Movements and dive behaviour of a toothﬁsh-depredating
killer and sperm whale
Jared R. Towers
, Paul Tixier
, Katherine A. Ross
, John Bennett
, John P. Y. Arnould
Robert L. Pitman
, and John W. Durban
Bay Cetology, Box 554, Alert Bay, BC V0N 1A0, Canada
School of Life and Environmental Sciences, Deakin University, Burwood, VIC 3125, Australia
Falklands Conservation, Stanley, Falkland Islands
Sanford Seafood, Hall Street, North Mole, Timaru 7910, New Zealand
Southwest Fisheries Science Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration, 8901 La Jolla Shores
Drive, La Jolla, CA 92037, USA
*Corresponding author: tel: þ1 250 902 1779; e-mail: firstname.lastname@example.org.
Towers, J. R., Tixier, P., Ross, K. A., Bennett, J., Arnould, J. P. Y., Pitman, R. L., and Durban, J. W. Movements and dive behaviour of a
toothﬁsh-depredating killer and sperm whale. – ICES Journal of Marine Science, doi:10.1093/icesjms/fsy118.
Received 15 May 2018; revised 24 July 2018; accepted 8 August 2018.
Depredation of demersal longlines by killer and sperm whales is a widespread behaviour that impacts ﬁsheries and whale populations. To bet-
ter understand how depredating whales behave in response to ﬁshing activity, we deployed satellite-linked location and dive-proﬁle tags on a
sperm and killer whale that were depredating Patagonian toothﬁsh from commercial longlines off South Georgia. The sperm and killer whale
followed one ﬁshing vessel for >180 km and >300 km and repeatedly depredated when longlines were being retrieved over periods of 6 and
7 d, respectively. Their behaviours were also sometimes correlated with the depths and locations of deployed gear. They both dove signiﬁ-
cantly deeper and faster when depredating compared with when foraging naturally. The killer whale dove >750 m on ﬁve occasions while
depredating (maximum: 1087 m), but these deep dives were always followed by long periods (3.9–4.6 h) of shallow (<100 m) diving. We hy-
pothesize that energetically and physiologically costly dive behaviour while depredating is driven by intra- and inter-speciﬁc competition due
to the limited availability of this abundant resource.
Keywords: competition, depredation, diving, foraging, killer whales, movements, Patagonian toothﬁsh, South Georgia, sperm whales
Killer (Orcinus orca) and male sperm whales (Physeter macroce-
phalus) are among the top predators in high latitude food webs.
They normally occupy different ecosystem niches, but in some
regions both will take advantage of opportunities to remove ﬁsh
from commercial longlines (Kock et al., 2006). This behaviour,
referred to as depredation, is a deviation from natural foraging
behaviour (Gilman et al., 2006) and reﬂects the behavioural plas-
ticity and adaptive capabilities of each species. However,
physically harmful interactions with ﬁshing gear and ﬁshers can
negatively impact the health of cetaceans that engage in this be-
haviour. Furthermore, depredation can reduce the accuracy of
stock assessments and have a major ﬁnancial impact on ﬁsheries
economies (Gilman et al., 2006;Read, 2008;Hamer et al., 2012;
Peterson et al., 2014). The severity of these impacts have been in-
creasing worldwide since depredation of commercial longlines
was ﬁrst reported in the 1960s (Hamer et al., 2012). This has
Former afﬁliation: Government of South Georgia and South Sandwich Islands, Government House, Stanley, Falkland Islands.
CInternational Council for the Exploration of the Sea 2018. All rights reserved.
For permissions, please email: email@example.com
ICES Journal of Marine Science (2018), doi:10.1093/icesjms/fsy118
Downloaded from https://academic.oup.com/icesjms/advance-article-abstract/doi/10.1093/icesjms/fsy118/5103434 by firstname.lastname@example.org on 28 September 2018
resulted in an increased need to understand and mitigate this
A number of operational and technological mitigation techni-
ques have been used to minimize the impacts of depredation by
killer and sperm whales (Gilman et al., 2006;Tixier et al., 2010;
Goetz et al., 2011;Thode et al., 2012;Straley et al., 2014;
O’Connell et al., 2015;Tixier et al., 2015a;Towers, 2015;Werner
et al., 2015;Janc et al., 2018). Some, such as modifying gear
(Moreno et al., 2008) or ceasing gear retrieval, buoying off the
line, and leaving the area to return several h after whales have left
(Tixier et al., 2015b) have proved to be successful at times.
However, none has come without an economic cost or been
completely effective at eliminating depredation. Effective mitiga-
tion is complicated further because killer and sperm whales often
depredate repeatedly and concurrently (Purves et al., 2004;Roche
et al., 2007;Tixier et al., 2010;Straley et al., 2015;Tixier et al.,
2016), spread knowledge of this behaviour to other whales via so-
cial transmission (Tixier, 2012;Fearnbach et al., 2014;Schakner
et al., 2014), appear to have unique depredation strategies
(Hucke-Gaete et al., 2004;Tixier et al., 2015b) and may compete
for opportunities to depredate (Nolan et al., 2000).
The limited efﬁcacy of depredation mitigation strategies and
the escalating impacts of this behaviour indicate that a better un-
derstanding of killer and sperm whale depredation is necessary to
reduce and discourage this behaviour. Studies using photo-
identiﬁcation (Tixier et al., 2010;Straley et al., 2015), hydro-
phones (Mathias et al., 2009,2012;Thode et al., 2015), and
underwater video (Mathias et al., 2009;Guinet et al., 2015) have
all provided insight, but the ﬁne-scale horizontal and vertical
movements of depredating whales have rarely been investigated.
In fact, only two studies have used telemetry to assess and
compare the movements and dive patterns of depredating whales
to their natural behaviour. Straley et al. (2014) found that some
tagged depredating male sperm whales followed a ﬁshing vessel
for several hundred km while others engaged in natural migratory
movements, and Mathias et al. (2012) discovered that some
tagged male sperm whales repeatedly dove under 200 m
while depredating, but that most dove between 400 and 700 m
both while depredating and naturally foraging. Male sperm
whales typically descend to depths up to 1900 m for as long as
60 min when foraging (Watkins et al., 2002;Teloni et al., 2008;
Fais et al., 2015;Guerra et al., 2017), but the species may be
capable of diving much deeper and longer (Clarke, 1976;Watkins
et al., 1985). Although no depredating killer whales have been
tagged, this species typically dives <300 m for under 4 min (Baird
et al., 2005;Miller et al., 2010;Wright et al., 2017), but can
descend to at least 767 m and remain submerged for nearly
16 min (Reisinger et al., 2015).
The dive capacity of both species, but especially of sperm
whales, indicates that they can access the depths at which some
longlines are set (700–2250 m; Government of South Georgia &
South Sandwich Islands, 2017). However, depredation by killer
and sperm whales has only been observed to take place during
gear retrieval (Dahlheim, 1988;Sigler et al., 2008;Goetz et al.,
2011;Gasco et al., 2015), and the depth range at which this be-
haviour occurs remains largely unknown. Under natural circum-
stances, both species dive to depths that correspond to where
their prey are found or chased to (Fais et al., 2015;Wright et al.,
2017). For sperm whales, these prey include several species of
cephalopods (Clarke, 1980;Whitehead, 2009) and large teleost
ﬁshes (Gaskin and Cawthorn, 1967;Martin and Clarke, 1986)
that they catch in epipelagic, mesopelagic, bathypelagic and ben-
thic zones (Teloni et al., 2008;Guerra et al., 2017). Killer whales
also prey on several cephalopod and high lipid content ﬁsh spe-
cies that they catch in different bathymetric zones (Guinet et al.,
2007;Hanson and Walker, 2014;Wright et al., 2017), but some
distinct killer whale ecotypes specialize on these, and/or other
prey, such as mammals, birds, and reptiles (Ford et al. 1998;
Pitman and Ensor, 2003;Ford, 2009;Morin et al., 2010;
Foote et al., 2016;Durban et al., 2017). Patagonian toothﬁsh
(Dissostichus eleginoides) are a large bottom-dwelling teleost ﬁsh
with high lipid contents that typically occur at depths ranging
from 500 to 2500 m throughout the sub-Antarctic (Collins et al.,
2010). Longlining for this species can therefore create an abun-
dant and easily accessed source of preferred prey for some popu-
lations of killer and sperm whales.
In the South Atlantic Ocean around the island of South
Georgia, six commercial longlining vessels remove up to 2200 t of
toothﬁsh from shelf edge waters each year in a sustainably man-
aged ﬁshery (Government of South Georgia & South Sandwich
Islands, 2017). Killer whales and male sperm whales have been
depredating from this ﬁshery since the 1990s (Ashford et al.,
1996;Kock et al., 2006) and impact 3–5% and 13–40% of lines re-
trieved each year, respectively (Purves et al., 2004;Clark and
Agnew, 2010;So¨ffker et al., 2015). It is estimated that these spe-
cies are responsible for reducing the total toothﬁsh catch at South
Georgia by up to 8% in some years (Clark and Agnew, 2010).
Although, the extent to which sperm and killer whales in this re-
gion depredate as compared with feeding on naturally obtained
prey is not known, depredation rates are generally reported to be
increasing at South Georgia (Towers, 2015).
To better understand the horizontal and vertical movements of
depredating killer and sperm whales and how depredation differs
from natural foraging behaviours, daily observations of depreda-
tion were recorded from a toothﬁsh longliner at South Georgia
and depth-recording satellite transmitter tags were applied to one
individual of each species while they were depredating in the area.
Here, we use data obtained from the tags to determine how often
the whales depredated when opportunities to do so were avail-
able, compare how their dive behaviours differed between depre-
dating and natural foraging, test whether they depredated
longlines that were not being retrieved, and describe their natural
foraging behaviour. The results provide new insights into the nat-
ural and depredatory foraging ecology of killer and sperm whales
that can be used to help develop depredation mitigation strategies
and improve ﬁsheries management practices.
Material and methods
Field effort around South Georgia was undertaken from the
Patagonian toothﬁsh longliner San Aspiring in May and June
2015. Longline sets made by this vessel consisted of a main line
ranging in length from 4 to 12 km. Each line was equipped with
1000 baited hooks km
that were tethered to the main line by
snoods (short lines) <1 m in length. Lines were secured on the
seaﬂoor by anchors positioned at each end, at depths ranging
from 700 to 1700 m. Each anchor was attached to a vertical
downline equipped with a buoy at the surface. Longlines were al-
ways set at night and were left deployed on the seaﬂoor for peri-
ods of 5–44 h. They were typically retrieved during the day at a
rate of 2 km of line h
2J. R. Towers et al.
Downloaded from https://academic.oup.com/icesjms/advance-article-abstract/doi/10.1093/icesjms/fsy118/5103434 by email@example.com on 28 September 2018
Observations of depredating whales were conducted with na-
ked eye and monopod mounted Swarovski 8X42 binoculars from
a height of 7 m above sea level from the bridge of San Aspiring
during 39 d between 7 May and 15 June, 2015. The number and
species of depredating whales and the number of intact and par-
tially eaten toothﬁsh recovered on the line were recorded.
Individual depredating whales were photo-identiﬁed at every op-
portunity with Nikon D800 and D300 SLR cameras outﬁtted with
a 300 mm F2.8 lens.
SPLASH10 transmitter tags (see Schorr et al., 2014) developed
by Wildlife Computers (Wildlife Computers, Redmond, WA,
USA) were used to collect location and dive data from depredat-
ing killer and sperm whales. These 54 g tags were outﬁtted with
two titanium posts, each with six barbs designed to penetrate and
anchor up to 6.5 cm into the dorsal ﬁn or ridge of the whale.
They were also outﬁtted with a wet/dry sensor to activate trans-
missions through the Argos satellite system via an antenna when
the whale surfaced. Tags were programmed to provide up to
600 locations d
between 00:00 and 04:00, 07:00 and 12:00, 15:00
and 20:00, and 23:00 and 00:00 UTC every d until 20 June and at
5 d intervals thereafter. Tags were also pre-programmed to only
record dives deeper than 15 m and longer than 30 s. Surface time
therefore included all activities that occurred between the end
and beginning of these dives. Tags were mounted on the end of a
crossbow bolt and deployed from a 150 lb draw Excalibur Vixen
crossbow onto the dorsal ﬁn of a killer whale and the dorsal
hump of a sperm whale (e.g. Reisinger et al., 2014). After contact,
the ﬂoating bolt bounced off the whale, leaving the tag attached.
Deployment effort was undertaken from the 4 m tender of the
San Aspiring at ranges <10 m whenever whale activity, weather,
and ﬁshing activity were conducive.
Tag data analyses were conducted in the R Statistical
Environment 3.4.1 (R Core Team, 2017). Location data calculated
with the Argos Kalman ﬁlter (Lopez and Malarde´, 2011) were
then further ﬁltered using the Speed-Distance-Angle (SDA) algo-
rithm and the SDAﬁlter function in R package “Argosﬁlter”
(Freitas et al., 2008). Maximum swim speeds used as a threshold
in this algorithm were 2.5 m s
for sperm whales (Whitehead,
2003) and 7 m s
for killer whales (Durban and Pitman, 2012).
Filtered tag locations were used to assign an estimated location to
each dive record through a linear interpolation (function interp1
in the “pracma” package). This interpolation was performed us-
ing the dates and times of dive records as the dates and times at
which locations were estimated.
Data collected on the ﬁshing operations of the San Aspiring in-
cluded the GPS coordinates and depths of both ends of each long-
line at setting, as well as the date and time of the beginning of
setting and the end of retrieval. Since longlines were set in a
straight line over the seaﬂoor, we calculated the coordinates of
the middle point (the mean position of the two ends) to provide
a single location per longline. The distance between the tagged
animals and the nearest longline set was calculated as the least
distance between the location of each dive record and the three
locations (two ends and middle point) of all longline sets.
Dive records were assigned to one of the following behaviour
states: depredating, natural foraging, non-foraging or uncertain.
Taking into account that tag settings prevented dives <15 m from
being recorded, dives were ﬁrst classiﬁed as non-foraging dives if
shallower than 25 m for the sperm whale (based on threshold esti-
mated by Fais et al., 2015) and 20 m for the killer whale (based on
threshold estimated by Wright et al., 2017). Deeper dives were
then classiﬁed as depredating if they completely or partially oc-
curred during gear retrieval when the tagged animal was photo-
identiﬁed near the San Aspiring and there was evidence that
toothﬁsh were being removed from the longline (catch was lower
than expected and/or some hooks were recovered with only par-
tially intact toothﬁsh on them). The gear retrieval process was de-
ﬁned for depredating dives as the period of time between the ﬁrst
and last hook reaching the surface. Foraging dives were non-
depredating dives that met two conditions. First, these dives were
conﬁrmed as occurring within a 50 km range from the nearest
ﬁshing gear deployed by the San Aspiring. This threshold was
chosen because although data on operations of other toothﬁsh
longliners were not available for this study, information on their
positions received regularly by the San Aspiring indicated that
none was within 50 km during the times that data were transmit-
ted from the tagged whales. This range therefore allowed us to ex-
clude dives that could have been made by the whales while
interacting with other vessels. Secondly, foraging dives were iden-
tiﬁed if they occurred in between phases of gear retrieval and dur-
ing phases of gear retrieval for which there was no visual evidence
that tagged whales were depredating. Dives were classiﬁed as un-
certain if they occurred: (i) during gear retrieval phases at times
that photo-identiﬁcation and visual effort could not be con-
ducted due to darkness or snow, (ii) during times that the loca-
tion and depth of the nearest longline set had been modiﬁed by
buoying off the line, or (iii) when the tagged whales were >50 km
from the nearest gear set by San Aspiring.
Depredating dives were ﬁrst compared with the depth at which
the longlines were set on the seaﬂoor. For each depredated set,
the correlation between the maximum recorded dive depth of
tagged whales and the depth of the longline set was tested using
standard least-squares regressions. As two depth records both
recorded by the sounder of the vessel were available per set (one
for each end), three tests were separately conducted using the
depth of each end and their mean depth. The same tests were
then performed on dives that were classiﬁed as foraging dives
when they occurred near a longline set deployed on the seaﬂoor.
Foraging dives were only selected for this analysis if they occurred
within the same range to deployed gear as gear that was being
depredated from while being retrieved. The depths of all remain-
ing foraging dives (not in proximity to gear) were examined in re-
lation to the local bathymetry. The bathymetry was retrieved
from the ETOPO1 database at a 1 min resolution using the mar-
map package in R (Pante and Simon-Bouhet, 2013), and assigned
to individual dive records based on the nearest interpolated loca-
tion at the start of the dive.
Differences in depths, durations and vertical velocities between
available depredating and foraging dives were statistically tested
using Generalized Least Squares models (GLS) using the function
gls in R package “nlme”. Data on dive depth and duration are
provided as maximum values from tags. For each dive, the verti-
cal velocity was calculated as twice the depth divided by the total
duration of the dive, and expressed in m s
. Velocities presented
do not account for any non-vertical movements and are therefore
estimated values. A square root transformation was applied to
dive depths and durations, and a log transformation was applied
to vertical velocities to meet the normality assumptions. GLS
models included an autoregressive (AR1) correlation structure to
Movements of depredating whales 3
Downloaded from https://academic.oup.com/icesjms/advance-article-abstract/doi/10.1093/icesjms/fsy118/5103434 by firstname.lastname@example.org on 28 September 2018
account for temporal autocorrelation between successive dives
(Zuur et al., 2009).
Tags were deployed on one adult male sperm whale and one adult
female killer whale near Shag Rocks to the west of South Georgia
on 28 May 2015 and 2 June 2015, respectively (Figure 1a and b).
The sperm whale tag provided 260 Argos locations for 17 d and
23 h, with an average of 14.4 60.7 SE locations d
. The killer
whale tag transmitted 348 locations for 14 d and 16 h, with an av-
erage of 23.2 61.3 SE locations d
. The SDA ﬁltering resulted in
the removal of 37 (14%) and 20 (5.7%) Argos records for the
sperm and killer whale, respectively.
The sperm and killer whale spent 156 and 79 h within 50km of
the nearest longline set by the San Aspiring, respectively. Over
these time periods, 90 SDA ﬁltered Argos locations were recorded
from the sperm whale and 87 were recorded from the killer whale.
The sperm whale remained within 25 km of the nearest set gear
for 5 consecutive d after being tagged (Figure 2a). After the killer
whale was tagged on 2 June the San Aspiring moved 75 km away
to begin ﬁshing further to the east in an effort to avoid depreda-
tion, but both tagged whales and the 20 associated pod members
of the killer whale (see Towers, 2015) followed. The swim speed
of the sperm whale increased to 1.9 m s
and it and the killer
whale came within 20 km and 30 km, respectively, of gear re-
trieved by San Aspiring on 3 June before the vessel moved 200 km
further to the east to ﬁnd a more productive ﬁshing area
(Figure 2a and b). Only the tagged killer whale and associated
pod members continued to follow and on 4 and 5 June were
documented near San Aspiring (Figure 2b). On 5 June, a set being
depredated was buoyed off and then the San Aspiring left the
area. The killer whale remained near the set gear for the next 5 h
and then moved off. On 7 June, the swim speed of the killer whale
increased to 4.2 m s
and then slowed as it and associated pod
members located San Aspiring and then spent 8 June in its vicin-
ity (Figure 2b). The last set the tagged whale depredated from this
day was buoyed off while the vessel transited away and then
returned the next morning to retrieve it. After retrieving the set,
the San Aspiring traveled 385 km west to avoid depredation by
ﬁshing in a different area before returning to port.
Overall, the tagged sperm whale followed the San Aspiring over
a distance of 182 km and the tagged killer whale and associated
pod members interacted with the vessel over a range of 302 km.
After the last time the tagged killer whale was veriﬁed in the vicin-
ity of San Aspiring on 9 June, it traveled west along the shelf edge,
and then from 13 to 16 June swam directly north away from the
ﬁshing grounds at 2.6 61.9 SD ms
(n¼74 locations) before
the tag stopped transmitting (Figure 1b). By comparison, after 3
June the sperm whale travelled back to Shag Rocks and then re-
versed course and travelled east along the shelf edge to the north
side of South Georgia, where tag transmissions ceased on 14 June
Dive types and totals
For the sperm whale, dive and surface data were available for
88% of the deployment time (Figure 1a). Of the available data,
24% were surface time and 76% were dives >15 m and longer
than 30 s. Information on a total of 611 dives was recorded, in-
cluding 239 dives performed within a 50 km range of the nearest
longline set (Figure 2a). Among the 239 dives, four (2%) were
considered non-foraging dives based on the 25 m depth thresh-
old. The tagged sperm whale was visually conﬁrmed depredating
from eight sets resulting in 87 dives being categorized as depre-
dating dives. Among the remaining dives, 65 were classiﬁed as
uncertain because they occurred near a set that was being re-
trieved at night, and 83 were classiﬁed as natural foraging dives
because they occurred when no gear was being retrieved
Figure 1. Full ﬁltered and interpolated tracks of (a) the sperm whale from 28 May to 14 June 2015 and (b) the killer whale from 2 to 16 June
2015 with recorded dive data (black) and missing dive data (white). The locations of the longline sets retrieved by the San Aspiring that were
depredated by the whales while tagged are depicted with white squares (n¼8 sets for the sperm whale, n¼3 sets for the killer whale).
4J. R. Towers et al.
For the killer whale, dive and surface data were recorded dur-
ing 86% of the tag transmission period (Figure 1b). Of the avail-
able data, 71% were surface time and 29% were dives >15 m and
longer than 30 s. A total of 489 dives (33% of all dives recorded)
were performed within 50 km of the nearest longline set
(Figure 2b). Among them, 133 (28%) were considered non-
foraging dives based on the 20 m depth threshold. The tagged
killer whale was visually conﬁrmed depredating during the
retrieval of three sets (Figure 2b), resulting in 37 dives being clas-
siﬁed as depredating dives. Among the remaining dives, 270 were
classiﬁed as natural foraging and 49 as uncertain.
While tagged, the sperm whale was visually conﬁrmed depre-
dating simultaneously with killer whales during only one set. On
this set, the mean dive depth of the tagged sperm whale was
1122 6327 SD m(n¼7 dives), including some of the maximum
depths recorded (1407 and 1439 m). While tagged, the killer
Figure 2. Swim speed, distance to nearest longline set and dive proﬁles of (a) the sperm whale from 28 May to 4 June 2015 and (b) the
killer whale from 2 to 9 June 2015. The swim speed was calculated from successive ﬁltered location data and is depicted as a smooth curve
using a “loess” method. The distance to nearest gear is presented as the least distance between the location of each dive record and the
three locations (two ends and middle point) of all longline sets. Whether the nearest set was deployed on the seaﬂoor or being retrieved is
depicted. Dive types were classiﬁed as depredating (red), foraging (blue), non-foraging (black), and uncertain (grey) depending on the
distance of the animal to the nearest set, depth thresholds and behavioural observations as described in the Material and methods.
Movements of depredating whales 5
whale was not observed depredating sets also depredated by
Dive depths compared with depth of set longlines
When conﬁrmed depredating, the maximum dive depth of the
sperm whale was not signiﬁcantly correlated to the depth at
which the retrieved longline was set, whether the shallowest
end, the deepest end, or the mean depth of sets were tested
¼0.24, p¼0.12; F
¼0.28, p¼0.17 respectively, n¼8 sets)
(Figure 3a). With only three sets depredated during retrieval, this
correlation could not be tested for the killer whale (Figure 3a).
During the conﬁrmed depredation events, the maximum dis-
tances of the sperm and killer whale to the depredated longline
set were 7.4 km (mean 2.6 60.1 SE, n¼87 locations) and 4.1 km
(mean 2.0 60.2 SE, n¼37 locations), respectively. The correla-
tion between dive depths of the two whales and the depths at
which longline sets were deployed was therefore examined using
foraging dives that occurred within 7.4 km of the nearest set. The
tagged sperm and killer whale were both recorded within 7.4 km
of eight sets deployed on the seaﬂoor. During these phases, the
maximum depths of foraging dives of the sperm whale were posi-
tively correlated to the shallowest of the two ends of set longlines
¼0.74, p¼0.003, n¼8 sets), to the mean depth
at which longlines were set (F
¼0.51, p¼0.027, n¼8
sets), but not to the depth of the deepest end of the longline
¼0.11, p¼0.225, n¼8 sets) (Figure 3b). No cor-
relations were found for the killer whale when using the shallow
¼0.16, p¼0.330) or deep (F
p¼0.361, n¼8 sets) or mean depths of sets (F
¼0.16, p¼0.322, n¼8 sets) (Figure 3b). However, while in
the vicinity of a line that was buoyed off on 5 June the tagged
killer whale made two dives >550 m within 1 h and then moved
away from the gear (Figure 2b). These dives were classiﬁed as un-
certain, however, when this line was retrieved again the next
morning there were no toothﬁsh on the ﬁrst 500 m or so of the
line, but several further along.
The correlation between dive depth and bathymetry was also
tested for foraging dives that occurred when the animals were be-
tween 7.4 and 50 km from the nearest longline set. There was no
correlation for the sperm whale (F
n¼24 dives) or the killer whale (F
p¼0.052, n¼90 dives).
Comparisons of depredating and foraging dives
The sperm whale dove signiﬁcantly deeper, for longer durations and
at greater vertical velocities when depredating during the retrieval of
gear, as compared with when foraging naturally (GLS t¼3.697,
p<0.001 for dive depth; t¼2.029, p¼0.04 for dive duration;
t¼4.622, p<0.001 for vertical velocity). Depredating dive depths
averaged 590 6398 SD m and the maximum dive depth during dep-
redation (1471 m) was 128 m deeper than the maximum recorded
depth of a foraging dive (1343 m) (Table 1;Figure 4a and b). The
maximum duration of a depre dating dive (55.4 min) was 13.3 min
longer than the maximum duration of a foraging dive (42.1 min).
The killer whale also dove signiﬁcantly deeper and at greater
vertical velocities when depredating than when foraging (GLS
t¼4.322, p¼0.002 for dive depth; t¼3.385, p<0.001 for verti-
cal velocity), but no difference was detected for dive duration
(Table 1;Figure 4c and d). The foraging dives showed a bimodal
distribution with 76% (n¼206) of the depths <100 m, and 17%
(n¼47) of the depths >200 m (Figure 4c). A total of 41 foraging
dives >200 m were performed successively on three occasions (2,
5, 8 June) during foraging bouts 2.2–3.7 h. During these continu-
ous natural foraging events, the between-dive variance in maxi-
mum depths was low, respectively averaging 17.5 m (6% of the
mean 292 66SEm,n¼11 dives), 39.6 m (15% of the mean
263 621 SE m, n¼13) and 57.5 m (21% of the mean 272 618
SE m, n¼17) for each of the three bouts. The depredating dives
of the killer whale also showed a bimodal distribution, but
Figure 3. Correlations between the maximum dive depths of the sperm whale (black) and killer whale (grey) and the depths of the nearest
longline set during (a) depredating dives, and (b) foraging dives when the animals were <7.4 km from the nearest set deployed on the
values from linear regression lines (dashed lines) are depicted.
6J. R. Towers et al.
consisted of one or two deep dives >750 m (n¼5 for the three
depredated sets representing 13% of all depredating dives) fol-
lowed by repeated shallow dives (n¼31 dives <100 m represent-
ing 84% of all depredating dives) over periods of 3.9–4.6 h
(Figure 2b). The whale dove to a maximum depth of 1087 m
when depredating set 128 on 8 June 2015 (Figures 2b and 5). The
duration of this dive was 11.4 min with a subsequent estimated
vertical velocity of 3.2 m s
. When combined with the other dep-
redating dives >750 m (n¼5), the average duration was 10.1 6
1.4 SD min and the average vertical velocity was 3.1 6
0.3 SD ms
. Contrastingly, depredating dives <100 m (n¼31)
were performed at a mean vertical velocity of 0.5 60.4 SD ms
Figure 4. Frequency histograms of the dive depths (left) and mean vertical velocities (right) of (a, b) the sperm whale, and (c, d) the killer
whale, for depredating dives (black, n¼87 dives for the sperm whale, 37 dives for the killer whale) and foraging dives (grey, n¼83 dives for
the sperm whale, 270 for the killer whale). Error bars are the Standard Deviation of the mean vertical velocity in m s
Table 1. Sample size and descriptive statistics (mean, standard deviation and maximum) of depth, duration and estimated vertical velocity of
depredating and foraging dives performed by the tagged sperm and killer whale.
Dive depth (m) Dive duration (min) Vertical velocity (m s
Mean 6SD Maximum Mean 6SD Maximum Mean 6SD Maximum
Sperm whale Foraging 83 345 6324 1 343 21.7 610.0 42.1 0.5 60.3 1.3
Depredating 87 590 6398 1 471 25.7 69.0 55.4 0.7 60.3 1.5
Killer whale Foraging 270 87 6100 451 3.7 61.6 7.8 0.7 60.6 3.0
Depredating 37 163 6316 1 087 3.6 62.8 11.5 0.9 60.9 3.5
Movements of depredating whales 7
For two of the depredated sets, the killer whale was within a
10 km range of the set before retrieval started. In both cases, the
deep dives coincided with the end of retrieval of the downline,
when the ﬁrst hooks reached the surface (sets 121 and 129—
Figure 5). These sets were eventually buoyed off to deter further
depredation by the killer whales. When retrieval of set 128 began,
the tagged killer whale was estimated to be >30 km distant. It
started to move towards the set when the ﬁrst hooks reached the
surface, as indicated by an increase of the swim speed and a de-
crease in the distance to the set (Figure 5). As retrieval of most of
this set was complete by the time the killer whales arrived, the re-
mainder of the line was retrieved while they were depredating.
The horizontal and vertical movements of depredating killer and
sperm whales were correlated with ﬁshing activity, indicating that
both species are extremely motivated by opportunities to depre-
date Patagonian toothﬁsh from demersal longlines at South
Georgia. The tagged killer and sperm whale undertook direct
movements to relocate the ship after it moved away to ﬁsh in a
new area and each species dove deeper, faster and longer while
depredating. Even when depredation could not be conﬁrmed, the
movements and dive behaviour of the sperm whale were often
correlated with the locations and depths of set longlines, respec-
tively. The killer whale travelled along the shelf edge while not
depredating, but some of its horizontal and vertical movements
were correlated with the location of gear that was buoyed off,
while other behaviours were indicative of natural foraging.
Horizontal movements of tagged whales
Associated with ﬁshing gear
Tag data, supported by identiﬁcation photos acquired from San
Aspiring (see Towers, 2015), indicate that the horizontal move-
ments of both species were often directly correlated with ﬁshing
activity. The sperm whale mostly depredated in the same general
area over several consecutive d but in total, followed San Aspiring
for >180 km, while the killer whale travelled >300 km in <50 h
Figure 5. Detailed proﬁles of the killer whale behaviour when conﬁrmed depredating three longline sets (sets 121, 128, and 129). The
distance of the whale to the depredated set, the swim speed, and the dive depth are plotted against the time since retrieval of these sets
began. The retrieval process is depicted by the period in which the downline was being hauled (grey) and the period after which the ﬁrst
hooks reached the surface (black).
8J. R. Towers et al.
to depredate even though two out of three depredated sets were
buoyed off soon after it arrived. This exploitation of even small
windows of opportunity to depredate toothﬁsh suggests that this
prey provides energetic beneﬁts outweighing the cost of some
long-distance travel. It also indicates that depredation is favoured
over natural foraging because toothﬁsh may not be easily accessi-
ble to killer whales under natural circumstances due to the great
depths and benthic habitat in which they live (Collins et al.,
Not associated with ﬁshing gear
Both tagged whales eventually disassociated from San Aspiring,
but while the sperm whale travelled along the shelf edge, the killer
whale headed north away from the ﬁshing grounds. Given the
previous behaviour of the sperm whale and the fact that toothﬁsh
longliners operate in the areas it went to (Purves et al., 2004), it is
possible that these movements were motivated by depredation
opportunities. In contrast, given the speed at which the killer
whale travelled and that waters around South Georgia lie within
the polar front (Moore et al., 1999), we believe its ﬁnal north-
bound movements were the beginning of a physiological mainte-
nance migration to warmer waters, as described by Durban and
Pitman (2012) and reported by Reisinger et al. (2015).
Dive behaviour of tagged whales
While depredating during gear retrieval
The two tagged individuals modiﬁed their diving behaviour to
depredate. Both species dove signiﬁcantly deeper and faster when
depredating than when naturally foraging. The sperm whale also
dove signiﬁcantly longer when depredating. However, its dive
depths and durations while depredating were similar to natural
dive behaviour of other male sperm whales tagged at high lati-
tudes (Teloni et al., 2008;Fais et al., 2015;Guerra et al., 2017).
This indicates that dive behaviour required for successful depre-
dation are well within the physiological limits of this species.
However, we believe that the motivation for this whale to access
prey more quickly and at greater depths while depredating was
driven by intra- and possibly, inter-speciﬁc competition. For ex-
ample, between 3 and 13 other sperm whales were present on all
sets that the sperm whale depredated from during the time it was
tagged (Towers, 2015) and some of the deepest dives this whale
made were performed when killer whales were also depredating.
Conﬁrming competition between cetaceans is difﬁcult, given that
their prey are often heterogeneous, widespread, and highly mo-
bile. Toothﬁsh provided by longliners on the other hand, are not
only energy-dense (Collins et al., 2010) and highly desired, but
their availability is temporally limited, static, and localized. The
characteristics of this resource may set the stage for inter-speciﬁc
interference competition and help explain why killer and sperm
whales have been observed acting aggressively towards each other
while depredating (Nolan et al., 2000;Hucke-Gaete et al., 2004).
Compared with the sperm whale, the dives of the killer whale
showed substantially greater variation in maximum depth be-
tween the naturally foraging and depredating states. In particular,
the deep dives made by the killer whale while depredating were
over 300 and 700 m deeper than any dives previously reported for
this species in the southern (Durban and Pitman, 2013;Reisinger
et al., 2015) and northern hemispheres (Baird et al., 2005;Miller
et al., 2010;Wright et al., 2017), respectively. As related killer
whales are known to share prey as an inclusive ﬁtness beneﬁt
(Wright et al., 2016), and no sperm whales were documented
depredating at the same time as the tagged killer whale, its deep
dives were likely not due to resource competition with other
whales. Instead, the fact that the killer whale only made deep di-
ves when it and associated pod members ﬁrst arrived in the vicin-
ity of a line being retrieved suggests that it has learned that this
may be the only opportunity to depredate, because longlines in
this ﬁshery are often buoyed off as soon as killer whales are ob-
served during gear retrieval (Clark and Agnew, 2010). However,
the relative infrequency of these deep dives, even in situations
when the retrieval of gear continued, suggests not only that deep
diving behaviour is energetically costly for killer whales and is
only conducted when a positive net gain is likely, but also, that
these dives may represent the physiological limits of this species.
The durations of two deep depredating dives >750 m (11.4
and 11.5 min) slightly exceeded the aerobic dive limit (cADL) for
adult female killer whales (10.2 min) calculated by Miller et al.
(2010) [This value was calculated from mean mass estimates of
captive adult female killer whales and considering there is much
variation in the size of adult females from different wild popula-
tions (Pitman et al., 2007;Ford, 2014;Durban et al., 2017),
this limit should be considered approximate.]. Tagged adult and
juvenile killer whales sometimes exceed cADL during natural
diving behaviour (Miller et al., 2010;Reisinger et al., 2015).
Furthermore, most small beaked whale species regularly exceed
cADL to pursue and obtain prey at depth (Tyack et al., 2006;
Joyce et al., 2017). For the tagged killer whale, acquisition of prey
from a previously undepredated set is highly likely, so signiﬁ-
cantly exceeding cADL may not be necessary despite the depths
to which deep depredating dives were occurring. However, as
time at depth was likely necessary to ﬁnd and remove prey from
the longline, this indicates that vertical commutes during deep
depredating dives were conducted at relatively high velocities.
Killer whales have been known to chase ﬁsh at speeds up to
6.7 m s
in the North Paciﬁc (Wright et al., 2017), but the me-
dian vertical descent and ascent velocities of these killer whales
while on foraging dives (0.7 and 0.6 m s
respectively) and their
mean velocity while chasing ﬁsh (2.7 m s
) are lower than the
mean vertical velocity of the tagged killer whale during all depre-
dating dives (0.9 60.9 m s
) and it’s estimated average vertical
velocity during only deep depredating dives (3.1 60.3 SD ms
The tagged killer whale made dives <100 m for long periods of
time (232–277 min) after making one or two consecutive deep di-
ves. It is hypothesized that other marine mammals that spend
long periods of time engaged in shallow dive behaviour following
a deep dive do so to ofﬂoad carbon dioxide (Gerlinsky et al.,
2014) and repay oxygen debt associated with the accumulation of
lactic acid due to exceeding cADL (Kooyman et al., 1980;Tyack
et al., 2006;Joyce et al., 2017). The deep dives made by this killer
whale were all near cADL, but deep diving behaviour alone may
lead to supersaturation of nitrogen in body tissue that could
make individuals vulnerable to diving related pathologies (Cox
et al., 2006). Additionally, the short intervals between some deep
dives recorded for this killer whale and rarely in some species of
beaked whales (Joyce et al., 2017) have been associated with
higher risk of decompression sickness (Wong, 1999). Although
marine mammals are thought to have evolved anatomical, physi-
ological and behavioural adaptations to reduce risk of decom-
pression sickness associated with nitrogen supersaturation, how
these features function is poorly understood (Kooyman et al.,
1972;Ridgway and Howard, 1979;Cox et al., 2006;Garcia
Movements of depredating whales 9
Pa´rraga et al., 2018). Durations of some of the deep dives
recorded for this killer whale are some of the shortest known for
any cetacean diving to such depths. It is possible that the limited
duration of these deep dives combined with following long peri-
ods of shallow diving help to mitigate any potential negative
In the vicinity of deployed gear
A positive correlation existed between the maximum dive depths
of the sperm whale and the depths of the nearest longline not be-
ing retrieved. This indicates that by remaining in close proximity
to a ﬁshing vessel over the course of several days, the sperm whale
may have learned the locations of deployed gear and, as docu-
mented in southern elephant seals (Mirounga leonina)(van den
Hoff et al., 2017), took advantage of opportunities to depredate
before gear retrieval began. However, as these dives were not al-
ways to the same depths as deployed longlines and depredation
took place when the gear was being retrieved, it is likely that they
also included natural foraging behaviour.
The lack of correlation between the killer whale’s dives and the
depths of gear deployed nearby veriﬁes that there was no depre-
dation from longlines before they were retrieved. This is not sur-
prising, because while toothﬁsh caught on lines are likely easy to
locate and capture due to their inability to ﬂee, those near the sea-
ﬂoor are less likely to be taken due to the limited amount of time
killer whales can spend at these depths. However, the depths of
two deep dives made by this whale in the vicinity of a line several
h after it was buoyed off were correlated with the depth to which
the longline had been stripped of toothﬁsh, suggesting that it dep-
redated from gear that had only been partially retrieved.
When foraging naturally
Characteristics of several of the dives made by the tagged killer
and sperm whales while not depredating are indicative of natural
foraging behaviour. For example, on several occasions while not
depredating, the tagged killer whale successively dove to depths
>200 m with little variance for periods up to 3.7 h. Similarly, the
tagged sperm whale engaged in repeated diving, but to a variety
of depths, when not depredating. The resolution of bathymetric
data for interpolated locations of the tagged whales was too low
to verify the exact depths of the seaﬂoor where they dove. Around
South Georgia, toothﬁsh occur in the benthic zone at a variety of
depths (Collins et al., 2010), however, cephalopods replace the
role of ﬁsh as mesopredators in the epipelagic zone (Rodhouse
and White, 1995) and are also abundant in the mesopelagic and
bathypelagic zones (Collins et al., 2004) where few, if any, large
ﬁsh species occur. Cephalopods have been documented in the
diets of several species around South Georgia, including sperm
whales and southern elephant seals (Clarke, 1980;Rodhouse
et al., 1992). Interestingly, the dive depths of southern elephant
seals over the shelf edge in this region (mean maximum—350 m;
McConnell and Fedak, 1996) are similar to the foraging dive
depths of the tagged killer whale. Many killer whale populations
feed at least in part on cephalopods (Nishiwaki and Handa, 1958;
Jonsga˚rd and Lyshoel, 1970;Ford et al., 1998;Yamada et al.,
2007;Hanson and Walker, 2014) and they have been documented
in the diet of killer whales in nearby Antarctica (Berzin and
Vladimirov, 1983) and the South Atlantic (Santos and Haimovici,
2001). Cephalopods are also predicted to constitute signiﬁcant
portions of the diets of killer whales known to feed on mammals
and birds, as well as depredate toothﬁsh from demersal longlines
off the Prince Edward Islands (Reisinger et al., 2015,2016).
However, several killer whale populations specialize on different
types of prey (Ford et al., 1998;Pitman and Ensor, 2003) and at
South Georgia, at least three distinct types are sympatric (Pitman
et al., 2010;Towers, 2015). Among them, only a population of
individuals hypothesized to be B2s based on morphology and be-
haviour depredates in the region (So¨ffker et al., 2015;Towers,
2015). Nitrogen isotope values indicate that B2s do not feed on
marine mammals (Durban et al., 2017), but aside from depre-
dated toothﬁsh, only penguins have been documented in their
diet (Pitman and Durban, 2010).
Implications and recommendations
This study provides key ﬁndings on the movements and dive be-
haviour of depredating killer and sperm whales that have implica-
tions for the toothﬁsh longline ﬁshery and its management. The
results also enhance our understanding of the behaviour and ecol-
ogy of killer and sperm whales off South Georgia. However, as
only one individual of each species was tagged, caution should be
taken when applying these results to larger populations due to the
potential for individual variations in behaviour.
Nevertheless, the deep dives made by the killer whale while
depredating were to depths this species was not previously
thought capable of attaining (Purves et al., 2004;Kock et al.,
2006;Clark and Agnew, 2010;Collins et al., 2010;Tixier et al.,
2010,2015b). However, the long recovery periods following these
dives may represent times that whales are physiologically con-
strained in their depredation capabilities. Additionally, whales
may be prone to lethal effects of acoustic disturbance during these
times. For instance, it is hypothesized that decompression sick-
ness documented in beaked whales occurs due to behavioural
responses to naval sonar when the whales are physiologically lim-
ited during their recovery periods following deep dives (Jepson
et al., 2004;Tyack et al., 2006). Both killer and sperm whales
change their dive behaviour in response to high intensity sound
(Sivle et al., 2012), thereby altering nitrogen levels in their bodies
and increasing risk of decompression sickness (Kvadsheim et al.,
2012). Therefore, the use of acoustic disturbance devices to deter
depredation, although seemingly ineffective (Tixier et al., 2015a;
Towers, 2015), may have implications for the health of deep div-
ing depredating whale populations at South Georgia.
Another key ﬁnding of this study is that the dive behaviour of
both species when not depredating is suggestive of a natural diet
that may include cephalopods. This indicates that depredating
killer and sperm whales at South Georgia only supplement their
natural diet with toothﬁsh obtained from commercial longlines.
However, it remains unclear if longlining has only beneﬁted
whale populations by providing easy access to toothﬁsh that was
not historically present, or if the effect that longlining has had on
the toothﬁsh stock in this region has reduced the natural avail-
ability of this prey resource for local whale populations. In any
case, since dietary preferences and associated foraging strategies
can evolve as different prey species become more or less available
in the environment, effective mitigation is therefore paramount
not only for protecting catch but also to ensure that whale popu-
lations do not become more reliant on depredated resources.
Data collected in this study indicate that some mitigation tech-
niques caused whales to disassociate from the ﬁshing vessel. For
instance, the horizontal movements of the killer and sperm whale
10 J. R. Towers et al.
were not always correlated with ﬁshing activity after gear was
buoyed off or when the San Aspiring left the area in which depre-
dation was occurring. This is consistent with previous ﬁndings by
Tixier et al. (2015b). Other studies have also shown that both spe-
cies exhibited a westward trend in occurrence throughout the
South Georgia ﬁshing season that was not correlated with ﬁshing
effort (Clark and Agnew, 2010;So¨ffker et al., 2015). This is sup-
ported by the photo-identiﬁcation data of Towers (2015) showing
that eight sperm whales moved 463 km west through the ﬁshing
grounds in 22 d. However, Towers (2015) and this study also
show that killer whales moved >300 km east through the ﬁshing
grounds in <50 h. Given the high mobility of these species and
the size of the ﬁshing grounds, further study into the large-scale
movements of depredating whales in relation to ﬁshing vessels
may help inform how, when, and where depredation can be
Further studies into the ﬁne-scale movements of whales while
gear is deployed are also necessary because some evidence from
this study indicates that sperm and killer whales may remove
toothﬁsh from longlines that are not being actively retrieved. For
instance, on one occasion the vertical and horizontal movements
of the tagged killer whale were correlated with the location of
ﬁshing gear that was buoyed off and the depth to which toothﬁsh
had been removed from it. This suggests that keeping hooked ﬁsh
at greater depths by attaching extra line before buoying off would
result in greater retention of catch. On the other hand, the behav-
iour of the tagged sperm whale was on several occasions corre-
lated with the depths and locations of deployed gear, indicating
that it may have been depredating at the seaﬂoor. However, this
practice cannot be too common or widespread or there would
not be much incentive for so many whales to depredate while
gear was being retrieved. Nevertheless, if depredation from
deployed gear were to become an effective means for sperm
whales to acquire prey, this practice could result in greater reduc-
tion of catch, increased uncertainty in stock assessments and fur-
ther difﬁculty utilizing efﬁcient mitigation techniques.
This study demonstrates the value of ﬁne-scale movement and
dive data to study depredation. However, considering that tags
are not without some risk to whale health and survival (Raverty,
2016), this technology should be used with caution whenever
conservation of the study population is of concern. Nevertheless,
continued research on the ecology and behaviour of depredating
whales at South Georgia will be important to help direct ﬁsheries
management practices and depredation mitigation procedures
and technology. To this end, we recommend that ﬁshing and pa-
trol vessels continue to be utilized at South Georgia to conduct
depredation studies in the region.
We thank the crew of the San Aspiring and Nick Wren for provid-
ing assistance in the ﬁeld, Dean Jurasovich and Trevor Joyce for
assistance with data preparation and providing helpful comments
on the manuscript, Jade Vacquie-Garcia and Mary-Anne Lea for
insightful comments on the methods, Marta So¨ffker and Martin
Collins for efforts towards study inception, Sue Gregory and Paul
Brewin for logistical support, Sanford Seafood, the team at King
Edward Point and Captain, and crew of Pharos SG for their hos-
pitality, and two anonymous reviewers for valuable feedback.
Permits for this work were provided by the Government of South
Georgia and South Sandwich Islands and the New Zealand
Department of Conservation. Funding for ﬁeldwork was provided
by the Government of South Georgia and South Sandwich
Islands and manuscript preparation was funded in part by St.
Ashford, J. R., Rubilar, P. S., and Martin, A. R. 1996. Interactions be-
tween cetaceans and longline ﬁshing operations around South
Georgia. Marine Mammal Science, 12: 452–457.
Baird, R. W., Hanson, M. B., and Dill, L. M. 2005. Factors inﬂuencing
the diving behaviour of ﬁsh-eating killer whales: sex differences
and diel and interannual variation in diving rates. Canadian
Journal of Zoology, 83: 257–267.
Berzin, A. A., and Vladimirov, V. L. 1983. A new species of killer
whale (Cetacea, Delphinidae) from Antarctic waters.
Zoologicheskii Zhurnal, 62: 287–295.
Clark, J. M., and Agnew, D. 2010. Estimating the impact of depreda-
tion by killer whales and sperm whales on longline ﬁshing for
toothﬁsh (Dissostichus eleginoides) around South Georgia.
CCAMLR Science, 17: 163–178.
Clarke, M. R. 1976. Observation on sperm whale diving. Journal of
the Marine Biological Association of the United Kingdom, 56:
Clarke, M. R. 1980. Cephalopoda in the diet of sperm whales of the
Southern Hemisphere and their bearing on sperm whale biology.
Discovery Reports, 37: 1–324.
Collins, M. A., Allcock, A. L., and Belchier, M. 2004. Cephalopods of
the South Georgia slope. Journal of the Marine Biological
Assocation of the United Kingdom, 84: 415–419.
Collins, M. A., Brickle, P., Brown, J., and Belchier, M. 2010. The
Patagonian toothﬁsh: biology, ecology and ﬁshery. Advances in
Marine Biology, 58: 227–300.
Cox, T. M., Ragen, A. J., Vos, E., Baird, R. W., Balcomb, K., Barlow,
J., and Caldwell, J. 2006. Understanding the impact of anthropo-
genic sound on beaked whales. Journal of Cetacean Research and
Management, 7: 177–187.
Dahlheim, M. E. 1988. Killer whale (Orcinus orca) depredation on
longline catches of sableﬁsh (Anoplopoma ﬁmbria) in Alaskan
Waters. NWAFC Processed Report, 88-14. Alaska Fish Sci Cent,
NMFS, NOAA, Seattle, Washington.
Durban, J. W., and Pitman, R. L. 2012. Antarctic killer whales make
rapid, round-trip movements to subtropical waters: evidence for
physiological maintenance migrations?. Biology Letters, 8: 274–277.
Durban, J. W., and Pitman, R. L. 2013. Out of Antarctica: dive data
support ‘physiological maintenance migration’ in Antarctic Killer
Whales. In 20th Biennial Conference on the Biology of Marine
Mammals. Society for Marine Mammalogy (9–13 December
2013, Dunedin, New Zealand).
Durban, J. W., Fearnbach, H., Burrows, D. G., Ylitalo, G. M., and
Pitman, R. L. 2017. Morphological and ecological evidence for
two sympatric forms of Type B killer whale around the Antarctic
Peninsula. Polar Biology, 40: 231–236.
Fais, A., Soto, N. A., Johnson, M., Pe´rez-Gonza´ lez, C., Miller, P. J. O.,
and Madsen, P. T. 2015. Sperm whale echolocation behaviour
reveals a directed, prior-based search strategy informed by prey
distribution. Behavioural Ecology and Sociobiology, 69: 663–674.
Fearnbach, H., Durban, J. W., Ellifrit, D. K., Waite, J. M., Matkin, C.
O., Lunsford, C. R., Peterson, M. J., et al. 2014. Spatial and social
connectivity of ﬁsh-eating “Resident” killer whales (Orcinus orca)
in the northern North Paciﬁc. Marine Biology, 161: 459–472.
Foote, A. D., Vijay, N., A
´vila-Arcos, M. C., Baird, R. W., Durban, J.
W., Fumagalli, M., Gibbs, R. A., et al. 2016. Genome-culture co-
evolution promotes rapid divergence of killer whale ecotypes.
Nature Communications, 7: 11693.
Ford, J. K. B. 2009. Killer whale: Orcinus orca.In Encyclopedia of
Marine Mammals. 2nd edn, pp. 650–657. Ed. by W. F. Perrin, B.
Wursig, and J. G. M. Thewissen. Academic Press, San Diego,
Movements of depredating whales 11
Ford, J. K. B. 2014. Marine Mammals of British Columbia. Royal BC
Museum, Victoria, British Columbia. 460 pp.
Ford, J. K. B., Ellis, G. M., Barrett-Lennard, L. G., Morton, A. B.,
Palm, R. S., Balcomb, K. C. III, 1998. Dietary specialization in two
sympatric populations of killer whales (Orcinus orca) in coastal
British Columbia and adjacent waters. Canadian Journal of
Zoology, 76: 1456–1471.
Freitas, C., Lydersen, C., Fedak, M. A., and Kovacs, K. M. 2008. A
simple new algorithm to ﬁlter marine mammal Argos locations.
Marine Mammal Science, 24: 315–325.
Garcia Pa´rraga, D., Moore, M., and Fahlman, A. 2018. Pulmonary
ventilation – perfusion mismatch: a novel hypothesis for how div-
ing vertebrates may avoid the bends. Proceedings of the Royal
Society B, 285: 20180482.
Gasco, N., Tixier, P., Duhamel, G., and Guinet, C. 2015. Comparison
of two methods to assess ﬁsh losses due to depredation by killer
whales and sperm whales on demersal longlines. CCAMLR
Science, 22: 1–14.
Gaskin, D. E., and Cawthorn, M. W. 1967. Diet and feeding habits of
the sperm whale (Physeter catadon L.) in the Cook Strait region of
New Zealand. New Zealand Journal of Marine and Freshwater
Research, 1: 156–179.
Gerlinsky, C. D., Rosen, D. A. S., and Trites, A. W. 2014. Sensitivity
to hypercapnia and elimination of CO
following diving in Steller
sea lions (Eumetopias jubatus). Journal of Comparative
Physiology B, 184: 535–544.
Gilman, E., Brothers, N., McPherson, G., and Dalzell, P. 2006. A re-
view of cetacean interactions with longline gear. Journal of
Cetacean Research and Management, 8: 215–223.
Goetz, S., Laporta, M., Martı´nez Portela, J., Santos, M. B., and Pierce,
G. J. 2011. Experimental ﬁshing with an “umbrella-and-stones”
system to reduce interactions of sperm whales (Physeter macroce-
phalus) and seabirds with bottom-set longlines for Patagonian
toothﬁsh (Dissostichus eleginoides) in the Southwest Atlantic.
ICES Journal of Marine Science, 68: 228–238.
Government of South Georgia & South Sandwich Islands. 2017.
South Georgia & the South Sandwich Islands Toothﬁsh Fishery
(48.3 and 48.4) Management Plan 2018. Government House,
Stanley, Falkland Islands.
Guerra, M., Hickmott, L., van der Hoop, J., Rayment, W., Leunissen,
E., Slooten, E., and Moore, M. 2017. Diverse foraging strategies by
a marine top predator: sperm whales exploit pelagic and demersal
habitats in the Kaik
oura submarine canyon. Deep Sea Research
Part I, 128: 98–108.
Guinet, C., Domenici, P., de Stephanis, R., Barrett-Lennard, L., Ford,
J. K. B., and Verborgh, P. 2007. Killer whale predation on blueﬁn
tuna: exploring the hypothesis of the endurance-exhaustion tech-
nique. Marine Ecology Progress Series, 347: 111–119.
Guinet, C., Tixier, P., Gasco, N., and Duhamel, G. 2015. Long-term
studies of Crozet Island killer whales are fundamental to under-
standing the economic and demographic consequences of their
depredation behaviour on the Patagonian toothﬁsh ﬁshery. ICES
Journal of Marine Science, 72: 1587–1597.
Hamer, D. J., Childerhouse, S. J., and Gales, N. J. 2012. Odontocete
bycatch and depredation in longline ﬁsheries: a review of available
literature and potential solutions. Marine Mammal Science, 28:
Hanson, B., and Walker, W. A. 2014. Trans-Paciﬁc consumption of
cephalopods by North Paciﬁc killer whales (Orcinus orca).
Aquatic Mammals, 40: 274–284.
Hucke-Gaete, R., Moreno, C. A., and Arata, J. 2004. Operational
interactions of sperm whales and killer whales with the
Patagonian toothﬁsh industrial ﬁshery off southern Chile.
CCAMLR Science, 11: 127–140.
Janc, A., Richard, G., Guinet, C., Arnould, J. P. Y., Villanueva, M. C.,
Duhamel, G., Gasco, N., et al. 2018. How do ﬁshing practices
inﬂuence sperm whale (Physeter macrocephalus) depredation on
demersal longline ﬁsheries?. Fisheries Research, 206: 14–26.
Jepson, P. D., Arbelo, M., Deaville, R., Patterson, I. A. P., Castro, P.,
Baker, J. R., Degollada, E., et al. 2004. Whales, sonar and decom-
pression sickness. Nature, 428: 2–576.
Jonsga˚rd, A., and Lyshoel, P. B. 1970. A contribution to the knowl-
edge of the biology of the killer whale. Norwegian Journal of
Zoology, 18: 41–48.
Joyce, T. W., Durban, J. W., Claridge, D. E., Dunn, C. A., Fearnbach,
H., Parsons, K. M., Andrews, R. D., et al. 2017. Physiological,
morphological, and ecological tradeoffs inﬂuence vertical habitat
use of deep-diving toothed-whales in the Bahamas. PLoS One, 12:
Kock, K. H., Purves, M. G., and Duhamel, G. 2006. Interactions be-
tween cetacean and ﬁsheries in the Southern Ocean. Polar
Biology, 29: 379–388.
Kooyman, G. L., Schroeder, J. P., Denison, D. M., Hammond, D. D.,
Wright, J. J., and Bergman, W. P. 1972. Blood nitrogen tensions
of seals during simulated deep dives. American Journal of
Physiology, 223: 1016–1020.
Kooyman, G. L., Wahrenbrock, E. A., Castellini, M. A., Davis, R. W.,
and Sinnett, E. E. 1980. Aerobic and anaerobic metabolism during
voluntary diving in Weddell seals: evidence of preferred pathways
from blood chemistry and behavior. Journal of Comparitive
Physiology B, 138: 335–346.
Kvadsheim, P. H., Miller, P. J. O., Tyack, P., Sivle, L. D., Lam, F. P.
A., and Fahlman, A. 2012. Estimated tissue and blood N2 levels
and risk of decompression sickness in deep intermediate and shal-
low diving toothed whales during exposure to naval sonar.
Frontiers in Aquatic Physiology, 3: 125.
Lopez, R., and Malarde´, J-P. 2011. Improving ARGOS Doppler
Location Using Kalman Filtering. CLS-DT-MEMO-11-65.
Ramonville Saint-Agne, France.
Martin, A. R., and Clarke, M. R. 1986. The diet of sperm whales
(Physeter macrocephalus) captured between Iceland and
Greenland. Journal of the Marine Biological Association of the
United Kingdom, 66: 779–790.
Mathias, D., Thode, A., Straley, J., and Folkert, K. 2009. Relationship
between sperm whale (Physeter macrocephalus) click structure and
size derived from videocamera images of a depredating whale.
Journal of the Acoustical Society of America, 125: 3444–3453.
Mathias, D., Thode, A., Straley, J., Calambokidis, J., Schorr, G., and
Folkert, K. 2012. Acoustic and diving behavior of sperm whales
(Physeter macrocephalus) during natural and depredation foraging
in the Gulf of Alaska. Journal of the Acoustical Society of
America, 132: 518–532.
McConnell, B. J., and Fedak, M. A. 1996. Movements of southern ele-
phant seals. Canadian Journal of Zoology, 74: 1485–1496.
Miller, P. J. O., Shapiro, A. D., and Deecke, V. B. 2010. The diving be-
haviour of mammal-eating killer whales (Orcinus orca): variations
with ecological not physiological factors. Canadian Journal of
Zoology, 88: 1103–1112.
Moore, J. K., Abbott, M. R., and Richman, J. G. 1999. Location and
dynamics of the Antarctic polar front from satellite sea surface
temperature data. Journal of Geophysical Research, 104:
Moreno, C. A., Castro, R., Mujica, L. J., and Reyes, P. 2008.
Signiﬁcant conservation beneﬁts obtained from the use of a new
ﬁshing gear in the Chilean Patagonian toothﬁsh ﬁshery.
CCAMLR Science, 15: 79–91.
Morin, P. A., Archer, F. I., Foote, A. D., Vilstrup, J., Allen, E. E.,
Wade, P., Durban, J., et al. 2010. Complete mitochondrial ge-
nome phylogeographic analysis of killer whales (Orcinus orca)
indicates multiple species. Genome Research, 20: 908–916.
Nishiwaki, M., and Handa, C. 1958. Killer whales caught in the
coastal waters off Japan for recent 10 years. Scientiﬁc Reports of
the Whales Research Institute, 13: 85–96.
12 J. R. Towers et al.
Nolan, C. P., Liddle, G. M., and Elliot, J. 2000. Interactions between
killer whales (Orcinus orca) and sperm whales (Physeter macroce-
phalus) with a longline ﬁshing vessel. Marine Mammal Science,
O’Connell, V., Straley, J., Liddle, J., Wild, L., Behnken, L., Falvey, D.,
and Thode, A. 2015. Testing a passive deterrent on longlines to re-
duce sperm whale depredation in the Gulf of Alaska. ICES Journal
of Marine Science, 72: 1667–1672.
Pante, E., and Simon-Bouhet, B. 2013. Marmap: a package for
importing, plotting and analyzing bathymetric and topographic
data in R. PLoS One, 8: e73051.
Peterson, M. J., Mueter, F., Criddle, K., and Haynie, A. C. 2014. Killer
whale depredation and associated costs to Alaskan sableﬁsh,
Paciﬁc halibut and Greenland turbot longliners. PLoS One, 9:
Pitman, R. L., and Durban, J. W. 2010. Killer whale predation on
penguins in Antarctica. Polar Biology, 33: 1589–1594.
Pitman, R. L., and Ensor, P. 2003. Three forms of killer whales
(Orcinus orca) in Antarctic waters. Journal of Cetacean Research
and Management, 5: 131–139.
Pitman, R. L., Durban, J. W., Greenfelder, M., Guinet, C., Jorgensen,
M., Olson, P. A., Plana, J., et al. 2010. Observations of a distinc-
tive morphotype of killer whale (Orcinus orca), type D, from sub-
antarctic waters. Polar Biology, 34: 303–306.
Pitman, R. L., Perryman, W. L., LeRoi, D., and Eilers, E. 2007. A
dwarf form of killer whale in Antarctica. Journal of Mammalogy,
Purves, M. G., Agnew, D. J., Balguerı´as, E., Moreno, C. A., and
Watkins, B. 2004. Killer whale (Orcinus orca) and sperm whale
(Physeter macrocephalus) interactions with longline vessels in the
Patagonian toothﬁsh ﬁshery at South Georgia, South Atlantic.
CCAMLR Science, 11: 111–126.
R Core Team. 2017. R: A Language and Environment for Statistical
Computing. https://www.R-project.org/ (last accessed 8 August
Raverty, S. 2016. Final Report AHC Case: 16-1760. 7 pp.
Read, A. J. 2008. The looming crisis: interactions between marine
mammals and ﬁsheries. Journal of Mammalogy, 89: 541–548.
Reisinger, R. R., Gro¨ cke, D. R., Lu¨bcker, N., McClymont, E. L.,
Hoelzel, A. R., and de Bruyn, P. J. N. 2016. Variation in the diet
of killer whales Orcinus orca at Marion Island, Southern Ocean.
Marine Ecology Progress Series, 549: 263–274.
Reisinger, R. R., Keith, M., Andrews, R. D., and de Bruyn, P. J. N.
2015. Movement and diving of killer whales (Orcinus orca)ata
Southern Ocean archipelago. Journal of Experimental Marine
Biology and Ecology, 473: 90–102.
Reisinger, R. R., Oosthuizen, W. C., Pe´ron, G., Toussaint, D. C.,
Andrews, R. D., and de Bruyn, P. J. N. 2014. Satellite tagging and
biopsy sampling of killer whales at Subantarctic Marion Island: ef-
fectiveness, immediate reactions and long-term responses. PLoS
One, 9: e111835.
Ridgway, S. H., and Howard, R. 1979. Dolphin lung collapse and in-
tramuscular circulation during free diving: evidence from nitro-
gen washout. Science, 206: 1182–1183.
Roche, C., Guinet, C., Gasco, N., and Duhamel, G. 2007. Marine
mammals and demersal longline ﬁshery interactions in Crozet
and Kerguelen Exclusive Economic Zones: an assessment of dep-
redation levels. CCAMLR Science, 14: 67–82.
Rodhouse, P. G., and White, M. G. 1995. Cephalopods occupy the
ecological niche of epipelagic ﬁsh in the Antarctic polar front
zone. Biology Bulletin, 189: 77–80.
Rodhouse, P. G., Arnbom, T. R., Fedak, M. A., Yeatman, J., and
Murray, A. W. A. 1992. Cephalopod prey of the southern elephant
seal, Mirounga leonina L. Canadian Journal of Zoology, 70:
Santos, R. A., and Haimovici, M. 2001. Cephalopods in the diet of
marine mammals stranded or incidentally caught along
southeastern and southern Brazil (21–34S). Fisheries Research,
Schakner, Z. A., Lunsford, C., Straley, J., Eguchi, T., and Mesnick, S.
L. 2014. Using models of social transmission to examine the
spread of longline depredation behavior among sperm whales in
the Gulf of Alaska. PLoS One, 9: e109079.
Schorr, G. S., Falcone, E. A., Moretti, D. J., and Andrews, R. D. 2014.
First long-term behavioural records from Cuvier’s beaked whales
(Ziphius cavirostris) reveal record-breaking dives. PLoS One, 9:
Sigler, M. F., Lunsford, C. R., Straley, J. M., and Liddle, J. B. 2008.
Sperm whale depredation of sableﬁsh longline gear in the north-
east Paciﬁc Ocean. Marine Mammal Science, 24: 16–27.
Sivle, L. D., Kvadsheim, P. H., Fahlman, A., Lam, F. P. A., Tyack, P.
L., and Miller, P. J. O. 2012. Changes in dive behaviour during na-
val sonar exposure in killer whales, long-ﬁnned pilot whales, and
sperm whales. Frontiers in Aquatic Physiology, 3: 400.
So¨ffker, M., Trathan, P., Clark, J., Collins, M. A., Belchier, M., and
Scott, R. 2015. The impact of predation by marine mammals on
Patagonian Toothﬁsh longline ﬁsheries. PLoS One, 10: e0118113.
Straley, J., O’Connell, V., Liddle, J., Thode, A., Wild, L., Behnken, L.,
Falvey, D., et al. 2015. Southeast Alaska Sperm Whale Avoidance
Project (SEASWAP): a successful collaboration among scientists
and industry to study depredation in Alaskan waters. ICES
Journal of Marine Science, 72: 1598–1609.
Straley, J. M., Schorr, G. S., Thode, A. M., Calambokidis, J.,
Lunsford, C. R., Chenoweth, E. M., O’Connell, V. M., et al. 2014.
Depredating sperm whales in the Gulf of Alaska: local habitat use
and long distance movements across putative population bound-
aries. Endangered Species Research, 24: 125–135.
Teloni, V., Johnson, M. P., Miller, P. J. O., and Madsen, P. T. 2008.
Shallow food for deep divers: dynamic foraging behavior of male
sperm whales in a high latitude habitat. Journal of Experimental
Marine Biology and Ecology, 354: 119–131.
Thode, A., Mathias, D., Straley, J., O’Connell, V., Behnken, L., Falvey,
D., Wild, L., et al. 2015. Cues, creaks, and decoys: using passive
acoustic monitoring as a tool for studying sperm whale depreda-
tion. ICES Journal of Marine Science, 72: 1621–1636.
Thode, A., Straley, J., Mathias, D., Wild, L., Falvey, D., O’Connell, T.,
and Behnken, L. 2012. Reducing sperm whale depredation via de-
coy deployments and active deterrent testing. North Paciﬁc
Research Board Final Report 0918, 235 pp.
Tixier, P. 2012. Depredation par les orques (Orcinus Orca) et les
cachalots (Physeter macrocephalus) sur les palangriers _a la legine
australe dans la ZEE de l’archipel de Crozet [Depredation of killer
whales (Orcinus orca) and sperm whales (Physeter macrocephalus)
on the Patagonian toothﬁsh longline ﬁshery in the Crozet and
Kerguelen EEZs]. Ph.D. dissertation, CEBC-CNRS, University of
Aix-Marseille II, Marseille, France. 367 pp. Available at: http://
Tixier, P., Garcia, J. V., Gasco, N., Duhamel, G., and Guinet, C.
2015b. Mitigating killer whale depredation on demersal longline
ﬁsheries by changing ﬁshing practices. ICES Journal of Marine
Science, 72: 1610–1620.
Tixier, P., Gasco, N., Duhamel, G., and Guinet, C. 2015a.
Habituation to an acoustic harassment device (AHD) by killer
whales depredating demersal longlines. ICES Journal of Marine
Science, 72: 1673–1681.
Tixier, P., Gasco, N., Duhamel, G., and Guinet, C. 2016. Depredation
of Patagonian toothﬁsh (Dissostichus eleginoides) by two sympatri-
cally occurring killer whale (Orcinus orca) ecotypes: insights on
the behavior of the rarely observed type D killer whales. Marine
Mammal Science, 32: 983–1003.
Tixier, P., Gasco, N., Duhamel, G., Viviant, M., Authier, M., and
Guinet, C. 2010. of Patagonian toothﬁsh ﬁsheries with killer and
sperm whales in the Crozet Islands Exclusive Economic Zone: an
assessment of depredation levels and insights on possible mitiga-
tion strategies. CCAMLR Science, 17: 179–195. Interactions
Movements of depredating whales 13
Towers, J. R. 2015. Final Report for Contract as Cetacean Research
Technician for the Government of South Georgia and South
Sandwich Islands in 2015.
Tyack, P. L., Johnson, M., Aguilar Soto, N., Sturlese, A., and Madsen,
P. T. 2006. Extreme diving of beaked whales. Journal of
Experimental Biology, 209: 4238–4253.
van den Hoff, J., Kilpatrick, R., and Welsford, D. 2017. Southern ele-
phant seals (Mirounga leonina Linn.) depredate toothﬁsh long-
lines in the midnight zone. PLoS One, 12: e0172396.
Watkins, W. A., Daher, M. A., Dimarzio, N. A., Samuels, A.,
Wartzok, D., Fristrup, K. M., Howey, P. W., et al. 2002. Sperm
whale dives tracked by radio tag telemetry. Marine Mammal
Science, 18: 55–68.
Watkins, W. A., Moore, K. E., and Tyack, P. L. 1985. Sperm whale
acoustic behaviors in the southeast Caribbean. Cetology, 49: 1–15.
Werner, T. B., Northridge, S., Press, K. M., and Young, N. 2015.
Mitigating bycatch and depredation of marine mammals in long-
line ﬁsheries. ICES Journal of Marine Science, 72: 1576–1586.
Whitehead, H. 2003. Sperm Whales: Social Evolution in the Ocean.
University of Chicago Press, Chicago. 456 pp.
Whitehead, H. 2009. Sperm whale: Physeter microcephalus.In
Encyclopedia of Marine Mammals. 2nd edn, pp. 1091–1097. Ed.
By W. F. Perrin, B. Wursig, and J. G. M. Thewissen. Academic
Press, San Diego, California.
Wong, R. M. 1999. Tarvana revisited: decompression illness after
breath-hold diving. SPUMS Journal, 29: 126–131.
Wright, B. M., Ford, J. K. B., Ellis, G. M., Deecke, V. B.,
Shapiro, A. D., Battaile, B. C., and Trites, A. W. 2017.
Fine-scale foraging movements by ﬁsh-eating killer whales
(Orcinus orca) relate to the vertical distributions and escape
responses of salmonid prey (Oncorhynchus spp.). Movement
Ecology, 1: 1181.
Wright, B. M., Stredulinsky, E. H., Ellis, G. M., and Ford, J. K. B.
2016. Kin-directed food sharing promotes lifetime natal philopa-
try of both sexes in a population of ﬁsh eating killer whales,
Orcinus orca. Animal Behaviour, 115: 81–95.
Yamada, T. K., Uni, Y., Amano, M., Brownell, Jr, R. L., , Sato, H.,
Ishikawa, S., Ezaki, I., et al. 2007. Biological indices obtained from
a pod of killer whales entrapped by sea ice off northern Japan.
Report SC/59/SM12 to the IWC Scientiﬁc Committee.
Zuur, A. F., Ieno, E. N., Walker, N. J., Saveliev, A. A., and Smith, G.
M. 2009. Mixed Effects Models and Extensions in Ecology with R
Mixed Effects Models and Extensions in Ecology with R. Springer,
Handling editor: Simon Northridge
14 J. R. Towers et al.