Satellite tracking of Western Australian humpback whales Page 1 Satellite tracking of northbound humpback whales (Megaptera novaeangliae) off Western Australia Final Report – May 2012 M.C. Double 1 , K.C.S. Jenner 2 , M-N. Jenner 2 , I. Ball 1 , S. Childerhouse 1 , S. Laverick 1 , N. Gales 1 1 Australian Marine Mammal Centre, Australian Antarctic Division, 203 Channel Highway, Kingston, Tasmania 7150 2 Centre for Whale Research (Western Australia) Inc., PO Box 1622, Fremantle WA 6959
23
Embed
Final Report - Satellite Tracking North Bound WA Humpback Whales (2)
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Satellite tracking of Western Australian humpback whales
Page 1
Satellite tracking of northbound
humpback whales (Megaptera
novaeangliae) off Western Australia
Final Report – May 2012
M.C. Double1, K.C.S. Jenner2, M-N. Jenner2, I. Ball1, S. Childerhouse1, S. Laverick1,
N. Gales1
1 Australian Marine Mammal Centre, Australian Antarctic Division, 203 Channel
Highway, Kingston, Tasmania 7150 2 Centre for Whale Research (Western Australia) Inc., PO Box 1622, Fremantle
WA 6959
Satellite tracking of Western Australian humpback whales
Page 2
Satellite tracking of south-bound female humpback whales in the
Kimberley region of Western Australia
Summary
This study was the first large-scale deployment of satellite tags on northbound humpback
whales and aimed to describe the migratory distribution and behaviour of these whales off
north-western Australia. A total of 28 tags were successfully deployed on migrating humpback
whales between the 8th and the 23rd July 2011. Of these tags, 15 performed poorly providing no
uplinks, only Z class data or the tag ceased transmitting within the first six days of deployment.
Whales were tracked from one to 47 days (mean = 16.04 days; SD = 15.1) for a total of 28,893
km (mean = 1,256 km; max: 3,612) and the total net distance moved from the first to the last
location was 11,702 km (mean = 509 km; max: 1,283). The tagged whales revealed the
consistently narrow, inshore distribution of migrating whales along the north-western coast of
Australia. Surprisingly many whales terminated their northern migration well short of the
recognised calving and breeding area for this population. Also while deploying tags many
females with calves were sighted off North West Cape. These observations highlight the diversity
of migration and calving behaviour that occurs through this region of Western Australia.
Satellite tracking of Western Australian humpback whales
Page 3
Introduction
The population of humpback whales that migrate along the west coast of Australia and calve in
the Kimberley region of Western Australia is increasing in size but has not yet fully recovered
from industrial whaling (Hedley et al., 2009). The latest abundance estimate suggests this
population is now over 20,000 individuals and is probably increasing at a rate greater than 10%
per year (Bannister & Hedley, 2001; Hedley et al., in press; Hedley et al., 2009). Although the
general timing of the migration and approximate distribution of this population of humpback
whales has been broadly described previously (Jenner et al., 2001), the specific migratory
routes, habitat use, resting areas, calving areas and the individual behaviours of the males,
females and mothers with calves are not well described.
Satellite telemetry has been used routinely in wildlife biology for nearly two decades and has
been applied successfully in studies of many marine organisms including penguins, albatrosses,
seals and even sharks (Bonadonna et al., 2000; Gifford et al., 2007; Jouventin et al., 1994;
Weimerskirch et al., 1993). The use of this technology in understanding the migration of large
whales has, however, lagged behind other taxa largely due to the inability to catch and attach
tags to such large species. It is only relatively recently that research groups developed reliable
tags that can be implanted into free-ranging whales (e.g. Gales et al., 2009; Heide-Jørgensen et
al., 2001b; Mate et al., 2007). Such tags have now been used in studies of many large whales
species such as blue (Heide-Jørgensen et al., 2001b; Mate et al., 2007), humpback (Dalla Rosa et
al., 2008; Gales et al., 2009; Lagerquist et al., 2008), sei (Olsen et al., 2009), right (Baumgartner &
Mate, 2005), bowhead (Mate et al., 2000) and minke whales (Heide-Jørgensen et al., 2001a).
Satellite tracking studies of whales or other cetacea in Australian waters are few and therefore
their movement patterns and migratory routes are generally poorly described. In one of the first
studies Gales et al. (2009) deployed 16 tags on humpback whales migrating south off New South
Wales and followed their migration paths around Tasmania and New Zealand and through to
Antarctic waters. In the same year five tags were attached to five long-finned pilot whales prior
to their release after stranding in north-western Tasmania (Gales, R. et al., in press). Although
these studies represent significant progress in the development of satellite tags for large whales,
the technology and reliability of attachment remains somewhat erratic and poor tag
performance remains common for all research groups in this field (e.g. Gales et al., 2010; Gales et
al., 2009; Garrigue et al., 2010; Mate et al., 2007)
In 2009 the first large-scale (>20) tagging study was conducted on the western Australian
population of humpback whales. This study aimed to describe the migratory distribution and
behaviour of nursing humpback whales on their southbound migration and so a total of 23 tags
were deployed on female humpback whales with young calves in Camden Sound, the Buccaneer
Archipelago, Pender Bay and Beagle Bay (Double et al., 2010). Although there was reasonably
poor tag performance (9 tags provided few or no data) tagged whales were tracked from one to
108 days (mean = 15.3 days; SD = 26.6) for a total of 23,243 km (mean = 1,367 km; min: 5.75;
max: 9,071) over the 1,250 locations received. The total net distance moved from the first to last
location was 16,912 km (mean = 994 km; min: 3.2; max: 7,328). One individual was tracked from
its breeding ground in north-western Western Australia to its feeding ground at the edge of the
sea ice around Antarctica; a migration distance of over 7000 km. Importantly, this data set
provided the first detailed individual movement data for migrating females with calves in the
north-west region of Western Australia.
Satellite tracking of Western Australian humpback whales
Page 4
The aim of the project described in this report was to expand on the 2009 study but focus on
northbound movements by deploying satellite tags on northbound male and female humpback
whales off North West Cape of Western Australia (see Figure 1). Data from these tags will
provide further individual and population level information on the whales’ distribution and
behaviour close to their recognised calving area near Camden Sound in the Kimberley region of
Western Australia. This north-west region of Western Australia is currently subject to expanding
marine-based industries. Information on the behaviour and distribution of this recovering
humpback whale population can underpin decisions by conservation managers and other
stakeholders so that potential impacts related to these marine based industries are managed
appropriately.
Methods
The satellite tags used in this study have a custom-designed, anchor section joined to a housing
manufactured by Wildlife Computers (Redmond, Washington, USA) containing the Spot 5
transmitter (see Gales et al., 2009). The tags were designed to implant up to a maximum of
290mm into the skin, blubber, interfacial layers and outer muscle mass of the whale (generally
just forward and to the left or right side of the dorsal fin). The front 80mm of the tag
disarticulates from back section of the tag post-deployment; a flexible 5mm multi-braided
stainless steel wire maintains a coupling between the two parts. Retention of the tag is
maintained through two actively sprung plates, and a circle of passively deployed ‘petals’. All
external components of the tag were built from stainless steel and the tags were surgically
sterilised prior to deployment.
Each tag was deployed using a compressed air gun (modified ARTS, Restech) set at a pressure of
between 7.5 and 10 bar (cf. Gales et al., 2009; Olsen et al., 2009). When fired from the air gun the
tags were attached to a purpose-designed projectile carrier. Retention teeth on the projectile
carrier grip a metal ring fitted to the end of the tag. When the tag makes contact with a whale,
the rapid deceleration of the tag and carrier withdraws the retention teeth and releases the
projectile carrier (Double et al., 2010). The metal ring used to attach the carrier is designed to
fall off in time to reduce the drag of the tag. Tags were deployed from the bow of RV Whale Song
or the bow-sprit of a 5.8m rigid-hulled inflatable boat at a range of 3-8m (see Table 1).
Once deployed, the first dive of the whale will activate the tag. The tag will then transmit upon
each surfacing if it has not transmitted during the previous 30 seconds. On the day of
deployment the tags were set to transmit until 00:00 hrs UTC; after that time they were set to
transmit on a 6hr on, 18hr off duty cycle until the tag falls off the whale, malfunctions or the two
AA lithium batteries are exhausted. Following each deployment, we used a hand-held receiver to
check that the tag was transmitting. Whenever possible all the whales encountered during the
research period were photographed for photo-identification studies and detailed photographs
were taken of each tag deployment. On deployment we also recorded deployment time,
deployment location, tagging distance, gun pressure, percentage implantation of the tag, pod
size, pod composition, the whale’s reaction to the deployment and the weather and sea
conditions.
During the transmission periods, locations were obtained via the Argos System of polar-orbiting
satellites (Argos, 1990). Each location was allocated a level of accuracy by the Argos System.
Categories 3, 2, 1 have associated error predictions of 250m, 500m and 1500m respectively
Satellite tracking of Western Australian humpback whales
Page 5
whereas categories 0, A, B have no associated error prediction. All Argos locations were filtered
using the Speed-Distance-Angle function in the R package (R Development Core Team, 2007)
‘Argosfilter’ (Freitas et al., 2008) which has been designed specifically for marine mammal
tracking data and is based on the algorithm developed by McConnell et al. (1992). This function
removes locations from the data set based on unrealistic swimming speeds, distances between
successive locations and turning angles. The conservative default setting of a maximum
swimming speed of 7.2 km/h was applied in this study.
Great circle distances between locations were calculated using an equation that assumes a
spherical Earth of radius 6371 km. Analyses based on distances used the general tagging
location -22.56 S, 113.62 E (WGS 1984) as the 0km reference.
The R package TRIP (Sumner, 2006) was used to interpolate locations at one hour intervals.
ArcGIS (ESRI Software) with Hawth's Analysis Tools for ArcGIS (Beyer, 2004) was used to
calculate the total distance traveled and net distance traveled from the first location in addition
to Kernel Density Estimates.
When possible, skin biopsies were collected for genetic analyses (Table 1) using a biopsy dart
fired from a modified .22 Paxarms rifle (Krutzen et al., 2002). Biopsies were usually collected
simultaneously with the deployment of the satellite tag. Biopsies were stored in 70% ethanol
and DNA subsequently extracted using the Tissue DNA purification kit for the Maxwell 16 DNA
extraction robot (Promega Corporation). The sexes of the tagged whale were determined using a
5’ exonuclease assay of the polymorphisms in the sex-linked Zinc Finger genes as described by
Morin et al. (Morin et al., 2005).
Results
Between the 8th and the 23rd July 2011, 28 tags were successfully deployed on migrating
humpback whales (Table 1). Of these, 25 were deployed from the bowsprit of R/V Whale Song
and 3 from the RHIB Orca. Three other tags were lost during this campaign due to unsuccessful
deployment attempts. Biopsy samples were obtained from 18 of the successfully tagged whales
from which genetic analyses revealed 5 were females and 13 were male. In addition, 2 whales
were accompanied by a young calf so there was a minimum of 7 females in the sample of 28
tagged whales.
Figure 1 shows the deployment locations off North West Cape for the 28 tagged whales. Of these
tags, 15 performed poorly providing either no uplinks at all (N=1), only Z class data (i.e. no
location, just an uplink - N=2), or stopped transmitting within the first six days (N=12; see Table
2). The poor data stream from five of these tags can be explained by poor deployments (see
Table 1). It is most likely the remaining tags stopped transmitting due to rapid tag loss or
damage.
Tag performance is summarised in Table 2 and the tracks provided by the tags are presented in
Figures 4 to 7. Whales were tracked from 1 to 47 days (mean = 16.04 days; SD = 15.1). A total of
4,404 locations were received. 3,615 of these locations were not removed by the filtering
process and therefore considered reliable (Table 2, see Methods). Whales that provided more
than 1 location were tracked for a total of 28,893 km (mean = 1,256 km; max: 3,612) and the
total net distance moved from the first to last location was 11,702 km (mean = 509 km; max:
1,283).
Satellite tracking of Western Australian humpback whales
Page 6
Generally the tag longevity in this study was shorter than previous AMMC deployments using a
similar tag design (Figure 2 – Exmouth Gulf, WA 2011 survival curve). However, the survival was
very similar to a campaign on the same population of humpback whales in 2009 (Figure 2 -
Kimberley, WA 2009 survival curve). In both studies the tags suffered a severe early failure rate
as shown by the steep survival curve in the first week after deployment. We suspect that the tag
performance was lower for these campaigns in particular because whales in this region are
highly social or aggressive and located in shallow waters where whales may damage the tag on
the sea floor.
North West Cape to Camden Sound
Despite the high tag attrition, 10 whales were tracked for over 3 weeks (Figures 4 - 6). Of these,
seven whales were clearly tracked to the northern most point on their migration (Figure 7).
Interestingly 2 whales (98107 female without calf; 98127 male) completed their northward
migration south of Karratha and passed the tagging location in a southerly direction between 10
and 20 days post-deployment (Figures 5, 6, 7a). This is the first direct evidence that some
whales terminate their migration well south of the recognised ‘calving ground’ for this
population in the Buccaneer Archipelago/Camden Sound region of Western Australia. Of the five
remaining whales where the northern terminus was clear, surprisingly, only one reached
Camden Sound proper while the remaining four reached the Dampier Peninsula or the
Buccaneer Archipelago (Figures 5, 6, 7b).
The whales tended to be within 50km of the coast in the region between North West Cape and
Camden Sound (approx. 1000 km of coastline - Figure 8). Consequently, the whales were
frequently in very shallow water (most medians <40 m; Figure 9). This pattern is similar to the
data from the south migration although on the north migration, there was no indication that the
whales tended to migrate into deeper waters off the western side of the Dampier Peninsula as
was reported in the study of their southern migration (Double et al., 2010).
Daily linear travel distances from the tagging location were highly variable both within and
between whales but throughout the region the median values fell between 40 and 60 km per day
(Figure 10). There does appear to be a trend towards lower daily travel speeds as the whales
progress north but this has yet to be assessed statistically. Data are few south of North West
Cape but the southward daily travel speed reached over 100km per day.
Although likely to be influenced strongly by deployment location and tag longevity, the relative
Humpback whales tagged during theirnorthbound migration in the month of July 2011.
This research was conducted by the Australian Marine Mammal Centre of the Australian Antarctic Division (DEWHA) in collaboration with the Western Australian Marine Science Institute (WAMSI), Woodside Energy Ltd., and the Centre for Whale Research (WA) Inc.
Woodside Energy Ltd. provided funds to WAMSI to support this project.
PTT Numbers
53360
53398
53431
53733
98101
98107
98110
98111
98112
98113
98118
98120
98121
98123
98124
98126
98127
98131
98133
98136
98140
98142
98144
WESTERN AUSTRALIA
PERTH
DERBY
KALBARRI
GERALDTON
CARNARVON
PORT HEDLAND
Broome
Exmouth
Karratha
Scott Reef
Pender Bay
Perth Canyon
Camden Sound
North West Cape
James Price Point
130°E
130°E
120°E
120°E
110°E
110°E
20°S 20°S
30°S 30°S
Satellite tracking of Western Australian humpback whales
Page 17
Figure 5. The tracks obtained off northern Western Australia from the 25 satellite-tagged whales that provided more than one reliable location.
SATELLITE TRACKING OF WESTERN AUSTRALIAN HUMPBACK WHALES
Humpback whales tagged during theirnorthbound migration in the month of July 2011.
This research was conducted by the Australian Marine Mammal Centre of the Australian Antarctic Division (DEWHA) in collaboration with the Western Australian Marine Science Institute (WAMSI), Woodside Energy Ltd., and the Centre for Whale Research (WA) Inc.
Woodside Energy Ltd. provided funds to WAMSI to support this project.
PTT Numbers
53360
53398
53431
53733
98101
98107
98110
98111
98112
98113
98118
98120
98121
98123
98124
98126
98127
98131
98133
98136
98140
98142
98144
WESTERN AUSTRALIA
DERBY
PORT HEDLAND
Broome
Exmouth
Karratha
Scott Reef
Pender Bay
Camden Sound
North West Cape
James Price Point
130°E
130°E
120°E
120°E
20°S 20°S
Satellite tracking of Western Australian humpback whales
Page 18
Figure 6. The tracks obtained off the Kimberley region of Western Australia from the 25 satellite-tagged whales that provided more than one reliable
location.
SATELLITE TRACKING OF WESTERN AUSTRALIAN HUMPBACK WHALES
Humpback whales tagged during theirnorthbound migration in the month of July 2011.
This research was conducted by the Australian Marine Mammal Centre of the Australian Antarctic Division (DEWHA) in collaboration with the Western Australian Marine Science Institute (WAMSI), Woodside Energy Ltd., and the Centre for Whale Research (WA) Inc.
Woodside Energy Ltd. provided funds to WAMSI to support this project.
PTT Numbers
53360
53398
53431
53733
98101
98107
98110
98111
98112
98113
98118
98120
98121
98123
98124
98126
98127
98131
98133
98136
98140
98142
98144
DERBY
PORT HEDLAND
Broome
Scott Reef
Pender Bay
Camden Sound
James Price Point
120°E
120°E
20°S 20°S
Satellite tracking of Western Australian humpback whales
Page 19
a)
b)
Figure 7. Linear distance from a) the tagging location and b) Camden Sound against the number
of days since the tag was deployed.
0 10 20 30 40
050
01
00
015
00
Days since tagged
Dis
tance
fro
m ta
gg
ed
loca
tion
(km
)
0 10 20 30 40
05
00
10
00
15
00
20
00
Time since tagged (days)
Dis
tan
ce
fro
m C
am
de
n S
ou
nd
(km
)
Satellite tracking of Western Australian humpback whales
Page 20
Figure 8. Distance between the locations from the 25 satellite-tagged whales and the main
coastline of Western Australia. Data are categorized relative to their linear (great circle) distance
from a single general tagging location (-22.56 S, 113.62 E). Indicative geographic locations are
provided above the graph (cf. Figure 8). The box plot displays the median, the 10th, 90th
(whiskers) and the 25th and 75th percentiles (box).
-900 -600 -300 0 300 600 900 1200
05
01
00
15
02
00
Distance from tagging location (km)
Dis
tan
ce
fro
m c
oa
st (k
m)
Exm
ou
th
Po
rt H
ed
lan
d
Pe
nd
er
Ba
y
Ca
md
en
So
un
d
Satellite tracking of Western Australian humpback whales
Page 21
Figure 9. The water depth at the locations provided by the 25 satellite-tagged whales. Data are
categorised relative to their linear (great circle) distance from a single general tagging location
(-22.56 S, 113.62 E). Negative distances are those to the South of the tagging location. The box
plot displays the median, the 10th, 90th (whiskers) and the 25th and 75th percentiles (box).
-600
-450
-300
-150 0
150
300
450
600
750
900
10
50
12
00
13
50
-1200
-1000
-800
-600
-400
-200
0
Distance from tagging location (km)
De
pth
(m
etr
es)
Exm
ou
th
Po
rt H
ed
lan
d
Pe
nd
er
Ba
y
Ca
md
en
So
un
d
Satellite tracking of Western Australian humpback whales
Page 22
Figure 10. Graph showing the linear (great circle) distance traveled in a 24-hour period by the
25 satellite-tagged whales. Data are categorised relative to their linear (great circle) distance
from a single general tagging location (-22.56 S, 113.62 E). Indicative geographic locations are
provided above the graph (cf. Figure 8). The box plot displays the median, the 10th, 90th
(whiskers) and the 25th and 75th percentiles (box) if sufficient data are available.
-600 -300 0 150 450 750 1050 1350
05
01
00
15
0
Linear distance from tagging location (km)
Lin
ea
r dis
tance
tra
ve
lle
d in
24
ho
urs
(km
)
Exm
ou
th
Port
He
dla
nd
Pen
de
r B
ay
Ca
md
en S
oun
d
Satellite tracking of Western Australian humpback whales
Page 23
Figure 11. Cumulative occupancy times of the 25 satellite-tagged whales determined by interpolating the whale tracks to daily points. The density is
determined using a Kernel Density Smoother with 2 degree sampling radius. Note that although the unit is in cumulative days this is a relative rather
than absolute measure of occupancy. Due to the generally short longevity of the tags these data are biased by the deployment location.