged in July
off north Cornwall (Figure 10b). This demonstrates that sharks observed at the
surface long distances away from historical fishing grounds are capable of travelling
to those areas within relatively short time periods, suggesting that sharks occupying
(albeit temporarily) these historical fishing grounds may also move through the entire
European shelf area.
Figure 16. Map showing historical
basking shark fishing areas (in dark blue;
adapted from MCS, 2003) and the 12 nm
limit of UK territorial waters (black line).
We have delineated the seas of the European continental shelf into three zones: U.K.
territorial waters (≤ 12 nm ), the European Exclusion Zone (> 12 nm, ≤ 200 nm), and
International waters (> 200 nm). Table 5 shows the occupancy time of seven basking
sharks within each of these areas and demonstrates individual sharks fall into two
main groups. Sharks 01-01, 01-09 and 01-07 spent most time (67 – 87%) in
international waters and relatively little time close to the UK coast. In contrast, sharks
01-10, 02-04, 02-01 and 02-07 spent between 30 and 58% in territorial waters,
24
between 27 and 69% in the 200 nm zone and less than 27% outside the 200 nm
zone.
Table 5. Occupancy times (%) of basking sharks within different sea areas.
Shark #
01-01
01-09
01-07
01-10
02-04
02-01
02-07
Within 12 nm
5.3
5.8
13.0
29.9
34.6
30.2
57.8
12 – 200 nm
10.5
7.0
20.1
46.1
38.7
68.7
26.6
> 200 nm
84.0
87.0
66.8
24.0
26.6
1.0
15.5
Although Table 5 shows differences as well as similarities in occupancy times
between different sea areas used by individual sharks, analysis of occupancy times
by season indicates a general trend (Table 6). It appears that the sharks occupied
areas on the European shelf within the 12 and 200 nm limits marginally more during
spring, summer and autumn than they did during winter. This pattern of occupancy
should enable them to take advantage of the most productive areas located in
relatively shallower waters of the shelf during warmer months.
Table 6. Mean occupancy times (%) of basking sharks by season.
Within 12 nm
12 – 200 nm
> 200 nm
n sharks
Winter
15
21
64
4
Spring
20
44
36
3
Summer
30
26
45
7
Autumn
17
36
47
6
These results show widely different movement patterns between sharks that were
initially located and tagged in the same area. This may reflect the variability of
basking shark movements as they search for their zooplankton prey which is patchily
distributed at a very wide range of spatial scales (Sims & Quayle, 1998; Sims & Reid,
2002) which, in turn, may lead to significant changes in shark distribution between
years.
3.6 Vertical movements
3.6.1 Summer behaviour
Shark 01-01
This shark spent between 49% (night) and 71% (day) of the time in the uppermost 10
m of water in the western English Channel in early June, although short-term dives to
between 90 and 120 m depth did occur, mainly at night (Figure 18). Near the shelf
edge, this shark spent progressively less time in the uppermost 10 m during the day
(39% decreasing to 12%), whilst more time was spent at depths below 90 m (Figure
18b, c), increasing in maximum depth from 120 to 180 m when near the shelf edge
(Figure 18a). The temperature-depth profile indicates that this shark dived offshore in
strongly stratified water masses with a vertical temperature range of approximately 9
to 15 oC (Figure 18a). When this shark moved into shallower shelf waters off western
Scotland in July (Figure 18d), it resumed a daytime shallow-diving pattern similar to
that seen in the English Channel in May and June.
25
May
0.8
(e) E
0.6
0.4
0.2
1.0
Proportion of time
1.0
0.0
1.0
June – 1 st half
0.8
(f)
0.6
0.4
0.2
0.0
June – 2 nd half
0.8
0.6
0.4
0.2
0.0
0.2
August
September
0.8
0.6
0.4
0.2
October
November
0.8
0.6
0.4
0.2
0.6
-1
0
0.8
0
July
5 0 50
.5
10 100
0.5
15 150
0.5
25 250
0.
5
35 350
0.5
45 450
0.5
55 550
0.
5
65 650
0.5
75 - 75
0.5
0
-1
00
0
0.0
1.0
Proportion of time
(d) D
0.4
0.0
1.0
(g) G
Proportion of time
1.0
Proportion of time
(c) C
0.6
0.0
1.0
F
Proportion of time
Proportion of time
(b) B
June - English Channel
July - Shelf edge
0.8
10
.5
Proportion of time
(a) A
Depth (m)
0.4
0.2
0
90
12 120
010
00
0
0
-9
80
0
-7
-6
-8
70
60
0
-4
-5
0
50
40
0
-2
-3
0
30
20
10
0-
10
0.0
Depth (m)
Figure 17. Depth preferences of sharks 01-01 (a-d) and 01-09 (e-g) over summer
and autumn in the northeast Atlantic. Columns in (a-d) represent depths preferred by
shark 01-01 during the day (white) and night (black). Columns and error bars on all
panels denote the mean ± 1 S.E.M.
Shark 01-09
This shark utilised frontal habitat in the western English Channel during June before
moving out to the shelf edge, a behaviour pattern also shown by shark 01-01. This
change in summer habitat from shelf to shelf edge-frontal regions also reflected a
shift from surface feeding in June (75% of time spent in 0 to 50 m depth), to deeper
vertical habitat selection during July and August on the shelf edge (modal time-atdepth: 100-150 and 150-250 m). There were also occasional dives to between 350
and 650 m depth (Figure 18e, f). This shark occupied shelf-edge habitat in the Bay of
Biscay (Fig. 12a) that was further south than areas selected by shark 01-01. The
Biscay region was characterised by warmer surface waters (16–17 oC) although
shark 01-09 continued to undertake frequent deep dives into cooler water (10–13 oC)
(Figure 18d).
Sharks 02-01 and 02-07
These sharks conducted similar horizontal movement patterns during the summer of
2002 (Figure 19). Both spent July and August off Land’s End, Cornwall before
moving to the Celtic Sea front where they stayed during September and October.
Around Land’s End, they ranged vertically between the surface and over 100 m
depth with most time spent either in the top 50 m (72% of the time; shark 02-07) or
between 10 and 50 m (76% of the time; shark 02-01) (Figure 19). After moving to the
Celtic Sea front both sharks spent more time nearer the surface with > 95 % of time
in the uppermost 10 m. This level of congruence in diving behaviour between two
individuals occupying the same habitats at similar times indicates they were probably
26
responding to the same vertical changes in abundance of zooplankton: prey was
evidently deeper off Land’s End than near the Celtic Sea front.
(a)
Figure 18. Depth preferences of sharks (a)
02-01 and (b) 02-07 when resident off
Land’s End during July-August and in the
Celtic Sea front during September-October
2002.
(b)
Shark 02-04
From August to October this shark spent between 41% and 90% of time in the top 10
m (Figure 20). During this period it travelled from the western English Channel
around the west of Ireland to the Hebridean shelf. There was evidence of some short
dives to very deep depths between 200 and 850 m during October when it was off
the shelf edge, however, this individual clearly remained in the upper layers when
ranging on the shelf and during its long distance movement north.
Figure 19. Depth preferences of shark
02-04 from July to October during a
northward movement from the English
Channel to the Outer Hebrides.
3.6.2 Autumn/winter behaviour
Shark 01-09
From mid-October to mid-November shark 01-09 was geolocated on the shelf edge
in the Bay of Biscay where possibly it had remained since the summer. It moved
through the water column between 10 and 250 m on a regular basis, with some dives
to between 350 and 450 m (Figure 18f). One dive in November apparently extended
27
from the surface down to between 750 and 1000 m depth at the mesopelagicbathypelagic boundary (Figure 18g).
Between 17 November and 7 December this shark moved back into the western
English Channel and conducted similar deep-diving behaviour in the Hurd Deep
(maximum dive depth = 176 m). At fine time-scales we also found this shark was
active during 6 h time periods on consecutive days in early December, ranging
between 50 and 250 m depth.
7
Shark 01-07
In November and December this female spent 80-90% of its time between 50 and
100 m depth, although it occasionally went shallower and dived from the surface to
between 250 and 350 m depth (Figure 21c). Large amplitude vertical movements
were less frequent than during summer because the maximum dive depth became
progressively shallower (~90 to 50 m) concomitant with movements into inshore
areas with a fairly constant temperature of approximately 11oC. In early January
2002 this shark also showed vertical movements on consecutive days ranging from
between 0-10 and to 50-100 m indicating this individual was also actively diving
during winter months (Figure 21d). However, even though basking sharks were
active during winter they generally spent less time near the surface compared to
depths preferred in summer months (Figure 21a-d).
Proportion of time
Jul - Clyde Sea
Aug
0.8
0.6
0.4
0.2
1 Jan (12:00-18:00 h)
2 Jan (00:00-06:00 h)
3 Jan (18:00-00:00 h)
4 Jan (00:00-06:00 h)
0.6
0.4
0.2
0 - 10
10.5 - 50
50.5 - 100
Depth (m)
1.0
Sep - Irish Sea
Depth (m)
Oct - Celtic Sea
0.8
Proportion of time
0.8
0.0
0.0
(b)
1.0
(d)
Proportion of time
1.0
(a)
0.6
0.4
0.2
0.0
1.0
(c)
Nov - Shelf edge
Dec - English Channel
0.6
0.4
0.2
10
.5
50 50
.5
10 10
0
0.
5
15 150
0.
5
25 250
0.
5
35 35
0
0.
5
45 450
0.
5
55 55
0
0.
5
65 650
0.
5
75 - 7
50
0.
5
-1
00
0
-1
0
0.0
0
Proportion of time
0.8
Figure 20. Depth preferences of shark 0107 over (a) summer, (b) autumn and (c,d)
winter in the northeast Atlantic. Columns
and error bars on all panels denote the
mean ± 1 S.E.M.
28
Shark 01-10
This juvenile shark moved from the Clyde Sea to the Celtic Sea front via the Irish Sea
between August 2001 and March 2002. Its depth preferences over this period show a
marked shift from shallow depths selected in summer (0 – 10 m) to depths between
50 and 75 m in winter (Figure 22). Over 70% of this shark’s time was spent in the
uppermost 10 m during August, decreasing to only 17 and 6% in December/January
and February/March respectively. This shift in vertical habitat selection occurred
during early November.
Overall, the depth data for sharks during winter shows that there is a general
preference for deeper depths in the water column (or deeper water) on the shelf
during winter compared with summer, a shift that at least in part explains why
basking sharks are rarely observed at this time.
Figure 21. Depth preferences of shark
01-10 show a gradual shift to deeper
habitat with the onset of winter.
3.6.3 Daily dive patterns
Using archival data downloaded from tags recovered from four basking sharks (nos.
01-01, 01-08, 01-09 & 02-08; total length range, 4.5–6.0 m) daily dive patterns in
different habitats were examined. The data represented a total of 333 tracking days
between May 2001 and June 2002. Two sharks exhibited daily diving behaviour
characterised by dusk ascent into shallow waters from greater depths occupied
during the day; so-called normal “diel vertical migration” (DVM). Sharks 01-01 and
01-08 exhibited normal DVM (dusk ascent-dawn descent) when occupying deep
water of the fjord-like Clyde Sea (maximum water depth, 200 m; Figure 23a) and
when located on the European shelf edge (max. depth, 1000 m; Figure 23b)
respectively. The mean daytime depths (72.5 & 82.5 m) selected by sharks 01-01
and 01-08 respectively were significantly deeper than depths selected at night (29.9
& 12.8 m) (Mann-Whitney U-tests with normal approximation: shark A, median
daytime depth = 72.5 m, median night-time depth = 29.9 m, Z = 29.1, P < 0.0001;
shark B, median daytime depth = 87.2 m, median night-time depth = 12.8 m, Z =
62.41, P < 0.0001). The deep-water habitats occupied by these sharks were strongly
stratified thermally, with maximum vertical gradients of 5.4oC (0–92 m depth) and
4.4oC (0–128 m depth) respectively (Figure 24 a, b). Track reconstructions showed
that specific patterns of DVM were associated with particular ocean habitat because,
by contrast, sharks 01-09 and 02-08 showed reverse DVM (nocturnal descent) in
inner-shelf areas of the western English Channel (< 80 m depth) (Figure 23c, d).
Median night time depths of 19.1 and 54.9 m selected by sharks 01-09 and 02-08
respectively were significantly deeper than depths selected during the day (4.1 & 5.6
m) (Mann-Whitney U-tests with normal approximation: shark C, median daytime
29
depth = 4.1 m, median night-time depth = 19.1 m, Z = 28.0, P < 0.0001; shark D,
median daytime depth = 5.6 m, median night-time depth = 54.9 m, Z = 25.6, P <
0.0001). The inner-shelf habitats occupied by sharks 01-09 and 02-08 were
characteristic of tidal fronts with shallow temperature gradients, with this change
occurring mostly in the upper 20 m. Maximum gradients encountered by sharks were
2.6oC (shark 01-09) and 0.8oC (shark 02-08) between 0 and 64 m depth (Figure 24c,
d).
0
A
20
40
60
80
100
120
140
0
B
20
A
40
C
Depth (m)
60
80
100
120
140
UK
0
20
40
60
B
80
100
C
120
D
140
0
D
France
20
40
60
80
100
120
140
Time
Figure 22. Diel vertical changes in swimming depths of four basking sharks (A, 01-08; B, 0101; C, 02-08; D, 01-09) in relation to thermal habitat occupied on the European continental
shelf (areas A-D on the false-colour sea surface temperature (SST) remote sensing image).
Sharks 01-08 and 01-01 followed a normal pattern (nocturnal ascent) in thermally stratified
water masses, whereas sharks 02-08 and 01-09 showed a reverse pattern (nocturnal
descent) in frontal waters. Periods of diving behaviour shown: shark 01-08 (6 m total length,
LT), 12–15 August 2001; shark 01-01 (4.5 m LT), 12–15 June 2001; shark 02-08 (6m LT), 18–
21 June 2002; shark 01-09 (6m LT), 5–8 June 2001. The black bars on panels A-D denote
night-times between dawn and dusk. The colour scale bar on the SST map is surface water
temperature in oC as derived from the advanced very high resolution radiometer aboard
NOAA satellites. The dotted line on the SST map denotes the spring-summer position of the
Ushant thermal front. Continous lines represent the 200 and 2,000-m isobaths.
Sea temperature (oC)
8
10
12
14
16
8
10
12
14
16
0
20
40
60
80
Depth (m)
100
120
A
B
C
D
140
0
20
40
60
80
Figure 23. Vertical sea-temperature profiles
showing thermally stratified waters occupied by
sharks A (01-08) and B (01-01) and transitional
(frontal) zones occupied by sharks C (02-08)
and D (01-09). Solid lines are maximum daily
temperatures and dotted lines denote minimum
daily temperatures. The grey lines on panels C
and D denotes the seabed depth.
100
120
140
30
3.6.4 Surfacing frequency
Surfacing frequency and sightings
per unit effort (ind. h-1)
The existence of normal and reverse DVM patterns in individual sharks from the
same population present in UK waters resulted in very different surfacing frequencies
(time spent at ≤ 1 m depth). The daytime-surfacing frequency of a tracked individual
(02-08) feeding in an inner-shelf area near a front was over 100 times higher than
another shark feeding in well-stratified water (01-08) (Figure 24). This large
difference in ‘basking’ behaviour between regions was reflected in our survey data:
11.5 times more sharks per unit effort were observed in frontal areas than in stratified
water (Figure 25).
0.7
Surfacing frequency
0.6
SPUE (ind./h)
0.5
0.4
0.3
0.2
0.1
0
Front
Stratified
Figure 24. The surfacing frequency (black bars) of sharks 01-08 (stratified) and 02-08
(front) during daylight hours compared with sightings of sharks per unit time (SPUE,
individuals h-1) (white bars) in each of these tracking locations.
4. Conclusions
4.1 The technical approach
The technology used to track the movements and behaviour of basking sharks in this
project represent the best available instrumentation to enable remote tracking of fish
at sea. The PAT tags used only became available for field deployment in 2000, thus
when deployed by us in 2001 there had been few other operational deployments
anywhere in the world. As with all new, cutting-edge technology the failure rate is
initially high; this is especially relevant to long-term deployments of electronic tags in
the sea. In the current study we found 43% (10/23) of PAT tags successfully released
from sharks and uplinked to Argos receivers. Of these, 4 were weakly transmitting
tags and either no or very few messages were received. Because of this, the
geolocation of 2 tags could not be determined by Argos receivers.
Although the pop-up performance of PAT tags was relatively low compared to
success rates reported for position-only, pop-up tags in other studies (95%, Block et
al., 1998; 79%, Sedberry & Loefer 2001), it is comparable to other recent studies
using PAT tags on large fish species. For example, 48% of PAT tags attached to
bluefin tuna in the Mediterranean successfully uplinked (G. Arnold, CEFAS
Lowestoft, pers. comm.), which is similar to ~50% success rate found for white
sharks off South Africa (R. Bonfil, WCS, pers. comm.). This suggests that the
performance of the more sophisticated PAT tags is some 30-40% lower than
position-only pop-up tags.
31
The reasons why some PAT tags did not uplink to receivers remains unknown for the
majority of tags. However during the study 2 tags (01-08 & 01-10) that failed to
achieve uplinks were found washed up on beaches on the Isle of Arran, Scotland and
the Lleyn Peninsula, North Wales, respectively (Table 3). Examinations of the tags
showed 01-08 released prematurely due to tether failure prior to the programmed
release time, whereas tag 01-10 released from the shark at the programmed time,
but failed to uplink with satellites. This shows that not only do some tags transmit
only weak signals when floating at the surface, but some even when reaching the
surface do not apparently transmit at all.
There were 6 PAT tags that were geolocated by Argos receivers and continued
transmitting large numbers of messages for up to 14 days. Significant proportions of
these however, could not be decoded. Between 45 and 64% of all messages
received were corrupted, that is, messages contained erroneous checksums
indicating individual bytes of data deviated from their original encoded value. It
seems plausible that these messages became corrupted during transmission
possibly due to physical environmental effects on frequency stability.
The approach of reconstructing movements of sharks using light-level and seasurface temperature data is heavily reliant upon receiving data of appropriate quality
and temporal coverage. We were able to reconstruct meaningful tracks for 7
individual sharks using 6 satellite-retrieved datasets and 1 archival dataset because
datasets were extensive. Having a large dataset means that rigorous filtering of the
data is possible; for example, longitude data can be filtered for anomalous positions
(those on land or two consecutive positions >3o apart) (Block et al. 2001), and large
anounts of SST data enables ground truthing of estimated positions from highresolution remote sensing images of SST. Calibration tests on tags at set moorings
show the accuracy of estimating longitude using light-level geolocation range from
0.15 – 1.2o (Welch & Eveson 1999).
Estimating latitude using light-level is more prone to error because small differences
(mins) in estimating the timing of dawn and dusk (daylength) results in large
differences in latitude (~2-5o). We fixed latitude using SST data by matching tagderived SST with remote sensing satellite-derived SST. Final position was fixed by
passing equally plausible locations through filters that detected anomalies in watermass type, depth and swim speed (furthest possible distance covered per unit time).
The accuracy of this method of latitude-position estimates varied over the range of
standard deviation per location 0.01 – 1.56o longitude (Sims et al., 2003), which were
similar error fields to those associated with longitude estimation. As part of this work
we developed a latitude-by-SST fixing software program that we will make freely
available to other research groups following appropriate publication.
Overall, the light-level and SST geolocation methods used in this project to estimate
positions and thus reconstruct movements of basking sharks are largely insensitive
to movement detection below 10 km at best and below 70 – 170 km at worst.
Therefore, meaningful interpretation of finer scale movement trajectories of sharks
from PAT tags is not possible below these error limits. Whilst the grouping of
consecutive geolocations of an individual in a particular region may not enable us to
say anything certain about the number and direction of small-scale (<50 km) step
lengths, the information is clearly valuable biologically. Data of this type provide
useful information on the amount of time an individual spends in a large area, such
as the western English Channel for example. It also enables determination of
residence times and when individuals leave particular areas because the larger-scale
accuracy of position estimation using the methods described is high.
32
4.2 Summary of results
The results of the current study provide a wealth of new information, significantly
adding to our understanding of the movements and behaviour of basking sharks
presented earlier in the project in the review of their biology and ecology (Appendix
I). Until this study only a single basking shark had been tracked for a limited period
by satellite. Moreover, movements within and between seasons, depth preferences
and winter behaviour were all unknown. There are no comparable studies in any part
of the world, so the results presented here pave the way for investigations of basking
sharks in other oceans to examine the generality of the current findings.
We found basking sharks move freely between south-west England and western
Scotland and vice versa. Sharks were capable of covering these distances in a few
weeks; movements between northern and southern sea areas of the UK occurred
within and between seasons. Basking sharks move widely, including out to the edge
of the European shelf, to exploit plankton-rich areas in addition to utilising tidal fronts
that aggregate zooplankton in coastal areas. This “resource tracking” behaviour may
result in different spatial distributions of feeding aggregations in different years and
this may impact on public sightings.
Despite circumstantial evidence for basking sharks hibernating during winter
(Matthews & Parker, 1950; Parker & Boeseman, 1954), and popular acceptance of
this idea for at least the last 40 years (Matthews, 1962), in contrast we found that
they do not hibernate on the sea bottom but instead remain active throughout the
year. The apparent disappearance of basking sharks from coastal waters in winter
was accounted for by examining the depth preferences recorded by PAT tags. These
showed that in general basking sharks spent more time deeper, or in deeper water
during winter months although the habitats occupied were similar throughout the
year.
Diel patterns of vertical movement differ substantially depending on ocean habitat;
sharks spent most time at the surface during the day in tidal-front regions of the
European shelf and most time at the surface at night in deep, strongly stratified
waters. These habitat-specific differences in surfacing time may impact on public
sightings and research surveys aimed at monitoring numbers in different areas,
including the UK protection zone.
4.3 Population monitoring
The behaviour of basking sharks to feed or cruise at the sea surface during summer
months has encouraged sightings surveys to be conducted with the aim to determine
numbers of animals using particular areas (Kenney et al., 1986; Sims et al., 1997;
MCS, 2003). This type of monitoring provides useful information on where basking
sharks occur at the surface, and clearly raises the profile of basking sharks in the
public’s consciousness, but it is now becoming apparent that any sightings trends
(even corrected for the efficiency of surveillance effort) do not reliably indicate
changes in population size and/or distribution. Until the present study, there has been
no attempt to determine biases (e.g. sea state, observer effort, diving behaviour) that
would enable reliable relative abundance estimates to be made, and no way in which
these measurements could be corrected for shark surfacing frequency and
distribution of “basking” sharks in relation to that of the whole population in order to
estimate absolute abundance.
The distribution of reconstructed tag geolocations for individual sharks was compared
with sightings of basking sharks compiled as part of the “Conserving endangered
basking sharks” (CEBS) project, which is co-ordinated by the Marine Biological
33
Association (MBA) and incorporates sightings submitted to the Marine Conservation
Society by members of the public, and sightings made during scientific surveys by
the MBA, Hebridean Whale and Dolphin Trust, The Shark Trust, UK Wildlife Trusts
and the International Fund for Animal Welfare. Figure 26 shows the distribution of
basking sharks from tag geolocations, surveys and public sightings.
Tag geolocations
2001-2002
Survey sightings
1994-2003
Public sightings
1987-2003
Figure 25. Distributions of basking sharks on the European continental shelf
determined using three independent methods (tag geolocations, scientific surveys,
public sightings) identify the same main areas where basking sharks aggregate.
Survey and sightings data from CEBS partners.
Whilst the broad distribution patterns revealed by these different methods are similar,
there are considerable differences in density distributions, with a strong emphasis in
the sightings data in the Hebridean Sea, Clyde Sea, Irish Sea and close inshore
around Devon and Cornwall, areas that clearly represent important habitat for
basking sharks, most probably in relation to feeding opportunities. Tag geolocations,
however, identified two areas where individuals spent considerable time outside the
distributions indicated by sightings: the Celtic Sea and Western Approaches. In
particular, tag geolocations show sharks undertaking persistent ranging movements
near the Celtic Sea front, whereas surface sightings would indicate relatively few
sharks in this area. The most likely reasons for this discrepancy are that there may
be fewer observers in these areas and ‘basking’ behaviour is reduced, resulting in
few surface sightings.
One explanation is the existence of normal and reverse DVM patterns in individual
sharks present in UK waters, and this results in very different surfacing frequencies in
different ocean habitats. This suggests sightings per unit effort do not reflect real
differences in geographic (horizontal) abundance between areas, because the
probability of sighting a basking shark shifts from about 0.6 in frontal areas to < 0.01
in well-stratified zones. This difference will result in under-estimating abundance in
stratified areas by about 60-fold, and this has profound implications for the use of
sighting data both in defining population distribution and estimating abundance
trends. These results suggest bias-reduction according to habitat type (and
zooplankton behaviour) should be incorporated into analyses of survey data when
attempting to estimate abundance.
4.4 Stock identification
Estimations of population size also require validation of the assumption that a single
population is sampled during surveys. The distributions obtained by sightings and
34
reconstruction of track geolocations suggest a restricted range for basking sharks
around Britain, with the majority of positions falling between the Hebrides and the
North coast of Brittany and relatively few occurrences on the west coast of Ireland
and the east coast of the UK. It was striking that the tagged sharks appeared to avoid
the low productivity in pelagic waters overlying very deep regions off the shelf (Le
Fevre 1986), and that the European shelf is used year-round by this species. Within
the above region, tagged sharks were shown to move widely, and it might be
suggested that they are part of a single stock. We do not know, however, whether
this stock area extends to the south into the Bay of Biscay, or further north into
Norwegian waters where basking shark were formerly common (Kunzlik, 1988). We
also have no knowledge of whether subpopulations exist within this area, which is an
important aspect of designing population surveys that will enable stock status to be
determined. It is possible, therefore, that population replacement may depend more
on recruitment of juveniles than immigration from other parts of the species range
and that populations of basking sharks may have extremely local distributions (many
shark species are characterised by large-scale migrations).
4.5 Recommendations and further work
The principal finding of the present study is that basking sharks use a much greater
geographical distribution range than sightings information alone would indicate.
However, whilst it is apparent that sharks using feeding areas off southwest England
and in the Clyde Sea occupy a similar distribution range over a period of a year or so,
we do not know how representative these fish are of the population living more
widely across the north-east Atlantic, nor whether there are longer term population
movements that might indicate a much wider stock range (c.f. spurdog, Squalus
acanthias; Vince, 1991). Further, since only 3 of the 20 sharks tagged during the
project were 7 m or more in length (i.e. reproductively mature), much of the new
information presented here on the movements and distribution on basking sharks
relates to juvenile or adolescent fish that may have seasonal scales and patterns of
movement that are different from sexually mature adults. For example, genetic
studies on white sharks have shown there to be sex-biased differences in dispersal,
with philopartic (non-roving) females and more widely roving males (Pardini et al.,
2001). This information is crucial if we are to provide robust scientific advice on the
conservation needs of basking sharks and management of their fisheries (if any).
There are two lines of research that will enable us to be confident that we know what
part of the north-east Atlantic population the basking sharks present around Britain
represent. Genetic studies are being carried out under GWD project CR 0288 aimed
at describing stock characteristics in the north-east Atlantic in comparison with
basking shark populations elsewhere in the world, but there is no guarantee that the
results will enable us to distinguish a “British” population. To achieve this, additional
archival tagging of sharks is required along different parts of the western seaboard,
such as south Brittany, Northwest Scotland and Norway to elucidate the extent of
movement and mixing between local groups of basking sharks. The results will allow
us to know whether exploitation outside the “normal” range of British basking sharks
will affect them, and that whether conservation measures around the UK will have
any effect either locally or on the north-east Atlantic population as a whole.
These tagging studies should be designed so as to enhance and verify the existing
behavioural information on basking sharks movement; seasonally, horizontally and
vertically, and in relation to oceanic conditions, e.g. fronts etc. These studies should
also be designed to provide more information on any differences in behaviour and
patterns of movement between males and females, and between mature and
immature fish. This will provide two benefits. It will allow us to provide weightings to
sightings data in order to determine changes in local abundance and stock trends, on
35
the assumption that sightings surveys should be encouraged to raise the basking
shark’s profile and as a long-term monitoring tool. Second, these data can be used to
parameterise spatially and temporally explicit bio-energetic models of movement and
migration that could be used to test climate change and exploitation-pattern type
scenarios, validated against existing information (including historic fisheries data) and
to make up for the inevitable data deficiencies. This may, for example, help to identify
the potential impact of environmental change on basking shark populations, and how
this compares with past or current directed fishing activities. Any future consideration
of conservation measures will need to take such factors into account. It should be
noted that this does not invoke any kind of assessment against biological models (life
table, production etc), and with little in the way of population structure information
(age, length, growth etc) or fisheries mortality on populations in the north-east
Atlantic, indicate that management measures as such are probably not needed.
36
5. References
Batten, S.D., Clark, R., Flinkman, J., Hays, G., John, E., John, A.W.G., Jonas, T.,
Lindley, J.A., Stevens, D.P. & Walne, A. (2003) CPR sampling: the technical
background, materials and methods, consistency and comparability. Progress in
Oceanography 58: 193-215.
Baum, J.K., Myers, R.A., Kehler, D.G., Worm, B., Harley, S.J. & Doherty, P. (2003)
Collapse and conservation of shark populations in the northwest Atlantic. Science
299: 389-392.
Block, B.A., Dewar, H., Farwell, C. & Prince, E.D. (1998) A new satellite technology
for tracking the movements of Atlantic bluefin tuna. Proceedings of the National
Academy of Sciences U.S.A. 95: 9384-9389.
Block, B.A., Dewar, H., Blackwell, S.B., Williams, T.D., Prince, E.D., Farwell, C.J.,
Boustany, A., Teo, S.L.H., Seitz, A., Walli, A. & Fudge, D. (2001) Migratory
movements, depth preferences, and thermal biology of Atlantic bluefin tuna. Science
293: 1310-1314.
Block, B.A., Costa, D.P., Boehlert, G.W. & Kochevar, R.E. (2003). Revealing pelagic
habitat use: the tagging of Pacific pelagics program. Oceanologica Acta. 25: 255–
266.
Boustany, A., Davis, S., Anderson, S., Pyle, P., & Block, B. (2002). Satellite tagging:
expanded niche for white sharks. Nature (London). 415:35-36.
Casey, J.M. & Myers, R.A. (1998) Near extinction of a large, widely distributed fish.
Science 228: 690-692.
Compagno, L.J.V. (1984) FAO Species Catalogue. Vol. 4, Part 1 Sharks of the
World. FAO Fisheries Synopsis 125: 1-249. Food and Agriculture Organisation of the
United Nations: Rome.
Dulvy, N.K., Metcalfe, J.D., Glanville, J., Pawson, M. & Reynolds, J.D. (2000) Fishery
stability, local extinctions and shifts in community structure in skates. Conservation
Biology 14: 283-293.
Eckert, S.A. & Stewart, B.S. (2001) Telemetry and satellite tracking of whale sharks,
Rhincodon typus, in the Sea of Cortez, Mexico and the north Pacific Ocean.
Environmental Biology of Fishes 60: 299-308.
English Nature (1999). UK Biodiversity Gropu Tranch 2 Action Plans Volume V –
Maritime Species and Habitats. English Nature, Peterborough.
Hill, R.D. (1994) Theory of geolocation by light levels. In: Elephant Seals: Population
Ecology, Behaviour and Physiology (eds B.J. Le Boeuf & R.M. Laws), pp. 227-236.
University of California Press: Berkeley.
Holden, M.J. (1973) Are long-term sustainable fisheries for elasmobranchs possible?
Rapports et Proces-Verbaux des Reunions du Conseil International pour
l’Exploration de la Mer 164: 360-367.
Holden, M.J. (1977) Elasmobranchs. In: Fish Population Dynamics (ed. J.A.
Gulland), pp. 117-137. Wiley: London.
37
IUCN (2004). IUCN Red List of Threatened Species. IUCN Gland, Switzerland and
Cambridge UK.
Kenney, R.D., Owen, R.E & Winn, H.E. (1985) Shark distributions off Northeast United
States from marine mammal surveys. Copeia 1985: 220-223.
Kunzlik, P.A. (1988) The basking shark. Aberdeen, UK: Department of Agriculture
and Fisheries for Scotland.
Le Fèvre, J. (1986) Aspects of the biology of frontal systems. Advances in Marine
Biology 23: 163-299.
Marine Conservation Society (2003) Marine Conservation Society Basking shark Watch
Report 1987-2001. Marine Conservation Society, Ross-on-Wye, UK.
Matthews, L.H. (1962) The shark that hibernates. New Scientist 280: 756-759.
Matthews, L.H. & Parker, H.W. (1950) Notes on the anatomy and biology of the
basking shark Cetorhinus maximus (Gunner). Proceedings of the Zoological Society
of London 120: 535-576.
OSPAR (2004). 2004 Initial OSPAR List of Threatened and/or Declining Species and
Habitats. OSPAR Commission, 2004.
Pardini, A.T., Jones C.S., Noble, L.R., Kreiser, B., Malcolm, H. & Bruce, B.D. (2001).
Sex-biased dispersal of great white sharks. Nature, 412: 139-140.
Parker, H.W. & Boeseman, M. (1954) The basking shark (Cetorhinus maximus) in
winter. Proceedings of the Zoological Society of London 124: 185-194.
Parker, H.W. & Stott, F.C. (1965) Age, size and vertebral calcification in the basking
shark, Cetorhinus maximus (Gunnerus). Zoologische Mededelingen, Leiden 40: 305319.
Pratt, H.L. & Casey, J.G. (1990) Shark reproductive strategies as a limiting factor in
directed fisheries, with a review of Holden’s method of estimating growth parameters.
In: Elasmobranchs as living resources: advances in the biology, ecology systematics
and status of fisheries (ed. H.L. Pratt, S.H. Gruber and T. Tanuichi) pp. 97-109.
NOAA Technical Report 90. Seattle, WA: National Oceanographic and Atmospheric
Administration.
Priede, I.G. (1984) A basking shark (Cetorhinus maximus) tracked by satellite
together with simultaneous remote-sensing Fisheries Research 2: 201-216.
Sedberry, G.R. & Loefer, J.K. (2001) Satellite telemetry tracking of swordfish, Xiphius
gladius, off the eastern United States. Marine Biology 139: 355-360.
Sims, D.W. (1999) Threshold foraging behaviour of basking sharks on zooplankton:
life on an energetic knife-edge? Proceedings of the Royal Society of London B 266:
1437-1443.
Sims, D.W. (2000) Filter-feeding and cruising swimming speeds of basking sharks
compared with optimal models: they filter-feed slower than predicted for their size
Journal of Experimental Marine Biology and Ecology 249: 65-76.
38
Sims, D.W. & Quayle, V.A. (1998) Selective foraging behaviour of basking sharks on
zooplankton in a small-scale front. Nature (London) 393: 460-464.
Sims, D.W., Fox, A.M. & Merrett, D.A. (1997) Basking shark occurrence off southwest England in relation to zooplankton abundance. Journal of Fish Biology 51: 436440.
Sims, D.W., Southall, E.J., Richardson, A.J., Reid, P.C., & Metcalfe, J.D. (2003)
Seasonal movements and behaviour of basking sharks from archival tagging: no
evidence of winter hibernation. Marine Ecology Progress Series 248: 187-196.
Vince, M.R. (1991). Stock identity in spurdog (Squalus acanthias L.) around the
British Isles. Fisheries Research, 12: 341-354
Weimerskirch, H. & Wilson, R.P. (2000). Oceanic respite for wandering albatrosses.
Nature (London) 406: 955-956.
Welch, D.W. & Eveson, J.P. (1999) An assessment of light-based geoposition
estimates from archival tags. Canadian Journal of Fisheries and Aquatic Sciences
56: 1317-1327.
Wilson, R.P., Ducamp, J.-J., Rees, W.G., Culik, B.M. & Niekamp, K. (1992)
Estimation of location: global coverage using light intensity. In: Wildlife Telemetry:
Remote Monitoring and Tracking of Animals (eds I.G. Priede & S.M. Swift), pp. 131134. Ellis Horwood: Chichester.
39
6. Outputs
Scientific papers
Sims, D.W., Southall, E.J., Richardson, A.J., Reid, P.C., Metcalfe, J.D. (2003)
Seasonal movements and behaviour of basking sharks from archival tagging:
no evidence of winter hibernation. Marine Ecology Progress Series, 248, 187196.
Sims, D.W., Southall, E.J., Tarling, G.A., Metcalfe, J.D. Habitat-specific normal and
reverse diel vertical migration in a fish megaplanktivore. Journal of Animal
Ecology, in press.
Press reports/articles
Gentle beast of the sea. The Times, 15 October 2001.
Basking sharks tracked by satellites. BBC News Online, 10 September 2002.
Sharks rethink (News in Brief). The Times, 11 September 2002.
Secret life of basking shark laid bare by tag. The Daily Telegraph, 19 September
2002.
Call to extend safety zone for the gentle giant of British seas. The Daily Telegraph, 1
October 2002.
A starring role for city sharks. Evening Herald, 13 December 2003.
No Hibernation for basking sharks. MBA News, No.32 October 2004
Scientific talks
Sims, D.W., Southall, E.J., Metcalfe, J.D. (2001). Behaviour of the basking shark in
the north-east Atlantic: A new study using pop-up satellite telemetry. European
Elasmobranch Association 5th Annual Science Meeting, 19-21 October,
University of Kiel, Germany.
Sims, D.W. (2004) The Secret Life of Sharks. Public Lecture, The Royal Institution of
Great Britain, 31 October 2001, London, UK.
Sims, D.W., Southall, E.J., Metcalfe, J.D. (2002). Migratory behaviour of basking
sharks revealed by pop-up archival telemetry. European Science Foundation
Workshop on Tracking of long-distance animal movements: navigation and
migration performance, 21-24 February, Department of Animal Ecology, Lund
University, Sweden.
Sims, D.W., Southall, E.J., Metcalfe, J.D. (2002). Movements and behaviour of
basking sharks as revealed by pop-up archival satellite transmitters. European
Marine Biology Symposium, 5-9 August 2002, University of Iceland, Reykjavik,
Iceland.
Sims, D.W., Southall, E.J., Metcalfe, J.D. (2002). Foraging and migratory behaviour
of basking sharks over seasonal scales determined using satellite archival
40
telemetry. European Elasmobranch Association, 6-8 September 2002, Cardiff,
UK.
Sims, D.W. (2002). Daily and seasonal patterns of migration in large and small
sharks. The British Association Annual Festival of Science, 9-10 September,
University of Leicester, Leicester, UK.
Sims, D.W., Southall, E.J. & Metcalfe, J.D. (presented by J Metcalfe) (2002)
Movements and behaviour of basking sharks (Cetorhinus maximus) as
revealed by pop-up archival transmitting tags. The Symposium Elasmobranch
Fisheries: Managing for Sustainable Use and Biodiversity Conservation,
hosted by the Scientific Council of the Northwest Atlantic Fisheries
Organization (NAFO), 11-13 September, 2002, Santiago de Compestela,
Spain.
Sims, D.W. (2003). Tractable models for testing theories about natural strategies:
foraging behaviour and habitat selection in free-ranging sharks. Fisheries
Society of the British Isles International Symposium Fish As Models of
Behaviour, 30 June-4 July 2003, University of East Anglia, UK.
Sims, D.W., Southall, E.J. & Metcalfe, J.D. (presented by J Metcalfe) (2002)
Movements and behaviour of basking sharks (Cetorhinus maximus) as
revealed by pop-up archival transmitting tags. ICES Symposium on Fish
Behaviour in Exploited Ecosystems, 23-26 June, 2003, Bergen, Norway.
Sims, D.W. (2004) Tracking fish with chips. Public Lecture, The Royal Institution of
Great Britain, 18 March 2004, London, UK.
Sims, D.W., Southall, E.J., Tarling, G.A., Metcalfe, J.D. (2004) Habitat-specific
differences in diel vertical migration of the basking shark and its conservation
implications. 8th European Elasmobranch Association Conference, 21-24
October 2004, Zoological Society of London, UK.
Southall, E.J., Sims, D.W., Metcalfe, J.D. (2004) Identifying critical habitat of basking
sharks in the northeast Atlantic Ocean: using electronic tags, surveys and
sightings data. 8th European Elasmobranch Association Conference, 21-24
October 2004, Zoological Society of London, UK.
Radio and TV items
Moving On - animal migration (with Matthew Parris), BBC Radio 4, 26 October 2003.
Animal Camera (with Steve Leonard), BBC1 Television, 7.30 p.m., 10 March 2004.
E-mail from a Shark. A 26-minute TV documentary by Shark-Bay Films. Winner of
the British Council Youth and Science Award at the Helsingborg Film Festival,
Sweden.
Haie in der Nordsee. A 24-min TV documentary by Tesche-Dokumentarfilm for ZDF,
Germany.
Riesenhaie vor Helgoland. A 24-min TV documentary by Tesche-Dokumentarfilm for
ZDF, Germany.
41
Appendix I
Review of the biology and ecology of the basking shark
(Cetorhinus maximus).
Note: This review was prepared as a deliverable of this project in May 2002. It was
used in developing the successful UK proposal in 2002 to include the basking shark
in Appendix II of the Convention on International Trade in Endangered Species. It
does not include the new information generated by this study on the movements and
behaviour of basking sharks that is included in the body of this report.
vi
Review of the biology and ecology of the basking shark (Cetorhinus maximus)
By D.W. Sims, Marine Biological Association, The Laboratory, Plymouth, U.K.
1. Introduction
The majority of over 380 species of shark are macropredators and scavengers, while only three
species obtain food by filtering seawater (Compagno, 1984). These however, are among the largest
living sharks, and among marine vertebrates only whales are larger. The basking shark (Cetorhinus
maximus) is the second largest known fish species attaining lengths approaching 10m and a weight of
approximately 4 tonnes. This species is greater in size than the rare megamouth (Megachasma
pelagios), but smaller than the whale shark (Rhincodon typus) of tropical regions.
Organised fisheries for basking shark have existed in the north-east Atlantic region since at least two
hundred years ago (McNally, 1976; Fairfax, 1998). Indeed, the earliest directed fisheries for pelagic
shark were probably for this species (Pawson & Vince, 1999). Despite the commercial interest, little is
known of the biology and ecology of the basking shark. The lack of basic information and the slow
progress in obtaining such data in recent times is because studying the basking shark at sea is
expensive, combined with the fact that their inshore presence during summer is unpredictable and for
the remainder of the year their whereabouts are unknown. It has been 14 years since the last review
of the scientific literature on the biology of C. maximus (Kunzlik, 1988). Because significant new
information has been added, particularly in the last five years, a new review is warranted (Fig. 1).
Furthermore, the basking shark is listed as ‘Vulnerable’ in the 1996 IUCN Red List of Threatened
Animals, so there are current concerns regarding its conservation status.
Fig. 1 The number of scientific articles by
5
C. maximus
year for the basking shark (C. maximus) with
a comparison to those for the whale shark
(R. typus). Source of data: ISI, Philadelphia,
USA.
R. typus
Number of papers
4
3
2
1
19
81
19
83
19
85
19
87
19
89
19
91
19
93
19
95
19
97
19
99
20
01
0
Year
2. Description of the species
2.1 Taxonomy
The basking shark was first scientifically described and named Squalus maximus (literally ‘largest
shark’) by Gunnerus in 1765. As Squalus was a catch-all genus for cartilaginous fish generally,
Blainville in 1816 erected a new sub-genus of Squalus named Cetorhinus (literally ‘whale shark’).
There were many objective synonyms of Squalus (Cetorhinus) maximus between 1765 and 1960,
including Halsydrus pontoppidani, Squalus pelegrinus, Squalus peregrinus, Squalus rhinoceros and
Cetorhinus maximus forma infanuncula (Compagno, 1984). For example, the latter name was erected
by Van Deinse and Adriani (1953) to describe a putative subspecies of basking shark which they
found to lack filtering gill-rakers. This proposition was successfully refuted by Parker and Boeseman
(1954) from observations that basking sharks shed gill-rakers on an apparently seasonal cycle.
Despite attempts to erect subspecies, especially for individuals found between different ocean basins,
1
it is generally considered that there is only a single species of basking shark. Springer and Gilbert
(1976) rejected the concept of at least four species subdivided on the grounds of differences in body
proportions between individuals in the north Atlantic/Mediterranean, south Atlantic and waters around
Australia. As such differences occur naturally during growth, it was considered that separation into
species merely reflected these differences and was therefore insufficient evidence for division
(Springer & Gilbert, 1976; Kunzlik, 1988).
The basking shark, Cetorhinus maximus, is the only species placed within the family Cetorhinidae,
which is considered a sister group to Lamnidae (Compagno, 1990; Martin & Naylor, 1997). These
families constitute two of the seven placed within the order Lamniformes (mackerel sharks)
(Compagno, 1984). Lamniformes is one of eight orders of shark within Class Chondrichthyes
(subclass Elasmobranchii). The interrelationships of shark taxa including those species within
Lamniformes is not without controversy. Maisey (1985) argued that the megamouth shark,
Megachasma pelagios should be included within Cetorhinidae on account of its similarities with C.
maximus jaw suspension and dental array, rather than forming a new monotypic family
(Megachasmidae) as proposed by Taylor et al. (1983). However, as noted in Dulvy and Reynolds
(1997), the cladistic phylogeny of the monophyletic Lamniformes (Compagno, 1990) is consistent with
the molecular phylogenies of Martin et al. (1992) and Naylor et al. (1997). Furthermore, recent
molecular analysis of cytochrome b gene sequences implies independent origins of filter-feeding
within Lamniformes, and hence argues against C. maximus and M. pelagios forming sister taxa within
Cetorhinidae (Martin & Naylor, 1997) (Fig. 2).
Species
From: Martin & Naylor (1997)
Mitsukurina owstoni (1)
Carcharias taurus (2)
Odontaspis ferox (2)
Odontaspis horonhai (2)
Pseudocarcharias kamoharai (3)
Megachasma pelagios (4)
Alopias vulpinus (5)
Alopias superciliosus (5)
Alopias pelagicus (5)
Cetorhinus maximus (6)
Lamna nasus (7)
Lamna ditropis (7)
Carcharodon carcharias (7)
Isurus paucus (7)
Isurus oxyrinchus (7)
Fig. 2 Interrelationships of species within Lamniformes derived from molecular data (from Martin &
Naylor, 1997) and which is consistent with the phylogeny derived from cladistic analysis (Compagno,
1990). The same number beside species names denotes placement within the same family: 1,
Mitsukurinidae; 2, Odontaspididae; 3, Pseudocarchariidae; 4, Megachasmidae; 5, Alopiidae; 6,
Cetorhinidae; 7, Lamnidae.
2.2 Morphology and structure
The basking shark is a large-bodied fish with a fusiform body shape. Detailed general descriptions of
external morphology and internal anatomy are given in Matthews and Parker (1950) and which are
summarised in the review by Kunzlik (1988). Nothing needs to be added to these treatments here
2
other than to provide the reader with a brief overview of the species field marks that are of particular
interest, and to describe the differences in fin dimensions between juvenile and adult individuals.
The colour of the body surface varies in descriptions, from black to dark grey through slate grey to
brown (Matthews & Parker, 1950; Kunzlik, 1988). When observed in sunlight in its natural habitat,
basking sharks appear grey-brown with lighter dappled or irregular longitudinal patterns along its
lateral surface (Fig. 3a). When dead and out of water, basking sharks appear slate or dark grey-black
(Fig. 3b). The variations in body colour reported may therefore reflect changes due to death and/or
removal from water (Kunzlik, 1988).
The large body size is a feature that helps distinguish this shark from all others (Matthews & Parker,
1950; Compagno, 1984). Basking sharks have been credited with maximum total lengths between
12.2 and 15.2 m (Compagno, 1984), whilst theoretical maxima have been given as 12.76 and 13.72 m
(Parker & Stott, 1965; Kunzlik, 1988). Compagno (1984) states that even if these are correct, most
specimens do not exceed 9.8 m total length. However, the longest reliable measurement of a shark
caught in static fishing gear in Newfoundland was found to be a 12.2-m long male (Lien & Fawcett,
1986). Consequently, the basking shark is the second largest shark species (elasmobranch, and fishlike vertebrate) in the world after the whale shark (R. typus). The body mass of basking sharks in
relation to total length is not well known on account of the difficulties associated with weighing large
specimens. Maximum body masses of 5-6 tonnes have been ascribed to adult sharks in popular
accounts. However, two Californian specimens measuring 8.5 and 9.1 m total length weighed 2991
and 3909 kg respectively (Bigelow & Schroeder, 1948). The body mass of an 8.3-m total length female
shark taken off Florida was found to be 1980 kg (Springer & Gilbert, 1976), and a 6.0 m individual from
Scotland weighed approximately 2000 kg (Stott, 1980). An adult female and an adult male basking
shark of 6 to 7 m length taken off Plymouth weighed 1678 and 1924 kg respectively (Bone & Roberts,
1969). Kruska (1988) measured the mass of a 3.75 m long specimen to be 385 kg.
a
b
c
d
Fig. 3 Photographs of four different basking sharks to illustrate the variation in colouration and
patterning among living (a, b, d) and dead (c) specimens.
3
Basking sharks have correspondingly large fins and a caudal peduncle with strong lateral keels. The
first dorsal fin measured in an adult female of 8.3 m total length (LT) was 1.1 m in height (Springer &
Gilbert, 1976). The pectoral fins were similar in length (~1.3 m) to the first dorsal fin height, whereas
the leading edge of the caudal fin was 1.7 m in length. In contrast, the length of the pectoral fins of a
2.6-m LT immature female C. maximus, was nearly twice that of the dorsal fin height (0.22 m),
whereas the leading edge of the caudal fin was nearly 0.7 m long (Izawa & Shibata, 1993).
Presumably these differences in fin proportions relate to ontogenic changes in gross morphology. A
table of fin measurements for 13 individuals ranging in total length from 2.6 to 8.5 m, together with a
table of relative differences in fin dimensions from three of these individuals are given in Appendix 1.
In addition to the pointed snout and huge sub-terminal mouth, there are minute hooked teeth (~5 mm
in height) arranged in three to seven functional rows on the upper and lower jaws respectively
(Matthews & Parker, 1950). The teeth are modified placoid dermal denticles. Small denticles of the
normal type point posteriorly over the entire skin surface which is also covered with a dark-coloured
mucus to the level of the summits of the denticles (Matthews & Parker, 1950). During behavioural
studies, this mucus has been deposited on ropes (used to deploy plankton nets) by basking sharks as
they brush past them during normal swimming (D. Sims, unpublished observations). Taking skin
swabs of this mucus may be an effective, non-invasive method for obtaining shark DNA in future
studies.
Basking sharks are also typified by their enormous gill slits that virtually encircle the head (Fig. 3a).
The five gill slits on each side of the pharyngeal area are openings between the gill arches upon which
there are two distinct structures: the gill lamellae that enable respiration by the exchange of oxygen
with seawater, and anteriorly, the gill rakers which are comb-like structures arranged in a single row
along the distal portion of each gill arch. When the mouth is open, two rows of gill rakers on separate
gill arches extend across each gill-slit gap and act to filter zooplankton prey from the continuous flow
of seawater produced by forward swimming (so-called ram-filter feeding). The rakers are erected
when the mouth opens by contraction of the muscle on the aboral half of the foot of the raker, and are
held in position against the water flow by elastic fibres in the connective tissue strip (Matthews &
Parker, 1950). When the slits are closed the rakers lie flat against the surface of the arches. The gill
rakers are about 0.1 m long in the centre of the gill arch and the inter-raker distance is about 0.8 mm
in an adult specimen (Matthews & Parker, 1950). Anatomical investigations of summer and wintercaught specimens suggest that the gill rakers are shed in late autumn or early winter, re-grow through
the winter and erupt through the gill-arch epidermis in late winter/early spring, in time for seasonal
feeding (Parker & Boeseman, 1954; Matthews, 1962).
The liver of the basking shark is large and makes up between 15-25 % of its body weight (Kunzlik,
1988). The hydrocarbons of zooplankton pass through the basking shark alimentary canal without
fractionation or structural modification, and are resorbed in the spiral valve and deposited in the liver
(Blumer, 1967). Even though squalene is present only in traces in zooplankton, it is abundant in the
liver of basking sharks, which contains between 11.8 and 38.0% squalene (Blumer, 1967; Kunzlik,
1988). The liver functions as both an energy store and as a hydrostatic organ for increasing static lift
(Bone & Roberts, 1969; Baldridge, 1972).
The skeleton of the basking shark is cartilaginous with varying degrees of calcification throughout, but
like that of other sharks, these structures are not ossified (Kunzlik, 1988). The paired sexual organs
(claspers) of male sharks are located ventrally at the base of the paired pelvic fins and these become
progressively calcified with maturity. Sharks have been aged by counting growth zones visualised in
structures such as dorsal spines and vertebral centra (for review see Cailliet, 1990). These growth
zones are comprised of opaque bands that have cells with high concentrations of calcium and
phosphorus and translucent bands that are less mineralised (Yudin & Cailliet, 1990). Species such as
the blue shark appear to deposit alternate dark and light concentric rings annually (Stevens, 1975), but
quantifying age at length has been verified in less than 10 species (Cailliet, 1990). The number of
4
rings present in the vertebra centra of basking sharks varies along an individual’s body length and so
ageing this species has proved problematic (Parker & Stott, 1965). Recent progress has been made
however in ageing filter-feeding whale sharks using X-radiography of vertebral centra (Wintner, 2000).
3. Distribution and habitat
3.1 Total area
The basking shark is a coastal-pelagic shark known to inhabit the boreal to warm-temperate waters of
the continental and insular shelves circumglobally. It has been recorded in the western Atlantic from
Newfoundland to Florida and from southern Brazil to Argentina (Wood, 1957; Compagno, 1984;
Tomas & Gomes, 1989). In the eastern Atlantic C. maximus is present from Iceland, Norway and as
far as the Russian White Sea (southern Barents Sea) to the Mediterranean and Senegal, western
Cape province and South Africa (Konstantinov & Nizovtsev, 1980; Compagno, 1984). They are also
present in the Pacific Ocean; from Japan, the Koreas, China, Australia (south of 25oN) and New
Zealand in the west, and from the Gulf of Alaska to Baja California, Peru and Chile in the east
(Compagno, 1984). The basking shark has been recorded primarily from coastal areas, however this
may not represent its entire habitat range as distribution throughout the epipelagic zone of ocean
basins is possible. The coastal distribution pattern may be misleading for two main reasons. Firstly,
basking sharks are more likely to be observed/captured in coastal-shelf areas on account of the
increased human occupation/usage of these areas compared to the open sea. Secondly, basking
sharks may spend more time at depth in deeper waters and away from the surface layers. The latter
scenario would likely lead to lower sightings and/or capture rates in fishing gear. Therefore, at best,
our knowledge of the total area distribution of this species is presently limited.
3.2 Differential distribution
Although population segregation by size and sex is a general characteristic of shark species
worldwide (Klimley, 1987), there is no substantial evidence to indicate differential distribution in the
basking shark. Juvenile (2-3 m total length, LT) and putative sub-adult (3-5 m LT) sharks have been
frequently observed in the same areas and summer feeding aggregations as adults (Berrow &
Heardman, 1994; Sims et al., 1997). There was some indication that juveniles and sharks <3 m LT
appeared at the surface to feed later in the summer compared to larger individuals (Sims et al., 1997),
but this may have been due to biotic factors such as zooplankton abundance rather than agesegregated distribution or migration per se. In the years since the observations of Sims et al. (1997)
were made, a shift towards smaller-sized sharks off Plymouth as the summer progresses has been
less obvious (D.W. Sims, unpublished observations).
Similarly, at present there is no substantial evidence to indicate whether sexual segregation of the
population occurs. Males and females have been observed in the same areas during summer
(Matthews & Parker, 1950; Maxwell, 1952; O’Connor, 1953; Watkins, 1958; Sims et al., 2000),
although more females than males are caught in directed fisheries (Kunzlik, 1988). Pregnant females
are virtually unknown from these same locations so differential habitat utilisation by males and females
at certain times in the reproductive cycle may well occur.
4. Bionomics and life history
4.1 Reproduction
Matthews (1950) gives a detailed account of reproduction in the basking shark based upon macro and
microscopic anatomical investigations of dissected specimens from Scotland. To summarise the main
points of interest briefly, Matthews (1950) suggests the basking shark is ovoviviparous, that is, live
young are produced from eggs that hatch within the body. This mode of reproduction is common
among large-bodied elasmobranchs, including the whale shark (Joung et al., 1996). What was
unusual however, was that in the female basking shark only the right ovary was functional, but this
contained at least six million ova each about 0.5 mm in diameter (Matthews, 1950). Smaller numbers
of more heavily yolked ova are more commonly found in sharks (Kunzlik, 1988). Fertilisation in the
5
basking shark, as in all other sharks is internal: the intromittent organs (claspers) are inserted via the
female’s cloaca into the vagina and transfer large quantities of sperm packets or spermatophores. In
male basking sharks, spermatophores are up to about 3 cm in diameter, each with a core of sperm
and a firm translucent cortex. The spermatophores float in a clear seminal fluid and Matthews (1950)
estimates that about 18 litres of them are transferred to the female during mating. The period of
gestation is not known with any certainty, but estimates as high as 3.5 years have been proposed
(Parker & Stott, 1965), although a period of just over one year has been estimated from the same
length-frequency data (Holden, 1974). There is only one published record of a pregnant female being
captured despite organised fisheries for basking sharks in the northeast Atlantic dating back at least
two hundred years. According to this single account, a female basking shark was caught in August
1936 off the mid-western coast of Norway and towed into Teigboden (Sund, 1943). Whilst being towed
the shark gave birth to six pups, each about 1.5-2.0 m LT, five of which began swimming openmouthed at the surface, presumably feeding. The sixth pup was stillborn. Therefore, if this number of
pups is representative of normal parturition rates, it seems the basking shark exhibits very low
fecundity even when compared to other ovoviviparous sharks. It has been suggested that female C.
maximus do not mature until at least their third year (Matthews, 1950), however as is the case for
embryo development and parturition, growth and age at maturity in this species is very poorly
understood.
Total length (m)
4.2 Growth and Ageing
The growth rate of basking sharks is not known exactly, but there have been attempts to estimate age
using two methods: (1) Length-frequency analysis has been used to derive length-at-age growth
curves (Matthews, 1950; Parker & Boeseman 1954; Parker & Stott 1965), and (2) vertebral centra
analysis has been used to relate observed numbers of ‘age rings’ to measured body length (Parker &
Stott, 1965). For the length-frequency analysis, which attempts to relate successive modes in the
length-frequency distribution with successive age groups, measurements of 93 fishery-caught
individuals from the north-east Atlantic were used. These data resulted in suggestions that a size of
not < 2 m length was typical of the first summer, a mean size of 3.09 m was attained in the following
summer and that the mean size in the next winter was 3.52 m (Parker & Stott, 1965). From these
empirical data and using the assumptions that growth was asymptotic and best described by a von
Bertalanffy growth function, that the length at parturition was 1.5 m LT, and that the maximum length
asymptote was 11.0 m LT, Parker and Stott (1965) derived a growth curve for C. maximus (Fig. 4).
10
9
8
7
6
5
4
3
2
1
0
Fig. 4. A hypothetical growth curve for
C. maximus based upon mean size
growth increment over a 0.5 year
period (triangles) (Parker & Stott,
1965), and an observed growth
increase for a female shark re-sighted
by Sims et al. (2000) after a 3-year
period (circles).
Sims et al. 2000
Parker & Stott (1965)
0
2
4
6
8
10
12
14
Age (years)
The growth curve indicates that a 5-m long shark will be about 4 years old whereas a 9-long individual
will be at least 12.5 years old. Pauly (1978) estimated longevity at about 40 years. There have been
no recent studies to evaluate the growth of basking sharks using length-frequency analysis. However,
an opportunistic re-sighting of a female shark by Sims et al. (2000) showed that this 5.0-m long shark
had increased in total body length by approximately 2.4 m in just over 3 years. Assuming a 5 m long
6
shark to be in its fourth year, then the growth increment ascribed to this individual seems well
predicted by Parker & Stott’s (1965) hypothetical growth relationship for C. maximus.
Validation of age-at-length using growth rings in verebral centra of basking sharks has proved difficult
because the number of rings decreases caudally suggesting uneven laying-down of rings and as a
function of body length and with respect to time (Parker & Stott, 1965). Furthermore, there appear to
be seven rings present at birth (Parker & Stott, 1965). The Pacific angel shark (Squatina californica)
has 6 or 7 bands present in the vertebral centra at birth and up to 42 in the largest adults (Natanson &
Cailliet, 1990). It was demonstrated that these bands were not deposited annually as they are in some
species (e.g. Prionace glauca; Stevens, 1975; Cailliet, 1990), but deposition was related to somatic
growth. Parker and Stott (1965) showed that vertebral centra of basking sharks between 3.5 and 5.5
m total body length contained 9 – 16 rings, whereas those from 7.5 to 9.0 m LT possessed between 26
and 32 rings. The latter authors suggested for basking sharks that 2 opaque bands were deposited
per year perhaps as a function of increased somatic growth during the two main periods of plankton
productivity in temperate waters. Despite these suggestions and the early studies described above,
there has been no contemporary work to progress age determination in basking sharks.
For the reasons stated above, the length (age) of basking sharks at sexual maturity is very uncertain.
Matthews (1950) and Matthews and Parker (1950) observed mature males at lengths between 6.8 and
8.1 m. Rapid increase in male clasper length occurred between 6.0 and 7.5 m body length with little
change thereafter (Francis & Duffy, 2002). Female length at maturity is uncertain, but females
between 7.7 and 8.2-m long were considered mature by Matthews (1950) and Matthews and Parker
(1950). Therefore, sexual maturity in females is unlikely to be reached until at least an age of 10
years.
4.3 Food and feeding
The basking shark feeds upon zooplankton prey it captures by forward swimming with an open mouth
so that a passive water flow passes across the gill-raker apparatus. Unlike the megamouth and whale
sharks that may rely upon suction or gulp feeding to capture swarms of zooplankton (Diamond, 1985;
Clark & Nelson, 1997), the basking shark is an obligate ram filter-feeder. But exactly how the
particulate prey is filtered remains unresolved. It has been assumed that the erect gill-rakers filter
particulate matter of a suitable size from the passive water flow directly, that is, like a ‘mechanical
sieve’ (dead-end filter) (Matthews & Parker, 1950; Kunzlik, 1988; for review see Gerking, 1991).
Apparently, when the mouth closes the rakers collapse on the gill arches and deposit zooplankton
onto mucus that is produced in vast quantities by cells at their base (Matthews & Parker, 1950).
However, the gill rakers are very thin, stiff bristles so it is not easy to see how these function to retain
plankton on their surfaces, because zooplankton are similarly of small diameter and unlikely to adhere
to them as the rakers contain no mucus-producing cells. It seems reasonable to assume that the small
gap between the rakers (the inter-raker distance), which is about 0.8 mm in adults, could prevent
particulate prey from passing through. However, basking sharks only swallow plankton every 30 to 60
secs (Hallacher, 1977; D.W. Sims, unpublished observations) so it remains unclear how plankton is
retained and trapped in position without loss for this length of time before swallowing. A recent study
of filter-feeding in small-bodied teleost fish suggests instead that rakers function as a crossflow filter
(Sanderson et al., 2001). Particles are not retained on rakers but are concentrated in the oral cavity
towards the oesophagus as water exits between the rakers. Apparently the crossflow prevents
particles from clogging the gaps between the rakers (Sanderson et al., 2001). Further study of the fluid
dynamics in basking shark models may elucidate a similar system.
Even though the actual mechanics of filter-feeding in basking sharks remains unknown, the prey
captured by them has been recorded for several specimens. Post mortem studies on basking shark
stomachs show that off Scotland calanoid copepods were generally the predominant prey group
(Matthews & Parker, 1950; Watkins, 1958). Matthews and Parker (1950) found Calanus and other
copepods, in addition to fish eggs, cirripede and decapod larvae. Records of the copepods Oithona,
7
Calanus, and Pseudocalanus have also been made from basking shark stomachs (Sproston, 1948
cited in Matthews & Parker, 1950). The main zooplankton species identified from shark feeding areas
in the English Channel off Plymouth were Calanus helgolandicus, Pseudocalanus elongatus, Temora
longicornis, Centropages typicus and Acartia clausi (Sims & Merrett, 1997). The density of total
zooplanktonts counted from samples taken in shark feeding areas was about 2320 per cubic metre
(Sims, 1999). The density of calanoid copepods ranged from 1050 to 1480 per cubic metre with C.
helgolandicus of 2 mm mean length making up about 70% of this total by number (Sims & Merrett,
1997). Mysid larvae, decapod larvae, chaetognaths, larvaceans, polychaetes, cladocerans, fish larvae
and post-larvae, and fish eggs were also recorded (Sims & Merrett, 1997). Calanoid copepods
dominated the stomach contents almost entirely of a 3.3-m long female shark found tangled in nets in
the English Channel (D.W. Sims, unpublished observations). However, in other regions basking
sharks utilise larger zooplankton prey. The stomach contents of an 8.1-m long basking shark off the
east coast of Japan was found to contain only specimens of the pelagic shrimp Sergestes similis,
which had been preyed upon by the shark at a depth below 100 m at night (Mutoh & Omori, 1978).
The shrimps in the shark’s stomach ranged in body length from 40 to 54 mm. The length-frequency
distribution for shrimps taken by the shark were similar to that sampled using trawl nets (Mutoh &
Omori, 1978).
The cardiac stomach contents of a large basking shark have been found to weigh over 0.5 tonnes, of
which only 30% was organic matter (Matthews & Parker, 1950). The rates of gastro-intestinal
evacuation in basking sharks are unknown, however filtration rates have been estimated using
measurements of swimming speed and mouth gape area. Using a swimming speed of 1.03 m s-1 for a
7 m shark with a mouth gape area of 0.4 m2, a maximum filtration rate of 1484 m3 h-1 was estimated
(Parker & Boeseman, 1954). This estimate has perpetuated in the literature and popular accounts,
however it fails to take into account the inefficiencies associated with filter-feeding, namely buccal flow
velocity was assumed to equal forward swimming velocity, and swallowing (prey handling) time was
not considered. A recent study in the English Channel measured the swimming speeds of 4.0 – 6.5-m
long basking sharks accurately and found that they filter feed at speeds some 24% slower than when
cruise swimming with the mouth closed (Sims, 2000a). Basking sharks were observed filter feeding at
a mean speed of 0.85 m s-1 (± 0.05 S.E.) and larger 9-m long sharks apparently do not swim
appreciably faster (Harden-Jones, 1973). Therefore using these recent studies, a more accurate
seawater filtration rate for a 7 m basking shark (mouth gape area ca. 0.4 m2) swimming at a speed of
0.85 m s-1 was calculated to be 881 m3 h-1, allowing for an observed swallowing (prey handling) time of
6 s min-1 (Hallacher, 1977) and assuming the actual buccal flow velocity to be 80% of the forward
swimming velocity (Sanderson et al., 1994). This suggests basking sharks filter seawater for food at a
rate some 41% lower than previously thought.
4.4 Behaviour
4.4.1 Foraging
Basking sharks are most frequently seen around the north, west and south-west coasts of Britain feeding
at the waters’ surface during summer months (Berrow & Heardman, 1994; Sims et al., 1997). In the
northwest and northeast Atlantic, surface foraging occurs from around April to October usually with a
peak in sightings from May until August (Kenney et al., 1985; Berrow & Heardman, 1994). The seasonal
increase in the surface sightings of basking sharks in British waters during May and early June coincides
with increased zooplankton abundance at this time (Sims et al., 1997; Sims, 1999). Similarly,
observations of surface-feeding basking sharks in Clayoquot Sound, British Columbia, were coincident
with the season of highest plankton productivity in the region (Darling & Keogh, 1994). In contrast,
basking sharks in the northeast Pacific off the central and southern California coast have been observed
at the surface from October to May, with peaks in October and March (Squire, 1990). Surfacing behaviour
in this particular region therefore occurred both well before and after the June peak in phytoplankton
abundance (Squire, 1990). Further studies are required to establish the timing of surface behaviours with
8
respect to seasonal trends in zooplankton abundance as relationships appear to differ between
geographic regions.
Basking sharks observed at the surface feed almost continuously, and frequently occur in large
aggregations. In the English Channel off Plymouth groups numbering between 3 and 12 individuals were
observed (Sims et al., 1997, Sims & Quayle, 1998), although aggregations of apparently up to 200
individuals have been reported off Cornwall by fishermen. There does not appear to be any social
organisation within these feeding groups. Basking sharks are primarily solitary, but their propensity to
exhibit prolonged feeding behaviour in specific areas probably results in the formation of foraging
aggregations. These have been shown to occur most often near oceanographic features (Sims & Quayle,
1998). A basking shark tracked by satellite was shown to remain close to a thermal boundary or front
between two water masses of different temperature (Priede, 1984). There have been similar sightings of
basking sharks feeding close to frontal features (e.g. Choy & Adams, 1995).
A thermal front is a region characterised by a larger-than-average horizontal gradient in water
temperature, which forms a boundary between warm, stratified and cold, mixed waters (Le Fevre, 1986).
Fronts can be formed by changes in tidal current speed as a function of depth, by underwater
topographical features that deflect currents to the surface, or by internal waves near shelf edges (Le
Fevre, 1986; Wolanski & Hamner, 1988). Fronts have biological significance because they are often
associated with enhanced primary and secondary production (plankton). This may be due to the
favourable conditions presented by nutrients diffusing from cold, mixed water into warmer water that can
confer higher rates of growth, or by aggregation of particulate plankton at these boundaries due to
complex upwelling and downwelling currents (Le Fevre, 1986). Fronts are of significance to marine
vertebrates generally (Wolanski & Hamner, 1988), and recent behavioural studies have demonstrated
their role as important habitat for foraging by basking sharks.
Basking sharks were thought to be indiscriminate planktivores that were unlikely to orientate to specific
plankton-rich waters (Matthews & Parker, 1950). However, Sims and Quayle (1998) tracked basking
sharks responding to zooplankton gradients and showed they were selective filter-feeders that chose the
richest, most profitable plankton patches. Basking sharks foraged along thermal fronts in the English
Channel and actively selected areas containing high densities of large zooplankton above a threshold
density. Surface-feeding basking sharks followed convoluted swimming paths along tidal slicks
associated with the front, and exhibited area-restricted searching (ARS) where zooplankton densities
were measured to be high (> 1 g m-3). As observed in other animals, ARS behaviour in basking sharks
was characterised by increased rates of turning and decreased swimming speeds (Sims & Quayle, 1998;
Sims, 1999; Sims, 2000a). Individually tracked sharks spent twice as long in areas with zooplankton
densities > 3 g m-3 compared with time spent in areas < 1 g m-3 (Fig. 5). Further study showed that
basking sharks surface-feed in areas in which the dominant calanoid copepod prey, Calanus
helgolandicus, was 2.5 times as numerous and 50% longer than in areas in which sharks do not feed
(Sims & Merrett, 1997). In the feeding areas there were also fewer numbers of smaller zooplankton
species, and therefore the biomass per cubic metre where sharks’ foraged was significantly increased.
These studies emphasise the role of tidal fronts as important annual habitat utilised by large numbers of
basking sharks. However, the fact that the duration of summer stratification in sea coastal areas is likely
to be altered by climate warming (Wood & McDonald, 1997) raises the question of how predicted
changes in the persistence of thermal fronts will affect the timing and location of foraging behaviour in this
species. Between years, the feeding locations of basking sharks indicated broad shifts in front-located
secondary production associated with a shift in location of the seasonally persistent front as a result of
local weather conditions (Sims & Quayle, 1998) (Fig. 6). Furthermore, basking sharks integrate a
planktivorous fish’s behaviour with zooplankton abundance directly. Therefore, it has been suggested that
basking sharks may be useful detectors of the distribution, density and characteristics of zooplankton in
fronts, and could provide high-trophic-level biological indication of fluxes in zooplankton assemblages that
are affected by oceanographic and climatic fluctuations of the north Atlantic (Sims & Quayle, 1998).
9
Future surveys of basking sharks where identifying large numbers of individuals becomes important
(perhaps using photographic identification; Sims et al., 2000b), efforts should be concentrated in these
areas.
Fig. 5 Feeding behaviour of basking sharks in relation to zooplankton density. (a) Representative finescale-foraging tracks of two basking sharks responding to zooplankton gradients, where track 1 denotes
a non-feeding shark and track 2 a feeding shark. Numbers along each track represent zooplankton
densities sampled in g m-3. (b) Positions of the tracks in relation to coastline and bathymetry off
Plymouth. (c) Relationship between zooplankton density class and the time basking sharks spent within
25 m of the zooplankton-sample locations. From Sims and Quayle (1998).
Fig. 6 Distribution of surfaceforaging basking sharks and
changes in summer sea surface
temperature (SST). The distribution
of foraging sharks in 1996 (open
circles) compared to 1997 (closed
circles) and the location of the
thermal front in June, 1996. The
inset chart shows the change in
summer SST at the sampling
station S1 in 1996 and 1997, with
brackets showing the periods when
SST fluctuated which is consistent
with decreases in frontal sharpness
due to calm weather conditions.
From Sims and Quayle (1998).
There is also evidence to indicate that within feeding aggregations the amount of time individual basking
sharks spend on the surface is proportional to the quantity of zooplankton present in surface waters (D.W.
Sims, unpublished data). Hence, the probability of sighting basking sharks in highly productive areas will
be greater because they spend longer at the surface. This suggests that future sightings schemes for
basking sharks should take zooplankton abundance in specific search areas into account. If zooplankton
10
abundance from year-to-year is not quantified in addition to the number of sharks sighted, then it will be
impossible to assess whether the number of sharks observed per unit time was due to enhanced
zooplankton abundance in that region rather than any other factors. A recent study on minke whales
(Baleanoptera acutorostrata) has shown that inaccurate population censuses can be obtained because
the probability of surface sightings can increase at certain times of the day, and in certain months (Stockin
et al., 2001; Young, 2001).
Recent studies have also investigated the effect decreases in zooplankton density have on the foraging
behaviour of basking sharks. It was found that individuals remain for up to 27 hours in rich patches that
are transported by tidal currents (Sims & Quayle, 1998). In one zooplankton patch monitored, up to 23
different sharks were observed to surface feed over a period of 224 h, during which time prey density
declined exponentially from between 1.47 – 8.29 g m-3 in the first 24 h to 0.50 – 0.80 g m-3 after 224 h
(Sims, 1999). This indicates basking sharks have the potential to influence the density and diversity of
plankton communities directly (Sims, 2000b). Furthermore, a lower threshold foraging level was also
determined using empirical data from behavioural studies of individual and group-feeding sharks and
theoretical calculations (Sims, 1999). This study showed that basking sharks tend to stop feeding and
leave patches when prey density reaches between 0.48 and 0.70 g m-3, values which were in good
agreement with the theoretical threshold prey density of between 0.55 and 0.74 g m-3 (Sims, 1999). A
previous study calculated the lower threshold to be 1.36 g m-3, a relatively high value that was then used
to argue that basking sharks could not derive net energy gain outside of summer months and so probably
hibernate during the winter in a non-feeding state (Parker & Boeseman, 1954; Matthews, 1962). Although
Parker and Boeseman’s (1954) threshold estimate was only roughly double that of Sims (1999), it was
found that the parameter values they used were not accurate in the light of modern data and
methodology, and that in turn the prey density estimate of 1.36 g m-3 cannot now be considered to be
correct (Sims, 1999; Weihs, 1999). The new prey threshold estimate of ~0.6 g m-3 is important because it
questions the validity of the ‘hibernation’ hypothesis. The results of Sims (1999) strongly suggest that
basking sharks are capable of utilising lower prey densities than 1.36 g m-3 for maintenance of growth
rates. Because zooplankton densities between 0.60 and 1.36 g m-3 occur in north-east Atlantic waters
outside summer months (Harvey et al., 1935; Digby, 1950), the implication of the work of Sims (1999) is
that sufficient productivity to support basking shark feeding and growth may not be as spatio-temporally
limited as suggested by Parker and Boeseman (1954). Therefore, basking sharks may not be limited to
feeding on high densities in summer alone (Sims, 1999).
Early anatomical studies demonstrated that winter-caught basking sharks often lacked gill-raker filtration
apparatus (Van Deinse & Adriani, 1953; Parker & Boeseman, 1954). This seasonal loss was used as
evidence to support the idea that when zooplankton densities decrease below 1.36 g m-3 they shed their
gill rakers and hibernate whilst re-growing their rakers during the winter months (Parker & Boeseman,
1954; Matthews, 1962). However, Sims (1999) stated that a significant proportion (~40%) of basking
sharks in winter have been found with full sets of gill rakers and zooplankton prey in their stomachs (Van
Deinse & Adriani, 1953; Parker & Boeseman, 1954). It appears that the chronology of autumn/winter
shedding of rakers, winter re-growth and eruption of new rakers in early spring suggested by Parker and
Boeseman (1954) was developed from detailed analysis of three individual sharks. Appraisal of the entire
dataset available to these workers suggests this chronology may not apply to all individuals in the
population (Sims, 1999). Basking sharks may have a shorter raker development time or shedding and regrowth may be asynchronous, which would account for sharks in winter possessing rakers and having
food in their stomachs (Sims, 1999).
4.4.2 Courtship
Courtship behaviours are used by animals to attract potential mates and as a prelude to mating.
Comparatively little is known about courtship and mating behaviours in wild sharks as it has proved
extremely difficult to study, especially in large pelagic sharks. Actual reproductive behaviours such as
courtship, pairing, copulation, or post-copulatory activities have been described in only nine out of the
380 or so species of sharks, and most of these have been for captive animals (Carrier et al., 1994).
11
Reproduction in the basking shark has been studied only from anatomical examinations of fisherycaught individuals (Matthews, 1950). The latter study supports the hypothesis that mating occurs
during summer months off the British Isles. Adult basking sharks caught off west Scotland during the
summer of 1946 were in breeding condition and showed signs of having recently copulated
(Matthews, 1950). Females bore recent or unhealed cloacal wounds inflicted by the claw on the
clasper of the male during copulation. A female examined closely contained many spermatophores,
while both males and females carried abrasions near the pelvic area possibly due to contact of the
roughly denticulated skin in this region made during pairing (Matthews, 1950). On the basis of these
data, Matthews (1950) concluded that the breeding season was in ‘full swing’ during the second half of
May off west Scotland. There have been anecdotal behavioural observations of interactions between
sharks before capture (Matthews & Parker, 1950), but until recently however, there have been no
detailed studies of social or courtship behaviour.
Elements of courtship and putative mating behaviours among a group of 13 basking sharks at the
surface over deep water (ca. 130 m) were recently recorded for a 5-min period off the coast of Nova
Scotia, Canada (Harvey-Clark et al., 1999). In the latter study, nose-to-tail following, flank approach,
close approach including rostrum-body contact, parallel and echelon swimming and possible pectoral
biting were observed and interpreted to be consistent with courtship and mating behaviours. There
are descriptions and observations of close-following behaviour in a number of shark species, including
blacktip (Carcharhinus melanopterus) and whitetip (Triaenodon obesus) reef sharks in the wild
(Johnson & Nelson, 1978), captive bonnethead (Sphyrna tiburo) and sandtiger (Carcharias taurus)
sharks (Myrberg & Gruber, 1974; Gordon, 1993), and captive (Klimley, 1980) and free-ranging nurse
sharks (Ginglymostoma cirratum) (Carrier et al., 1994).
Despite the timeliness of the opportunistic observation of putative courtship and mating in basking
sharks made by Harvey-Clark et al. (1999) in view of the general lack of information, there were
several obvious shortcomings. It was not possible for the authors to verify the sex of individuals
exhibiting following behaviours, to characterise the behaviours over longer time-periods for
quantitative comparison with those seen in other sharks species, or to determine courtship duration
and its spatio-temporal occurrence. However, annual courtship-like behaviour in basking sharks from
25 separate episodes was observed and tracked during a five-year study (1995-1999) off southwest
England (Sims et al., 2000). Social behaviours were observed between paired, or three or four sharks
and were consistent with courtship behaviours seen in other shark species, namely nose-to-tail
following, close following, close flank approach, parallel and echelon swimming (Fig. 7). Behaviours
were recorded between individuals of 5 to 8 m total body length (LT), whereas smaller sharks (3 – 4 m
LT) did not exhibit these behaviours. In this study, lead individuals were identified as females and
interactions were prolonged; the longest continuous observation of socialising was 1.8 h, although
intermittent track data indicated bouts may have lasted up to 5-6 h (Sims et al., 2000).
Breaching behaviour, signified by basking sharks leaping completely clear of the water also occurred
during observed social interactions in the western English Channel (Sims et al., 2000). This behaviour
by basking sharks was at first thought to be improbable (Matthews & Parker 1950), however was
frequently observed between May and June by shark fishermen off Scotland (Matthews & Parker,
1951). Breaching is thought to act as social communication between predatory white sharks
(Carcharodon carcharias) when entering their seasonal reproductive mode (Pyle et al., 1996), and
between filter-feeding whales, where it may also be used as a courtship display (Whitehead, 1985).
Similarly, breaching behaviour may be linked to courtship in basking sharks (Sims et al., 2000).
Courtship behaviour between basking sharks in the western English Channel over a five-year period
occurred between May and July (Sims et al., 2000). These observations are consistent with the
summer breeding period suggested by Matthews (1950) from anatomical studies (May), and for
observed breaching events (May and June) (Matthews & Parker, 1951). It also appears that basking
12
shark courtship events are significantly associated with seasonally persistent fronts rather than mixed
or stratified water (Sims et al., 2000). This spatial distribution was similar to that recorded for surface
foraging locations of this species (Sims & Quayle, 1998). Close-following behaviours were only
observed when large sharks were aggregated in relatively rich zooplankton patches, which indicated
patch aggregation and the resultant close proximity of mature individuals was a controlling factor in
whether courtship was observed (Sims et al., 2000). Therefore, courtship probably occurs as a
consequence of individuals aggregating to forage in rich prey patches before initiating courtship. This
suggests that locating the richest prey patches along fronts may be important for basking sharks to
find mates as well as food in the pelagic ecosystem. As courtship-like behaviours occur annually off
southwest England, this region may represent an annual breeding area for this protected species,
although mating itself probably takes place at depth as it has yet to be observed at the surface (Sims
et al., 2000).
Furthermore, as courtship and foraging occurs at the surface annually between May and July near
fronts that are close to shore, there is the potential that these important behaviours may be at future
risk of increased disturbance from anthropogenic sources, such as commercial shipping, leisure and
ecotourism vessels.
a
Fig. 7 Close flank approach and
following behaviour between two
C. maximus off Plymouth (a), and
the frequency distribution of
behaviour type observed between
lead and following sharks (b). From
Sims et al. (2000).
)
nd
be
BL
0
2.
.5
(1
in
g
(1
llo
w
in
g
Fo
lo
w
Fo
l
hi
nd
hi
be
1.
.0
-
1.
.5
(0
in
g
w
Fo
llo
)
)
nd
hi
BL
BL
be
eto
0
5
ac
h
N
os
pp
ro
in
ea
os
Cl
w
im
m
al
le
ls
Pa
r
-ta
il
20
18
16
14
12
10
8
6
4
2
0
g
Frequency
b
4.4.3 Local movements
Basking sharks have been tracked continuously over fine spatial scales (0.1 – 1.0 km) (Sims &
Quayle, 1998), but only intermittently over meso-scale (1.0 – 10 km) and broad-scale (10 to >100 km)
distances (Priede, 1984; Sims & Quayle, 1998). In tracking studies undertaken off Plymouth, three
basking sharks were relocated (separately) feeding in different zooplankton patches 18 – 28 h after
initial trackings and 5 – 11 km distant from the foraging areas of the previous day (Sims & Quayle,
13
1998). Two sharks that were originally found feeding in the same patch moved in similar directions
along a zooplankton gradient from low to higher density (range: 0.47 – 1.11 g m-3 to 1.06 – 1.43 g m-3),
covering minimum distances of 9.5 km and 10.6 km in 27.6 and 23 h respectively. A basking shark
tracked by satellite spent 17 days moving in an approximately circular course and showed no signs of
moving out of the Clyde Sea, Scotland (Priede, 1984). In support of this, individual sharks have been
resighted in the same area after periods of up to between 14 and 45 days in studies undertaken during
summer in the western English Channel and off Vancouver Island, Canada, respectively (Darling and
Keogh, 1994; D.W. Sims, unpublished data). These studies indicate that basking sharks move
between patches, probably in response to low prey densities encountered previously. Tracks of nonfeeding sharks demonstrate that they swim on relatively straight courses and at significantly higher
speeds after leaving patches where the zooplankton density has decreased to threshold levels (Sims
& Quayle, 1998; Sims, 1999). It seems foraging movements may keep them within a localised area for
some considerable time, but only if prey densities remain high.
4.4.4 Migrations
Basking sharks are rarely encountered outside of summer months and several theories have been
forwarded to account for this apparent disappearance. One historical theory suggested that basking
sharks migrate south at the end of the summer and spend the winter as a single population off the coast
of Morocco, before making the return journey into northern coastal waters in spring (Kunzlik, 1988;
Fairfax, 1998). However, this chronology of gradual appearance from the south in spring was disputed on
the grounds that sharks were not observed first off Portugal, then Spain, France, the British Isles and
Ireland, and finally off Norway as the season progressed (Stott, 1982). Subsequently, there was no
southward increase in abundance at the end of the summer and during early autumn as expected in this
scenario. Matthews and Parker (1950) proposed another theory based upon their own and historical
observations around Britain and Ireland. They suggested that because basking sharks appeared at
similar times off Ireland, southwest England and Scotland during early spring and summer, then a west to
east seasonal movement pattern was more likely than a south-north migration. This idea was supported
by the observation that C. maximus off the west coast of Ireland apparently arrive there to feed a few
weeks earlier than further east (Watkins, 1958; McNally, 1976). Since then, a number of studies have
speculated on the pattern of basking shark migration, but they all seem in agreement that a west to east
migration occurs in spring and the reverse course is taken by sharks in late summer (Kunzlik, 1988).
The high squalene content and large size of the liver in basking sharks (Blumer, 1967) was put forward as
evidence that they may occupy a seasonal deep-water habit because squalene is found in large
quantities only in the livers of deep-water sharks (Baldridge, 1972). The anatomical observations made by
Parker and Boeseman (1954) are described in detail in section 4.3. Briefly, they showed there to be a
lack of gill raker apparatus in winter-caught or stranded sharks, indicating a seasonal cessation of
feeding. They coupled this observation with calculations demonstrating that winter densities of
zooplankton would be too low to enable basking sharks to derive net energy gain outside summer
months. Furthermore, anecdotal information from fishermen suggested that shark livers in early summer
were lighter than in sharks taken later in the season (O’Connor, 1953). Taking this information together,
Parker and Boeseman (1954) and Matthews (1962) hypothesised that basking sharks undergo a winter
‘hibernation’ by migrating into deep water away from coastal areas at the end of summer. The
conjectured that by remaining inactive in deep, cold water in canyons on the continental slope they could
survive this non-feeding period by subsisting entirely on the energy reserves stored in their liver for the
five or more months before they emerge from this habitat to feed in spring in productive coastal areas
(Parker & Boeseman, 1954; Matthews, 1962). This interpretation of basking shark migration has
remained largely unchanged in the subsequent scientific literature and popular accounts for almost 50
years. There is still no direct evidence for hibernation in basking sharks. Similarly, there have been no
studies on the migratory movements of basking sharks over seasonal scales to test these theories
directly (Weihs, 1999).
14
As discussed in section 4.4.1, a recent behavioural and theoretical study showed that basking sharks
have much lower prey density thresholds than previously thought, suggesting that they are not reliant on
the ‘migration-hibernation’ strategy for energy conservation (Sims, 1999; Weihs, 1999). More recent
quantitative results of seasonal distribution patterns indicate the difference in appearance between west
Ireland and inshore areas further to the east to be less pronounced than stated previously (Berrow &
Heardman, 1994). Therefore, the west to east migration pattern, from deep to shallow shelf seas in spring
may not in fact be tenable. Despite this new information, there is some incidental evidence from
biochemical studies of shark tissue and trawling surveys that basking sharks do occupy deep water (>
200 m), perhaps moving from oceanic areas to shallow water in spring or early summer (Blumer, 1967;
Berrow, 1994; Francis & Duffy, 2002).
5. Population
5.1 Structure
The sex ratio from fisheries data indicates there to be approximately one male for every 18 females
(Watkins, 1958), which is rather less than the 30 to 40 females per male suggested by Matthews
(1950). There is no reason to expect a population-level deviation from a 1:1 sex ratio, so this disparity
in sex ratio may indicate pronounced spatial and seasonal segregation by sex (Compagno, 1984), or
may in fact be due to fishery bias towards surface basking individuals. It is possible that females may
engage in this activity more often than males and hence make up a greater percentage of the catch. In
contrast, examination of 128 individual sharks caught incidentally in inshore fishing gear in
Newfoundland, Canada, showed males comprised 70% of the sample (Lien & Fawcett, 1986). This
suggests that sexual segregation may well occur in basking shark populations, although evidence is
far from substantial.
The size composition of basking sharks from different geographic areas has not been studied in detail.
The size distribution of 93 sharks caught off Scotland in the 1950s ranged from 1.7 to 9.5 m total body
length (Parker & Stott, 1965). The size distribution was bimodal with peaks centred on sharks with
body lengths between 3 and 4 m, and between 7.5 and 9.0 m. Incidental catches of basking sharks off
Newfoundland showed that males ranged in size from 3.0 to 12.2 m body length with a mean length of
7.5 m (± 1.87 S.D.) (Lien & Fawcett, 1986). The size of females examined in the same study was
slightly smaller, with a mean length of 6.9 m (± 1.82 m S.D.; range, 2.4 to 10.7 m). Recent fisheryindependent studies of size composition indicate that sharks sighted in the western English Channel
range from 1.5 to 7.5 m body length, but that the distribution is unimodal with individuals between 4
and 5 m being most common (Sims et al., 1997; D.W. Sims, unpublished data).
Apart from some observations on sex ratio and size composition, the structure of basking shark
populations has not been studied. It is not known whether basking sharks form separate populations in
the north and south Atlantic, and whether these in turn are different from sharks found in the north and
south Pacific Ocean. It has yet to be established whether basking sharks in the Mediterranean
(Valeiras et al., 2001) are a distinct ‘stock’ compared to other members of the species found
throughout the remaining north Atlantic Ocean. Attempts to separate basking sharks found in each of
these five regions into separate species according to apparent morphological differences have been
rejected (Springer & Gilbert, 1976; Kunzlik, 1988).
It has been strongly suggested that basking sharks form local populations or stocks (Parker & Stott,
1965). As with hibernation, this idea has perpetuated in the literature despite there being only
circumstantial evidence supporting it. Local stocks have been proposed for basking sharks on account
of sharply declining fishery catches in certain, spatially-limited areas, e.g. Keem Bay on Achill Island,
west Ireland (Parker & Stott, 1965; McNally, 1976). The rapid decline in the number of sharks caught
in Keem Bay after only about 10 years of the fishery commencing was interpreted as over-exploitation
of a limited stock of sharks inhabiting a locally discrete area annually. This hypothesis seems to be
15
supported by the observations that individually-identifiable basking sharks remain in localised areas
often for many days (e.g. Darling & Keogh, 1994). However, this apparent resident is probably more
closely related to high zooplankton abundance than with population structure (see sections 4.4.1 and
4.4.3 for discussion). The prospect of highly philopatric stocks existing along the entire western-shelf
edge of the northeast Atlantic, and which remain faithful to the same bays for summer feeding year
after year seems rather improbable when compared with the behaviour patterns of better-known
pelagic filter-feeders. Satellite tracking studies of whale sharks (Rhincodon typus) show wide dispersal
over ocean-basin scales (Eckert & Stewart, 2001). One individual tracked for 37 months moved a
minimum distance of 12,620 km, whilst distances of between 60 and 820 km were covered over time
periods of 1 to 39 days. It seems whale sharks show migratory movements consistent with locating
areas of high productivity, which in reality may be very widely spaced.
For basking sharks, there is evidence that few if any individuals return to the same local areas in
consecutive years. In a 7-year study off Plymouth in the western English Channel, there have been no
identified sharks recorded in the same area in consecutive years, although at least one individual is
known to have returned after a two-year absence (Sims et al., 2000). Furthermore, a recent
investigation has demonstrated congruent trends between the long-term zooplankton decline in the
northeast Atlantic near Achill Island and the decline in fishery catches of basking sharks at that
location (Sims & Reid, 2002; see section 6.4 for discussion). Taken together, the available evidence
suggests against the formation and persistence of localised stocks in basking shark populations.
Without directed research to elucidate long-range movements and population genetic structure
however, the hypothesis that basking sharks form local stocks will no doubt persist.
5.2 Abundance and density
The population abundance and density of basking sharks in any sea area of the world is not precisely
known. Fishery catches provide information on the numbers caught in particular years, but an absence
of information on the variability in search times (fishing effort) prevents a systematic evaluation of
relative abundance by area or year (see section Exploitation). The best available assessment of
absolute basking shark abundance was provided by marine mammal aerial surveys flown between
October 1978 and January 1982 (Owen, 1984; Kenney et al., 1985). Individual counts of basking
sharks were made in U.S. continental shelf waters (shoreline to 9 km beyond the 1,829 m isobath) off
New England, northwest Atlantic (Hudson Canyon to the Gulf of Maine) (Kenney et al., 1985). These
surveys indicated an abundance there of between 6,671 and 14,295 animals. Similar aerial surveys
were flown along the central and southern U.S. Californian coast between 1962 and 1985 (Squire,
1990). The number of sharks sighted varied greatly between different ‘block’ areas (each block = 220
km2). Up to 6,389 sharks were observed over the 23 year study period in the Morro Bay area, with a
mean of 96.8 sharks per sighting. Lower numbers of sharks and fewer sharks per sighting occurred
north of Morro Bay towards Point Sur (between 1.0 and 9.5 sharks per sighting). Whereas in Monterey
Bay, there were between 14.4 and 42.1 sharks per sighting. Further south however, the greatest
number observed south of Point Conception was a mean of 6.7 individuals per sighting (Squire, 1990).
Over a smaller spatial scale, a sightings scheme was established in Ireland, mainly from fishing boats,
to determine the distribution and abundance of sharks throughout Irish waters (Berrow & Heardman,
1994). The results showed that basking sharks were sighted only between April and October, with the
number seen per month ranging from between 1 and 60 individuals. The total number sighted in 1993
was 425 individuals, and the abundance of sharks ranged from 1 to > 40 per 2500 km2 area (Berrow &
Heardman, 1994).
Basking shark abundance can be very high in productive inshore areas and will probably determine to
a large degree the surface sightings of sharks (Sims et al., 1997; Darling & Keogh, 1994). Annual
studies operating over small spatial scales (~500 km2) in specific locations have provided information
on the number of individual sharks observed per unit time (e.g. Sims et al., 1997). In the western
English Channel off Plymouth, the number of sharks observed from May to August in each year
16
between 1995 and 2001 varied from 0.01 to 0.35 per hour (D.W. Sims, unpublished data). The years
1998 and 1999 yielded uncharacteristically few sightings (0.01 and 0.02 per h), compared to 19951997 (0.10 to 0.35 per h) and 2000/2001 (0.30 and 0.14 per h). The abundance of basking sharks
over these years have been related to prey density, with a higher number per h observed in years
when the zooplankton density was high at the surface (D.W. Sims, unpublished data). As discussed in
section 4.4.1, the abundance of zooplankton must be assessed in parallel with surveys for basking
sharks if the method of finding sharks depends upon their surface occurrence.
5.3 Recruitment
The number of female basking sharks in all sea areas of the world remains completely unknown.
Similarly, because to date there has been only a single capture of a pregnant female (Sund, 1943),
estimates of fecundity and hence probable recruitment rates are extremely difficult. The pregnant
female captured in August off central Norway gave birth to six pups, five of which began swimming
and feeding at the surface almost immediately (Sund, 1943). If this number of pups is representative
of normal parturition rates, then the rate of recruitment in basking sharks must be considered to be low
even compared to other shark species (Pratt & Casey, 1990). In a preliminary demographic analysis,
an effective annual female fecundity of 1.0 was determined using an assumption that each female
gives birth to six pups every third year (Mollet, 2001).
It is generally agreed that total body length at parturition probably lies between 1.5 and 2.0 m (Sund,
1943; Parker & Stott, 1965). The frequency with which putative young-of-the-year basking sharks of
this body length are sighted undoubtedly varies between years, but they were shown to never make
up more than 2.8 % of all sightings in the western English Channel off Plymouth (Sims et al., 1997). In
a study of the incidental catches of C. maximus in inshore fishing gear in Newfoundland, immature
sharks made up only 2.6 % of captures (Lien & Fawcett, 1986). Interestingly, the frequency of
sightings and capture of small-bodied basking sharks was very similar between these two studies in
the north Atlantic. In addition, it was shown in both studies that these young sharks only occurred later
in the summer.
5.4 Mortality
The natural mortality rates of basking sharks are not known for any geographic region. In a preliminary
demographic analysis for the basking shark, a mortality value (M) of 0.10235 yr-1 (S = 0.9027) was
assumed by Mollet (2001) based on a longevity of 40 years and using the fecundity estimate of 1 per
female per year. However, this author concluded that a lack of information on fecundity was not
particularly problematic, rather, better mortality data and a better estimate of age-at-maturity was
urgently needed.
Little is known about natural mortality from behavioural studies for obvious reasons associated with
the difficulty of prolonged observation. But because of their large body size, natural mortality of adult
basking sharks through predation is probably quite low. There have been anecdotal reports from
fishermen in southwest England that killer whales (Orcinus orca) sometimes predate on basking
sharks. There is a record of a small, 2-3-m long basking shark being found in the stomach contents of
a sperm whale (Physeter macrocephalus) caught off the Azores (Clark, 1956). However, such records
in the literature are extremely rare.
6. Exploitation
6.1 Fishing gear and boats
The gear and boats used to hunt basking sharks have been reviewed in considerable detail by
Myklevoll (1968), Kunzlik (1988) and most recently by Fairfax (1998). Briefly, most fishing operations
have utilised harpoons or harpoon guns mounted on boats to catch basking sharks, although one
fishery used tethered nets across an embayment to entangle sharks (Went & Suilleabhain, 1967).
17
6.2 Fishing areas and seasons
Basking sharks have been exploited by organised fisheries dating back to at least the 18th Century.
Several nations have prosecuted fisheries at the time when basking sharks are present in inshore
areas, which in the northeast Atlantic occurs from April to September. Fisheries have operated off the
U.S. Californian coast, and perhaps most importantly in the northeast Atlantic, have been undertaken
annually by Norway, Ireland and Scotland (Kunzlik, 1988). The Norwegian fleet, which by 1987
numbered only seven boats, was known to hunt for basking sharks throughout the Norwegian Sea,
and in areas around Scotland and Ireland outside the 12 miles belt. This was not always the case
however, because Norwegian boats were frequently observed catching sharks in the Minch in
Scotland during early 1950s (O’Connor, 1953).
6.3 Fishing results
Between 1946 and 1986, directed basking shark fisheries in Norway, Scotland and Ireland took a
recorded 77,204 individuals (mean number per year, range, 164 – 1,495) (Kunzlik, 1988). In more
recent years between 1989 and 1997, Norway landed 14,263 metric tonnes (mt) of basking shark liver
(FAO, 2000). Assuming a mean liver weight of 0.5 mt per shark, this gives the number caught over
this 9-year period as approximately 28,526 individuals. Taken together, the landings records in the
northeast Atlantic indicate that 105,730 sharks were captured over a 51 year period. Clearly, without
any knowledge of population size and inter-annual fluctuations in abundance there is no way of
assessing whether capture rates were high in relation to population numbers in any one year. The
geographical areas in which sharks were taken between 1946 and 1997 varied between years
indicating that broad-scale locations of aggregations may also have changed between years in the
northeast Atlantic.
Less is known about the number of basking sharks caught incidentally in fishing gear where other
species were the primary target. In Newfoundland between 1980 and 1983, 371 basking sharks were
captured in inshore fishing gear (Lien & Fawcett, 1986). By contrast, in over forty gill-netting fishing
trips in Irish waters between February 1993 and January 1994, totalling 1,167 km and 19,760 km h
observed fishing effort, only one basking shark was caught incidentally (Berrow, 1994). This large
difference in capture rate may reflect geographic differences in shark numbers, variations in the
amount of gear deployed, and/or the fishing method employed.
6.4 Decline in numbers
An example often cited as demonstrating clear evidence for over-fishing of basking sharks (e.g.
Anderson, 1990) was the fishery conducted at Achill Island, Co. Mayo, Republic of Ireland, between
1947 and 1975. After a few years of peak catches in the early 1950s the number of sharks captured at
the surface (using harpoons and nets) declined sharply (Kunzlik, 1988). Between 1947 and 1975,
there were 12,360 sharks taken in the fishery. The number caught over the period 1950-56 accounted
for 75% of this total (mean, 1,323 sharks yr-1 ± 380 S.D.), whereas between 1957 and 1961 a mean
of 345 sharks were caught per year (± 129 S.D.), and from 1962 to 1975 the mean number caught
declined to 60 per year (± 29 S.D.) (Fig. 8). This downward trend was suggested as a result of a stock
collapse due to over-exploitation of a localised population (Parker & Stott, 1965). A newly-published
study however, related the trend in basking shark fishery catches off Achill Island to zooplankton (total
copepod) abundance in four adjacent sea areas over a 27-year period (Sims & Reid, 2002). The
number of basking sharks caught and copepod abundance showed similar downward trends and were
positively correlated (r-value range, 0.44-0.74) (Fig. 8). A possible explanation for the downward trend
in shark catches was that progressively fewer basking sharks occurred there between 1956 and 1975
because fewer copepods, their food resource, occurred near the surface off west Ireland over the
same period. It was suggested by the latter authors that the decline in basking sharks may have been
due to a distributional shift of sharks to more productive areas, rather than a highly philopatric,
localised stock that was over-exploited (Sims & Reid, 2002). In support of this conclusion, Sims and
Reid (2002) note that the catches of basking sharks in the Norwegian Sea, the main hunting ground
18
for the Norwegian fleet (Myklevoll, 1968), remained relatively low between 1949 and 1958 when
catches were highest off Achill Island. However, after 1958 the Norwegian catches increased to levels
greater than those made off Achill, and remained fairly constant until 1980 (Sims & Reid, 2002). This
may indicate that basking shark distribution shifted northwards in the mid-1950s, perhaps to areas
with relatively higher utilisable productivity.
3
2.5
1
0.8
b.
3
2.5
0.8
0.6
0.6
1.5
1
1.5
0.4
Area C5
r = 0.44
0.2
0.5
0
1
3.5
c.
3
0.2
0
0
1949 1952 1955 1958 1961 1964 1967 1970 1973
1
3.5
1.4
d.
3
1.2
0.8
2.5
2.5
0.6
2
1.5
0.5
0.4
Area C4
r = 0.58
0.5
1949 1952 1955 1958 1961 1964 1967 1970 1973
1
1
2
2
0
C. maximus caught (log10(x+1))
1.2
3.5
0.4
Area D5
r = 0.58
0
1
2
0.8
1.5
0.6
1
0.2
0
1949 1952 1955 1958 1961 1964 1967 1970 1973
Total copepod abundance (log10(x+1))
1.2
a.
Area D4
r = 0.74
0.4
0.2
0.5
0
0
Total copepod abundance (log10(x+1))
C. maximus caught (log10(x+1))
3.5
1950 1953 1956 1959 1962 1965 1968 1971 1974
Fig. 8 The relationships between numbers of basking sharks caught of west Achill Island (continuous
line) and abundance of total copepods (dashed line) in Continuous Plankton Recorder (CPR) areas
C4, C5, D4 and D5. The CPR areas were the four surrounding Achill Island. From Sims & Reid (2002).
In other parts of the world however, there is some evidence to indicate that basking shark populations
may take very many years to recover from exploitation. Basking sharks were the subject of an
eradication program in Barkley Sound, Vancouver Island, Canada in the 1940s and 1950s (Darling &
Keogh, 1994). The program was set up by the Canadian Department of Fisheries and Oceans and
entailed sharks being rammed by a fishery vessel armed with a blade mounted on the bow below the
waterline. About 100 sharks were killed in the summers of 1955 and 1956, with perhaps several
hundred being killed in the area up to 1959 (Darling & Keogh, 1994). Apparently, basking sharks are
still rarely observed in Barkley Sound or in other areas of Vancouver Island, although Darling and
Keogh (1994) describe a small population in Clayoquot Sound. It is unclear whether the eradication
program was responsible for the decline and persistent low number of sharks seasonally present off
Vancouver Island in the years following the program, or whether other factors such as food availability
are responsible.
19
7. Management and protection
7.1 Management
Membership of the U.K. and Ireland in the European Union (EU) and the extension of fishery limits
formalised Norwegian fisheries in these countries waters (Kunzlik, 1988). The Norwegian catch of
basking sharks in EU waters has been limited since 1978 by quota which stands at approximately 400
mt of liver weight annually. This quota amounts to approximately 800 individual basking sharks
annually.
7.2 Protection
Sharks and rays are particularly vulnerable to exploitation on account of slow growth rates, long times
to sexual maturity, long gestation periods, and relatively low fecundity (Brander, 1981; Pratt & Casey,
1990; Dulvy et al., 2000). The basking shark may take as long as 10-12 years to reach sexual
maturity, probably has a gestation period of between 1 and 2 years, and has a very low fecundity rate
even among elasmobranchs. Because of these basic aspects of its biology, there has been concern
that past fishing activities may have affected populations.
As a result of these concerns, and predominantly under the precautionary principle, the basking shark
is listed as Vulnerable in the 1996 World Conservation Union (IUCN) Red List of Threatened Animals.
In April 1998 it was designated a protected species in U.K. waters within 12 miles of the shore under
the Wildlife and Countryside Act (1981), and has been listed on Appendix III of CITES by the U.K.
National protection has also been afforded to this species in waters surrounding the Isle of Man (12
mile limit), Guernsey in the Channel Islands, in U.S. Florida state waters (3-9 miles), U.S. Atlantic and
Gulf federal waters (3-200 miles), and in New Zealand. The basking shark is listed for international
protection in the Mediterranean but the convention concerned has yet to be ratified.
8. References
Baldridge, H.D. (1972) Accumulation and function of liver oil in Florida sharks. Copeia 1972: 306-325.
Berrow, S.D. (1994) Incidental capture of elasmobranchs in the bottom-set gill-net fishery off the South
coast of Ireland. Journal of the Marine Biological Association of the U.K. 74: 837-847.
Berrow, S.D. & Heardman, C. (1994) The basking sark Cetorhinus maximus (Gunnerus) in Irish
waters - Patterns of distribution and abundance. Biology And Environment - Proceedings Of
The Royal Irish Academy 94B: 101-107.
Bigelow, H.B. & Schroeder, W.C. (1948) Fishes of the western North Atlantic. Memoir Sears
Foundation for Marine Research, 1, 576 pp.
Blumer, M. (1967) Hydrocarbons in digestive tract and liver of a basking shark. Science 156: 390-391.
Bone, Q. and Roberts, B.L. (1969) The density of elasmobranchs. Journal of the Marine Biological
Association of the U.K. 49: 913-937.
Brander, K. (1981) Disappearance of common skate Raia batis from Irish Sea. Nature 290: 48-49.
Cailliet, G.M. (1990) Elasmobranch age determination and verification: An updated review. In:
Elasmobranchs as living resources: advances in the biology, ecology systematics and status of
fisheries (ed. H.L. Pratt, S.H. Gruber and T. Tanuichi) pp. 157-165. NOAA Technical Report
90. Seattle, WA: National Oceanographic and Atmospheric Administration.
20
Carrier, J.C., Pratt, H.L. & Martin, L.K. (1994) Group reproductive behaviours in free-living nurse
sharks, Ginglymostoma cirratum. Copeia 1994: 646-656.
Choy, B.K. and Adams, D.H. (1995) An observation of a basking shark, Cetorhinus maximus, feeding
along a thermal front off the east-central coast of Florida. Florida Scientist 58: 313-319.
Clark, E. and Nelson, D.R. (1997) Young whale sharks, Rhincodon typus, feeding on a copepod
bloom near La Paz, Mexico. Environmental Biology of Fishes 50: 63-73.
Clark, R. (1956) The biology of sperm whales in the Azores. Discovery Reports 28: 1-200.
Compagno, L.J.V. (1984) FAO species catalogue. IV. Sharks of the world. 1. Hexanchiformes to
Laminiformes. Rome: Food and Agriculture Organisation of the United Nations.
Compagno, L.J.V. (1990) Relationships of the megamouth shark, Megachasma pelagious
(Lamniformes: Megachasmidae), with comments on its feeding habits. In: Elasmobranchs as
living resources: advances in the biology, ecology systematics and status of fisheries (ed. H.L.
Pratt, S.H. Gruber and T. Tanuichi) pp. 357-379. NOAA Technical Report 90. Seattle, WA:
National Oceanographic and Atmospheric Administration.
Darling, J.D. and Keogh, K.E. (1994) observations of basking sharks, Cetorhinus-maximus, in
Clayoquot Sound, British Columbia. Canadian Field-Naturalist 108: 199-210.
Diamond, J.M. (1985) Filter-feeding on a grand scale. Nature 316: 679-680.
Digby, P.S.B. (1950) The biology of the small plantonic copepods of Plymouth. Journal of Marine
Biological Association of the U.K.29:393-438.
Dulvy, N.K. & Reynolds, J.D. (1997) Evolutionary transitions among egg-laying, live-bearing and
material inputs in sharks and rays. Proceedings of the Royal Society of London Series B Biological Sciences. 264: 1309-1315.
Dulvy, N.K., Metcalfe, J.D., Glanville, J., Pawson, M. & Reynolds, J.D. (2000) Fishery stability, local
extinctions and shifts in community structure in skates. Conservation Biology 14: 283-293.
Eckert, S.A. & Stewart, B.S. (2001) Telemetry and satellite tracking of whale sharks, Rhincodon typus,
in the Sea of Cortez, Mexico and the north Pacific Ocean. Environmental Biology of Fishes 60:
299-308.
Fairfax, D. (1998) The Basking Shark in Scotland. Tuckwell Press: Scotland.
Francis, M.P. & Duffy, C. (2002) Distribution, seasonal abundance and bycatch of basking sharks
(Cetorhinus maximus) in New Zealand, with observations on their winter habitat. Marine
Biology, in press.
Gerking, S.D. (1991) Feeding ecology of fish. Academic Press: San Diego.
Gordon, I. (1993) Pre-copulatory behaviour of captive sandtiger sharks, Carcharias taurus.
Environmental Biology of Fishes 38: 159-164.
Hallacher, L.E. (1977) On the feeding behaviour of the basking shark, Cetorhinus maximus.
Environmental Biology Fishes 2: 297-298.
21
Harden-Jones, F.R. (1973) Tail beat frequency, amplitude, and swimming speed of a shark tracked by
sector scanning sonar. Journal du Consiel pour l’International Exploration de la Mer 35: 95-97.
Harvey, H.W., Cooper, L.H.N., Lebour, M.V. & Russell, F.S. (1935) Plankton production and its control.
Journal of the Marine Biological Association of the U.K. 20: 407-441.
Harvey-Clark, C.J., Stobo, W.T., Helle. E. & Mattson, M. (1999) Putative mating behavior in basking
sharks off the Nova Scotia coast Copeia 1999: 780-782.
Holden, M.J. (1974) Problems in the rational exploitation of elasmobranch populations and some
suggested solutions. In: Sea Fisheries Research (F.R. Harden Jones, ed.). Paul Elek.
IUCN, 1996. IUCN Red List of Threatened Animals. Gland, Switzerland: International Union for the
Conservation of Nature and Natural Resources.
Izawa, K. & Shibata, T. (1993) A young basking shark, Cetorhinus maximus, from Japan. Japanese
Journal Of Ichthyology 40: 237-245.
Johnson, R.H. & Nelson, D.R. (1978) Copulation and possible olfaction-mediated pair formation in two
species of carcharhinid sharks. Copeia 1978: 539-542.
Joung, S.J., Chen, C.T., Clark, E., Uchida, S. & Huang, W.Y.P. (1996) The whale shark, Rhincodon
typus, is a livebearer- 300 embryos found in one ‘megamamma’ supreme. Environmental
Biology of Fishes 46: 219-223.
Kenney, R.D., Owen, R.E & Winn, H.E. (1985) Shark distributions off Northeast United States from
marine mammal surveys. Copeia 1985: 220-223.
Klimley, A.P. (1980) Observations of courtship and copulation in the nurse shark, Ginglymostoma
cirratum. Copeia 1980: 878-882.
Klimley, A.P. (1987) The determinants of sexual segregation in the scalloped hammerhead shark,
Sphyrna lewini. Environmental Biology of Fishes 18: 27-40.
Konstantinov, K.G. & Nizovtsev, G.P. (1980) The basking shark Cetorhinus maximus, in Kandalaksha
Bay of the White Sea. Journal of Ichthyology 19: 155-156.
Kruska, D.C.T. (1988) The brain of the basking shark (Cetorhinus maximus) Brain Behavior and
Evolution 32: 353-363.
Kunzlik, P.A. (1988) The basking shark. Aberdeen, UK: Department of Agriculture and Fisheries for
Scotland.
Le Fèvre, J. (1986) Aspects of the biology of frontal systems. Advances in Marine Biology 23: 163-299.
Lien, J. & Fawcett, L. (1986) distribution of basking sharks, Cetorhinus-maximus, incidentally caught in
inshore fishing gear in Newfoundland. Canadian Field-Naturalist 100: 246-252.
Maisey, J.G. (1985) Relationships of the megamouth shark, Megachasma. Copeia. 1985: 228-231.
Martin, A.P. & Naylor, G.J.P. (1997) Independent origins of filter-feeding in megamouth and basking
sharks (Order Lamniformes) inferred from phylogenetic analysis of cytochrome b gene
22
sequences. In: Biology of the Megamouth Shark (Yano, K., Morrissey, J.F., Yabumoto, Y.,
Nakaya, K., eds.), pp. 39-50. Tokai University Press: Tokyo.
Martin, A.P., Naylor, G.J.P. & Palumbi, S.R. (1992) Rates of mitochondrial DNA evolution in sharks
are slow compared with mammals. Nature 357: 153-155.
Matthews, L.H. (1950) Reproduction in the basking shark Cetorhinus maximus (Gunner).
Philosophical Transactions of the Royal Society of London B 234: 247-316.
Matthews, L.H. (1962) The shark that hibernates. New Scientist 280: 756-759.
Matthews, L.H. & Parker, H.W. (1950) Notes on the anatomy and biology of the basking shark
Cetorhinus maximus (Gunner). Proceedings of the Zoological Society of London 120: 535-576.
Matthews, L.H. & Parker, H.W. (1951) Basking sharks leaping. Proceedings of the Zoological Society
of London 121: 461-462.
Maxwell, G. (1952) Harpoon at a venture. R. Hart-Davis: London.
McNally, K. (1976) The Sun-Fish Hunt. Blackstaff Press: Belfast.
Mollet, H. (2001) Demographic analysis of the basking shark Cetorhinus maximus (Gunnerus, 1765).
Website publication: http://homepage.mac.com/mollet/cm/demography.html
Mutoh, M. & Omori, M. (1978) Two records of patchy occurrences of the ocean shrimp Sergestes
similis (Hansen) off east coast of Honshu, Japan. Journal of the Oceanographical Society of
Japan 34: 36-38.
Myklevoll, S. (1968) Basking shark fishery. Commerical Fisheries Review 30: 59-63.
Myrberg, A.A. & Gruber, S.H. (1974) The behaviour of the bonnethead, Sphyrna tiburo. Copeia 1974:
358-374.
Natanson, L.J. & Cailliet, G.M. (1990) Vertebral growth zone deposition in Pacific Angel sharks.
Copeia 1990: 1133-1145.
Naylor, G.J.P., Martin, A.P., Mattison, E.G. & Brown, W.M. (1997) The interrelationships of lamniform
sharks: testing phylogenetic hypotheses with sequence data. In: Molecular systematics of
fishes (Kocher, T.D. & Stephian, C.A., eds.), pp. 195-214. Academic Press: London.
O’Connor, P.F. (1953) Shark-O! Secker & Warburg: London.
Parker, H.W. & Boeseman, M. (1954) The basking shark (Cetorhinus maximus) in winter. Proceedings
of the Zoological Society of London 124: 185-194.
Parker, H.W. & Stott, F.C. (1965) Age, size and vertebral calcification in the basking shark, Cetorhinus
maximus (Gunnerus). Zoologische Mededelingen, Leiden 40: 305-319.
Pauly, D. (1978) A critique of some literature data on the growth, reporduction and mortality of the
lamnid sharks Cetorhinus maximus (Gunnerus). International Council for the Exploration of the
Sea. Council Meeting 1978/H:17 Pelagic Fish Committee, 10 pp.
23
Pawson, M. & Vince, M. (1999) Management of shark fisheries in the northeast Atlantic. Case studies
of the management of elasmobranch fisheries (Shotton, R., ed.). pp. 1-46, FAO Fisheries
Technical Paper 378/1. Food and Agriculture Organization of the United Nations: Rome.
Pratt, H.L. & Casey, J.G. (1990) Shark reproductive strategies as a limiting factor in directed fisheries,
with a review of Holden’s method of estimating growth parameters. In: Elasmobranchs as living
resources: advances in the biology, ecology systematics and status of fisheries (ed. H.L. Pratt,
S.H. Gruber and T. Tanuichi) pp. 97-109. NOAA Technical Report 90. Seattle, WA: National
Oceanographic and Atmospheric Administration.
Priede, I.G. (1984) A basking shark (Cetorhinus maximus) tracked by satellite together with
simultaneous remote-sensing Fisheries Research 2: 201-216.
Pyle, P., Anderson, S.D., Klimley, A.P. & Henderson, R.P. (1996) Environmental factors affecting the
occurrence and behavior of white sharks at the Farallon Islands, California. In: Great white
sharks: the biology of Carcharodon carcharias (Klimley, A.P. & Ainley, D.G., eds.), pp. 281291. Academic Press: San Diego.
Sanderson, S.L., Cech, J.J. & Cheer, A.Y. (1994) Paddlefish buccal flow velocity during ram
suspension feeding and ram ventilation. Journal of Experimental Biology 186:145-156.
Sanderson, S.L., Cheer, A.Y., Goodrich, J.S., Graziano, J.D. & Callan, W.T. (2001) Crossflow filtration
in suspension-feeding fishes. Nature 412:439-441.
Sims, D.W. (1999) Threshold foraging behaviour of basking sharks on zooplankton: life on an
energetic knife-edge? Proceedings Of The Royal Society Of London Series B-Biological
Sciences 266: 1437-1443.
Sims, D.W. (2000a) Filter-feeding and cruising swimming speeds of basking sharks compared with
optimal models: they filter-feed slower than predicted for their size Journal of Experimental
Marine Biology and Ecology 249: 65-76.
Sims, D.W. (2000b) Can threshold foraging responses of basking sharks be used to estimate their
metabolic rate? Marine Ecology Progress Series 200: 289-296.
Sims, D.W. & Merrett, D.A. (1997) Determination of zooplankton characteristics in the presence of
surface feeding basking sharks Cetorhinus maximus. Marine Ecology Progress Series 158:
297-302.
Sims, D.W., Fox, A.M. & Merrett, D.A. (1997) Basking shark occurrence off south-west England in
relation to zooplankton abundance. Journal of Fish Biology 51: 436-440.
Sims, D.W. & Quayle, V.A. (1998) Selective foraging behaviour of basking sharks on zooplankton in a
small-scale front. Nature 393: 460-464.
Sims, D.W., Southall, E.J., Quayle, V.A. & Fox, A.M. (2000a) Annual social behaviour of basking
sharks associated with coastal front areas Proceedings of the Royal Society of London, Series
B - Biological Sciences 267:1897-1904.
Sims, D.W., Speedie, C.D. & Fox, A.M. (2000b) Movements and growth of a female basking shark resighted after a three year period. Journal of the Marine Biological Association of the U.K. 80:
1141-1142.
24
Sims, D.W. & Reid, P.C. (2002) Congruent trends in long-term zooplankton decline in the north-east
Atlantic and basking shark (Cetorhinus maximus) fishery catches off west Ireland. Fisheries
Oceanography 11: 59-63.
Springer, S. and Gilbert, P.W. (1976) The basking shark Cetorhinus maximus, from Florida and
California, with comments on its biology and systematics. Copeia 1976: 47-54.
Squire, J.L. (1990) Distribution and apparent abundance of the basking shark Cetorhinus maximus off
the central and southern California coast 1962-1985. Marine Fisheries Review 52: 8-11.
Stevens, J.D. (1975) Vertebral rings as a means of age determination in the blue shark (Prionace
glauca L.). Journal of the Marine Biological Association of the U.K. 55: 657-665.
Stockin, K.A., Fairbairns, R.S., Parsons, E.C.M., & Sims, D.W. (2001) Effects of diel and seasonal
cycles on the dive duration of the minke whale (Balaenoptera acutorostrata). Journal of the
Marine Biological Association of the U.K. 81: 189-190.
Stott, F.C. (1980) A note on the spaciousness of the cavity around the brain of the basking shark,
Cetorhinus maximus (Gunnerus). Journal of Fish Biology 16: 665-667.
Stott, F.C. (1982) A note on catches of basking sharks, Cetorhinus maximus (Gunnerus), off Norway
and their relation to possible migration paths Journal of Fish Biology 21: 227-230.
Sund, O. (1943) Et brugdebarsel. Naturen 67: 285-286.
Taylor, L.R., Compagno, L.J.V. & Struhsaker, P.J. (1983) Megamouth: a new species, genus and
family of lamnoid shark (Megachasma pelagios, family Megachasmidae) from Hawaiian
islands. Proceedings of the Californian Academy of Sciences 43: 87-110.
Tomás, A.R.G. & Gomes, U.L. (1989) Observacoes sobre a presenca de Cetorhinus maximus
(Gunnerus, 1765) (Elasmobranchii, Cetorhinidae) no sudeste a sul do Brasil. Bolletin Inst.
Pesca – Sao Paulo 16: 111-116.
Valeiras, J. Lopez, A. & Garcia, M. (2001) Geographical seasonal occurrence and incidental fishing
captures of basking shark Cetorhinus maximus (Chondricthyes : Cetorhinidae) Journal of the
Marine Biological Association of the U.K. 81: 183-184.
Van Deinse, A.B. & Adriani, M.J. (1953) On the absence of gill rakers in specimens of basking shark,
Cetorhinus maximus (Gunner). Zoologische Mededelingen, Leiden 31: 307-310.
Watkins, A. (1958) The sea my hunting ground. William Heinemann: London.
Weihs, D. (1999) Marine biology: No hibernation for basking sharks. Nature 400: 717-718.
Went, A.E.J. & Suilleabhain, S.O. (1967) Fishing for the sun-fish or basking shark in Irish waters.
Proceedings of the Royal Irish Academy C 65: 91-115.
Whitehead, H. (1985) Why whales leap. Scientific American 252: 70-75.
Wintner, S.P. (2000) Preliminary study of vertebral growth rings in the whale shark, Rhincodon typus,
from the east coast of South Africa. Environmental Biology of Fishes 59: 441-451.
25
Wolanski, E. & Hamner, W.M. (1988) Topographically controlled fronts in the ocean and their biological
significance. Science 241: 177-181.
Wood, C.M. & McDonald, D.G. (eds) (1997) Global warming: implications for freshwater and marine fish.
Cambridge University Press: Cambridge.
Wood, F.G. (1957) Southern extension of the known range of the basking shark, Cetorhinus maximus
(Gunnerus). Copeia 1957: 153-154.
Young, E. (2001) Minke whales out for the count. New Scientist 16 June, No. 2295: 12.
Yudin, K.G. & Cailliet, G.M. (1990) Age and growth of the gray smoothhound, Mustelus californicus,
and the brown smoothhound, M. henlei, sharks from central California. Copeia 1990: 191-204.
26
Appendix II
Seasonal movements and behaviour of basking sharks from
archival tagging: no evidence of winter hibernation.
D.W. Sims, E.J. Southall, A.J. Richardson, P.C. Reid and J.D. Metcalfe,
2003, Marine Ecology Progress Series, 248: 187-196
vii
The Centre for Environment, Fisheries & Aquaculture Science
Lowestoft Laboratory, Pakefield Road,
Lowestoft, Suffolk NR33 0HT UK
Tel: +44 (0) 1502 562244
Fax: +44 (0) 1502 513865
www.cefas.co.uk