1 1
Corresponding author: Leigh J. Klatsky, Dolphin Quest Inc. / Quest Global
2
Management, 1880 Harbor Island Drive, San Diego, CA 92101 USA E‐mail:
3
[email protected]
4
5
Running head: Offshore bottlenose dolphins near Bermuda
6
7
8
9
Offshore Bottlenose Dolphins (Tursiops truncatus): Movement and Dive Behavior
10
near the Bermuda Pedestal
11
12
13
14
Leigh J. Klatsky, Randall S. Wells, Jay C. Sweeney
15
16
17
18
Dolphin Quest Inc. / Quest Global Management, 1880 Harbor Island Drive, San Diego,
19
CA 92101, (LJK, JCS)
20
Chicago Zoological Society c/o Mote Marine Laboratory, 1600 Ken Thompson Parkway,
21
Sarasota, FL, 34236, (RSW)
2 ABSTRACT
22 23
24
Little is known about the behavior of offshore Atlantic bottlenose dolphins (Tursiops
25
truncatus) in deep water and near oceanic islands. Using satellite‐linked, time‐depth
26
recorders, the movements and dive behavior of offshore Atlantic bottlenose dolphins in
27
the deep waters surrounding the Bermuda Pedestal were investigated. Three dolphins
28
were tracked from 5 to 45 days and traveled a mean distance of 28.3 km/day where
29
mean water depth was ‐1402.0±1120.7 m. Regular dives during the night (2100‐0259
30
local time) to depths greater than 450 meters (mean±SD, 8.5±3.88 m), 46.4% of night
31
dives lasting longer than five minutes, and high hematocrit values reveal the deep
32
diving capabilities of offshore bottlenose dolphins. Shallow dives, 93.3% within 26 m of
33
the surface, and short dive durations, 52.7% lasting less than 1 min, during the day
34
(0900‐1459 local time) followed by an increase in the total number of dives at dusk
35
(1500‐2059 local time) (mean = 72.4±19.61 m), indicate a possible diel dive cycle. The
36
dive patterns of bottlenose dolphins in Bermuda waters correlates with the reported
37
nightly vertical migrations of mesopelagic prey along the steeply‐sided Bermuda
38
Pedestal.
39
40
Key Words: Bermuda, bottlenose dolphin, dive behavior, movements, offshore, Tursiops
41
truncatus
3 42
INTRODUCTION
43
44
Bottlenose dolphins (Tursiops truncatus) are found in most temperate and tropical
45
waters around the world, including coastal waters of all continents, around most
46
oceanic islands and atolls, and over shallow offshore banks and shoals (Reynolds et al.
47
2000; Rice 1998; Wells and Scott 1999). Although they are common in coastal and
48
pelagic waters, the majority of research on this species has been conducted in nearshore
49
waters due to logistical considerations and the lack of survey effort in offshore waters.
50
The behavior of offshore and island populations of bottlenose dolphins has not been
51
well‐studied.
52
There are two distinct ecotypes of bottlenose dolphins in the northwest Atlantic
53
Ocean: a nearshore, shallow, warm water ecotype and an offshore deep, cold water
54
ecotype (Kenney 1990; Leatherwood and Reeves 1982; Mead and Potter 1995; Wells and
55
Scott 1999). Most of what is known about the offshore ecotype has been based on
56
studies of stranded individuals along the east coast of the United States. Stranded
57
offshore dolphins have a 15% greater total adult body length, shorter and wider rostra,
58
smaller pectoral flippers, wider skulls, and larger nares than those displayed by the
59
coastal ecotype (Hersh and Duffield 1990; Mead and Potter 1995). These two forms also
60
exhibit differences in the number of electrophoretically distinct hemoglobin profiles; the
61
offshore ecotype expresses two hemoglobin types while the coastal ecotype exhibits
4 62
only one type (Duffield et al. 1983; Hersh and Duffield 1990). In addition, the two
63
ecotypes have been found to be genetically distinct based on mitochondrial DNA
64
(Curry and Smith 1997; Dowling and Brown 1993; Hoelzel et al. 1998), and nuclear
65
DNA (Hoelzel et al. 1998). Mead and Potter (1995) were also able to differentiate
66
ecotypes based on preferences in prey and differences in parasitic load. Results from an
67
opportunistic tracking study of two stranded and rehabilitated offshore dolphins from
68
Florida using satellite‐linked radio telemetry provide the only information about the
69
movement patterns of offshore dolphins in deep water (> 5000 meters) (Wells et al.
70
1999).
71
Because most studies of bottlenose dolphins have focused on the shallow water
72
ecotype, which have limited deep‐diving opportunities and are not known as deep
73
divers, little is known about the dive behavior of the offshore ecotype. Physical
74
characteristics found in the offshore ecotype indicate the possibility of deep diving
75
capabilities. Hersh and Duffield (1990) suggested offshore dolphins might possess an
76
adaptation for greater oxygen‐binding capacity and related breath‐hold capacity due to
77
higher values of hematocrits, hemoglobin concentration, and red blood cell counts
78
(Duffield et al. 1983). Along with these oxygen‐binding characteristics, an increased
79
respiratory function could be associated with the relatively greater nareal diameter
80
found in the offshore ecotype (Mead and Potter 1995).
5 81
Although marine resource distributions in the open ocean are highly variable, there
82
is a predictable increase in primary production and fish aggregations near areas with
83
steeply sloping benthic topography, such as shelf breaks and seamounts (Fiedler 2002;
84
Rogers 1994). Relationships between geographical patterns of the sea floor (depth and
85
depth gradient) and distributions have been found for several cetacean species (Gregr
86
and Trites 2001; Hooker et al. 1999; Norris et al. 1994; Waring et al. 2001). Currently,
87
there is very little information about offshore bottlenose dolphins inhabiting areas near
88
steep topographical structures in the open ocean. Acevedo‐Gutierrez (1999) reported
89
that a large population of bottlenose dolphins near Coco Island, offshore of Costa Rica,
90
is most likely transitory since most individuals were sighted only once, while work by
91
Baird et al. (2001, 2002) in the Hawaiian island chain, suggest small populations of
92
bottlenose dolphins are island‐associated in near‐shore waters (<200 m) along each of
93
the four main islands.
94
The goal of this study is to describe the occurrence, movements and dive behavior of
95
offshore bottlenose dolphins in the northwest Atlantic Ocean near the Bermuda
96
Pedestal. Sightings of bottlenose dolphins in Bermuda waters were not reported prior
97
to March 2001. Unfortunately, most sightings have been described by fishermen in
98
search of pelagic fish between the island of Bermuda and two banks to the southwest
99
(see Figure 1) and do not provide adequate coordinates for mapping purposes. Using
100
satellite‐linked time‐depth recorders, this study presents the first free‐ranging
Comment [Dolphin1]: Not in Lit Cited section.
6 101
movements and dive information for this species in offshore waters near an oceanic
102
island.
103
104
MATERIALS AND METHODS
105
106
Study Area.–This study was conducted in the deep waters surrounding the Bermuda
107
Pedestal, a steep‐sided volcanic peak, approximately 1000 km2 in area, and rising over
108
4000 m off the seafloor. The study area is made up of the island of Bermuda; two banks,
109
Challenger and Argus, located 19 and 32 km to the southwest with summits that reach
110
within 50 m of the surface; and the Bowditch seamount, located approximately 40 km to
111
the northeast that reaches within 800 m of the surface at its shallowest point (Fig. 1).
112
The closest major landmass is the North American continent, specifically Cape Hatteras,
113
North Carolina, located approximately 965 km to the northwest.
114
Animal capture and telemetry.–In June 2003, a 13.3 m converted lobster boat with a
115
bow pulpit for hoop‐netting was used to locate and catch bow‐riding dolphins (Asper
116
1975). Swimmers maneuvered the dolphins in the net into a sling suspended over the
117
side of the boat from a boom, and then a hydraulic lift brought each dolphin onto the
118
research vessel. Body weight was determined using a load cell and full body
119
morphometrics were taken. Blood samples were taken to evaluate health and
120
reproductive status, and for genetic analysis. Ultrasound was performed on the testes,
7 121
uterus, and ovaries to determine reproductive status. Results from blood hematology
122
and hemoglobin electrophoresis were used to verify that the three tagged animals were
123
the offshore ecotype.
124
Dolphins were fitted with dorsal fin mounted satellite‐linked time‐depth recorders
125
(SLTDRs, Wildlife Computers, STR‐T16, 10.1 x 5.7 x 2.3 cm, 135 g, Wildlife Computers,
126
Redmond, WA). SLTDRs consisted of a satellite transmitter (PTT) (ST‐16, Telonics), 22‐
127
cm semi‐rigid antennae, controller board (Wildlife Computers), pressure transducer
128
(Wildlife Computers), and one M1 battery cast in epoxy. SLTDRs were bolted to the
129
dorsal fin with three 0.64‐cm‐diameter Delrin pins and secured on the opposite side of
130
the dorsal fin with non‐stainless steel corrosible nuts and washers. Closed‐cell foam
131
attached to the tag and washers reduced the likelihood of abrasion to the dorsal fin.
132
The capture and release process was approved by the San Diego State University
133
Animal Care Committee.
134
Dive data were collected by a pressure transducer in the tag and compressed with a
135
microprocessor‐controlled dive recorder (Wildlife Computers, Redmond, WA).
136
Recorded data were grouped into four 6‐hr periods set to local Atlantic Standard Time:
137
Dawn = 0300‐0859, Day = 0900‐1459, Dusk = 1500‐2059, and Night = 2100‐0259. The
138
time‐depth recorders had a maximum depth range of 500 m and a 2 m resolution. A
139
dive was defined as any submergence greater than 6 m to differentiate between actual
140
dives and time spent at the surface. The time‐depth recorders measured three variables:
8 141
maximum depth of each dive, time spent at depth, and duration of dive. Data for
142
maximum depth and time at depth of each dive were compiled into bins for each 6‐hr
143
period. The maximum depth range bins and time spent at depth bins were initially set
144
to measure to 250 m but were adjusted in the field to 500 m after the first dolphin
145
tagged repeatedly exceeded 250 m. Duration of dive data were collected in six bins (1
146
min, 2 min, 3 min, 4 min, 5 min, and > 5 min). In addition, a maximum daily dive depth
147
was recorded for each 24‐hr period.
148
Location and dive data were transmitted with an 8‐hr duty cycle and 45 s repetition
149
rate to the Argos satellite system (Service ARGOS, Inc., Landover, Maryland, USA).
150
Position accuracy was designated to 6 location classes (LC). An LC 3 has an accuracy
151
estimate of less than ≤ 150 m, LC 2 ≤ 350 m, LC 1 ≤ 1000 m, LC 0 > 1000 m. Location
152
classes A and B require only two transmissions to provide a location with unknown
153
quality and therefore do not give an accuracy estimate (Service Argos 2003).
154
Locational analysis.–Locational data were examined using Argos_Filter V6.2
155
(Prepared by Dave Douglas, USGS, Alaska Science Center, Alaska, USA) to filter all
156
plausible locations. The program identifies implausible locations based on two
157
different filtering methods. The first is a minimum‐redundant‐distance (MRD), a user‐
158
defined distance utilized as a threshold for determining redundant locations. The
159
second method is the distance, angle, and rate measurement (DAR) that attempts to
160
remove implausible locations based on the fact that most suspicious ARGOS locations
9 161
cause the animal to incorrectly move a substantial distance and then return with the
162
subsequent location being more correct (Douglas 2000). A rate of 20 km/hr was used as
163
the mean rate of travel threshold to evaluate the plausibility of the locations. Three
164
kinds of output are produced by this program, one for each of the filtering strategies
165
and a third more‐experimental output that pools selected DAR locations to the MDR
166
results. This program also defines the best daily location per dolphin based on the
167
pooled MDR and DAR outputs. From the filter output, distance traveled is calculated
168
as the distance (km) from the previous plausible location and mean rate of travel is
169
calculated as the rate of movement (km/hr) between two successive locations. These
170
distances do not describe the actual distance traveled in a day, but rather the distances
171
between successive points.
172
Analysis of movement data was performed using Arcview Geographic Information
173
System (GIS) (ESRI 1994). Exploratory data analysis was performed on location data
174
and environmental variables. Bathymetry was obtained from gridded ETOPO‐2 data,
175
2‐min by 2‐min cells in the study area (National Geophysical Data Center, ETOPO‐2
176
Bathymetry/Topographic Data). Eight‐day composite sea surface temperature data
177
from MODIS‐Terra satellite at 4 km resolution were obtained from an internet site
178
(http://podaac.jpl.nasa.gov/sst), maintained by the NASA Physical Oceanography
179
Distributed Active Archive Center (PO.DAAC), at the Jet Propulsion Laboratory,
180
Pasadena, CA, March, 2003. Only the best position per day as determined by the
10 181
Argos_Filter output for each dolphin was included for analysis with environmental
182
variables to ensure independence (Swihart and Slade 1985). A Chi‐square goodness of
183
fit test (Zar 1999) was used to test the hypothesis that dolphins used certain depth
184
ranges and percent slopes within the study site in proportion to availability. Water
185
depths were assigned to 1 of 7 categories (0 m to ‐534 m, ‐535 m to ‐1085 m, ‐1086 m to ‐
186
1635 m, ‐1636 m to ‐2186 m, ‐ 2187 m to ‐2736 m, ‐2737 m to ‐3287 m, and ‐3288 m to ‐
187
4939 m). Percent slope was categorized into 9 categories as determined by Spatial
188
Analyst extension in ArcView (21.2%, 23.9%, 27.3%, 31.8%, 38.2%, 47.7%, 63.6%, 95.3%,
189
and 109%) Minimum convex polygons (MCP) calculated by Animal Movement were
190
used to delineate home ranges for determining habitat availability (Hooge 1999).
191
Dive analysis.–Dive data output was imported into Excel v. 2000 (Microsoft, Seattle,
192
WA) and Systat v. 10 (SYSTAT Software Inc., Chicago, IL) for statistical analysis.
193
Standard descriptive statistics were used to describe dive behavior. Means are
194
expressed with standard deviations. To test for the differences between four time‐
195
periods (i.e. night, dawn, day, and dusk) for depth of dives and duration of dives,
196
analysis of variance and Tukey post‐hoc comparisons were performed for data
197
exhibiting normal distributions (Zar 1999). If data were non‐normally distributed,
198
Kruskal‐Wallis analysis of variance and post‐hoc, non‐parametric multiple comparisons
199
were employed (Zar 1999).
200
11 201
RESULTS
202
203
Three dolphins (1 female, 2 males) were tagged and released with SLTDRs and VHF
204
tags in June 2003 (Table 1). Positional data were obtained from tagged dolphins for
205
periods ranging from 5 to 45 days. The mean number of positions per day per dolphin
206
for all location classes ranged from 7.6±2.0 to 8.2±1.9 and a mean daily distance of 28.3
207
km/day. After running all received locations for all three dolphins through the
208
Argos_Filter, a total of 293 plausible locations accounted for 75.1% of all position
209
estimates.
210
Movements.–The three tagged dolphins did not show considerable variability in their
211
movement patterns. Dolphin 40000 traveled in a clockwise rotation, starting in the
212
southwest corner, around the island during 4 of the 7 days it transmitted with a mean
213
daily distance of 33.7±13.0 km and a mean rate of 1.5±0.6 km/hr. Due to the 8‐hr duty
214
cycle, it is not clear if the dolphin completed a circle around the island or reversed
215
direction (Fig. 2a). Once released, dolphin 39999 remained near Challenger Bank over
216
the next day and then moved to the northeastern edge of the island for the remaining 4
217
tracking days (Fig. 2b). It traveled a mean daily distance of 23.2±19.1 km and a mean
218
rate of travel of 1.0±0.8 km/hr. The third dolphin, 40001, transmitted plausible locations
219
for 40 days over a 45 day tracking period, utilizing most of the area around the
220
Bermuda Pedestal. Twenty‐six of 40 (65%) of the best daily locations received from
12 221
dolphin 40001 occurred in the southwest corner of the study area on or near Challenger
222
and Argus Banks (Fig. 2c). Its maximum distance traveled was 98.4 km on day 25 when
223
it traveled northeast of Bermuda to Bowditch Seamount, but traveled a mean daily
224
distance of 28.1±21.8 km and a mean rate of travel of 1.1±0.9 km/hr.
225
Environmental variables.–The mean depth of all locations for all dolphins was ‐
226
1402.0±1120.7 m and no significant differences in means were found between the three
227
dolphins (F=1.13, d.f.=2, P=0.324). Chi‐square analyses for dolphin 40001 showed a
228
significant difference between the observed and expected depth categories within the
229
study area (χ2 =139.2, d.f.=6, P<0.001) and the MCP (χ2 =16.6, d.f.=6, 0.025
230
These results illustrate a preference for water depths between 0 to ‐2186 m and a lower
231
than expected use for water depth greater than ‐2187 m. Chi‐square analyses showed
232
that the observed percent slope use by dolphin 40001 did not differ significantly from
233
that expected based on available habitat within the MCP home range (χ2 =10, d.f.=8,
234
P>0.05). The weekly composite sea surface temperature rose steadily throughout the
235
study period from 25° C to 28° C with minimal variation within the study area. Weekly
236
movements for dolphin 40001 were not affected by sea surface temperature.
237
Dive behavior.–A total of 396 six‐hour periods of dive information were received from
238
dolphin 40001. Dolphin 40000 transmitted 36 six‐hour periods of dive information,
239
while dolphin 39999 transmitted 13 six‐hour periods of dive data. Due to the small
13 240
sample sizes of six‐hour periods transmitted by dolphins 39999 and 40000, statistical
241
analyses were only performed on data received by dolphin 40001.
242
Depths of dives for dolphin 40001 were recorded for 804 hours (134 six‐hour
243
periods). Of the total mean number of daily dives (208.1± 83.7), the greatest number of
244
dives (about 35% of the total) occurred during dusk (72.4±19.6) followed by night
245
(51.1±13.1, 25%), day (42.9±29.5, 21%) and dawn (41.7±21.5, 20%).
246
As expected, a majority of the dives made by all three dolphins did not exceed 50 m.
247
About 75% of all dives made by dolphin 40001 and 39999, and almost 85% of dives by
248
dolphin 40000 were made within 50 m of the surface (Table 2). All three dolphins were
249
also found to make repeated dives to depths beyond 250 m. Dolphin 40000 made a total
250
of 116 dives (10.5% of total) beyond 250 m during 90 hours, dolphin 39999 logged a total
251
of 31 dives (8.4% of total) beyond 250 m over 30 hours, and dolphin 40001 recorded
252
1137 dives to depths greater than 250 meters over 810 hours (Table 2).
253
Most dives made by dolphin 40001 to depths greater than 150 m occurred during the
254
period of night (Fig. 3). Of these dives, almost 17% were made to depths greater than
255
450 m (8.49±3.88). In addition, dives made to mid‐range depths between 100 m and 350
256
m accounted for approximately 25% of the total nightly dives (Fig. 3). During day,
257
dives were made to considerably shallower depths than those made during the other
258
three periods (Fig. 3).
14 259
Post‐hoc comparisons for dive depths by dolphin 40001 indicate significant
260
differences between the mean number of dives made at night compared to other time
261
periods for the following depths: 150 m (F=20.95, p<0.0001), 200 m (KW=38.29,
262
p<0.0001), 300 m (F=16.45, p<0.0001), and 350 m (F=30.49, p<0.0001). A significant
263
difference was found between all time periods for the mean number of dives to depths
264
greater than 450 m (F=84.10, p<0.0001) and for all depths beyond 350 m, the mean
265
number of dives during the period of day was significantly lower than all other periods
266
(Fig. 3). Some of the maximum daily dive depths appeared to exceed the limits of our
267
recorders. A total of 34 maximum daily dive depths were recorded for dolphin 40001,
268
with approximately 87% recorded to 492 m and the remaining 13% measured to 488 m.
269
Dolphin 40000 recorded four maximum daily dives depth, 492 m, 492 m, 488 m, and 452
270
m. No maximum daily dive depths were received from dolphin 39999. As mentioned
271
earlier, 492 m is believed to be a default recording since the time‐depth recorders used
272
in this project were only capable of measuring depth to 500 m. Thus it is possible that at
273
least two of the three dolphins were likely diving to depths in excess of 500 m.
274
A total of 792 hours (132 six‐hour periods) was recorded for time spent at depth for
275
dolphin 40001 with about 77% of the time spent within 50 m of the surface and over
276
11% spent beyond 250 m (Table 2). During 48 recorded hours of time spent at depth by
277
dolphin 40000, about 80% of the time was spent within 50 m of the surface and 7%
15 278
below 250 m, while dolphin 39999 spent over 82% of the time above 50 m and almost
279
6% below 250 m during 30 recorded hours (Table 2).
280
During day, dolphin 40001 spent more than twice as much time near the surface
281
compared to night (Fig. 4). While during night, over 40% of time was spent at depths
282
greater than 150 m and less than 2% was spent during day at those depths. Almost 9%
283
of the time was spent between 100 m and 150 m and beyond 450 m during the period of
284
night (Fig. 4). Although approximately 18% of the time during both dawn and dusk was
285
spent at depths below 150 m, a slight difference in dive behavior was observed. Below
286
150 m, the time spent during dusk was spread equally among the different depths,
287
while a greater amount of time during dawn was spent at depths beyond 450 m
288
compared to the other depths below 150 m (Fig. 4).
289
While a majority of dive durations for all three dolphins did not exceed 1 min, a
290
good proportion of dives for all dolphins surpassed 5 min (Table 3). For dolphin 40001,
291
less than half of all the dives durations during day lasted less than 1 min, while almost
292
half of the night dive durations surpassed 5 min (Fig. 6). About 23% of dive durations
293
during dawn and 14% during dusk lasted longer than 5 min (Fig. 6). Of the 6819 dive
294
durations recorded for dolphin 40001, the greatest number of dives per period were
295
made during dusk (70.9±22.2), followed by night (49.13±12.4), day (48.8±31.9), and dawn
296
(42.1±19.5).
297
16 298
DISCUSSION
299
300 301
The 3 dolphins tagged in this study were confirmed to be the offshore ecotype based on their high hematocrit values (ranging between 51% ‐ 59%) and two
302
electrophoretically distinct hemoglobin profiles. Hematocrit values for offshore
303
bottlenose dolphins have been observed in a range between 47% ‐ 56%, while coastal
304
animals have values ranging between 37% ‐ 47% (Hersh et al. 1983). During the 45 days
305
of this study, the dolphin locations and movements were found in close association
306
with the steep edges of the Bermuda Pedestal. These findings correspond with
307
previously recorded sightings by fishermen and help to expand our understanding of
308
these dolphins’ movements around the edge of the Pedestal.
309
Distance and rate of travel.–The mean daily distances traveled in this study (28.3
310
km/day) are similar to previous studies of bottlenose dolphin movements. In Tampa
311
Bay, Florida, Mate et al. (1985) reported an average traveling distance of 23.7 km/day for
312
a female bottlenose dolphin tracked for 25 days, while Tanaka (1987) found a distance
313
traveled of 33.6 km/day for a bottlenose dolphin tracked for 18 days off the coast of
314
Japan. Estrada and Hohn (2003) observed a mean daily distance of 32.7 km/day for
315
migrating bottlenose dolphins along the east coast of the U.S. compared to a mean daily
316
distance of 24.5 km/day for non‐migrating dolphins. However, Wells et al. (1999)
317
reported mean daily distances of 89 km/day and 48 km/day for two rehabilitated and
Comment [Dolphin2]: These details should be given in the results section, rather than in the Discussion section.
17 318
released offshore bottlenose dolphin that were tracked for 47 and 43 days, respectively,
319
off the southeastern coast of the U.S.
320
The rate of travel calculated from the straight line distance between the best daily
321
locations and the average swim speeds for all three dolphins ranged between 1.0±0.8 to
322
1.5±0.6 km/hr, which is consistent with those reported in previous studies of satellite‐
323
monitored bottlenose dolphins. Tanaka (1987) found average swim speeds between 0
324
and 7.4 km/hr for 11 bottlenose dolphins tracked off of Japan over 18 days, while Mate
325
et. al (1995) observed the average swim speed for a bottlenose dolphin in Tampa Bay,
326
Florida, to be 1.2 km/hr over 25 days, with a fastest mean swim speed of 4.9 km/hr.
327
Wells et al. (1999) tracked one rehabilitated bottlenose dolphin for 47 days off the east
328
coast of the United States and north of the Antilles and reported average swim speeds
329
of 1.8 – 4.5 km/hr. A second rehabilitated dolphin tracked by Wells et al. (1999)
330
exhibited an average swim speed of approximately 2.0 km/hr over 43 days. Estrada and
331
Hohn (2003) tracked 16 bottlenose dolphins off the east coast of the United States from 5
332
to 263 days and reported a difference in average swim speeds between non‐migratory
333
and migratory movements. A slightly faster average swim speed (1.27 km/hr) was
334
recorded during migratory movements between New Jersey and North Carolina
335
compared to 1.0 km/hr during non‐migratory movements (Estrada and Hohn 2003).
336
The daily mean rate of travel between all plausible successive locations during each
337
of the 8‐hour duty cycles, ranged from 7.8±1.4 to 8.6±3.0 km/hr for the three dolphins in
18 338
this study. These swim speeds are faster than the minimum mean rate of travel
339
between the best daily locations. This may be due to the dolphins’ preference for
340
specific locations around the plateau and the time elapsed between locations. If
341
locations were received 20 km from each other after 24 hours had elapsed, the rate of
342
travel would be remarkably slower than a 20 km distance between locations over 4
343
hours had elapsed. For example, 65% of the best daily locations for dolphin 40001 were
344
reported on or around Challenger and Argus Banks, an area approximately 390 km2.
345
During the beginning of dolphin 40000’s duty cycle (0700 AST), over 57% of all the
346
plausible locations were located in a 90 km2 area and followed a similar movement
347
pattern clockwise towards the north of the island throughout its 8‐hour duty cycle. A
348
mean number of 8 plausible successive locations were observed for each dolphin during
349
their 8‐hour duty cycles, making it possible that the faster mean swim speeds between
350
successive locations could be a better estimate of true speed than the mean travel speeds
351
found between only the best daily locations.
352
Bathymetric relationship.–A preference for water depths within 2186 m over and
353
around the Bermuda Pedestal was observed with a lower than expected use of water
354
depths greater than 2187 m. It is believed that seamounts are areas of high productivity
355
due to upwelling around the seamount, for example the formation of eddies called
356
Taylor columns which are closed circulations that can persist for several weeks (Rogers
357
1994). The increase in productivity can result in high levels of both plankton biomass
Comment [Dolphin3]: Awkward sentence.
Comment [Dolphin4]: Refresh for the reader why these values were lower than expected. Why were deeper waters expected?
19 358
and their predators in these areas, such as the well‐documented concentration of
359
commercially valuable fish species. Lanternfish, shrimps, and squid have been reported
360
feeding above the seamounts on vertically migrating plankton at night, but are found in
361
deeper water along the sides of the seamount during the day (Rogers 1994). These
362
species, particularly lanternfish, are believed to be an important food source of
363
bottlenose dolphins found in deep water (Mead and Potter 1995). It is possible that the
364
movements reported in this study may be related to the presence of vertically migrating
365
prey species along side and on top of the Bermuda Pedestal. For example, it is possible
366
that some of the locations reported on top of the banks in shallower water could be
367
associated with the presence of prey that are swept over the banks by prevailing
368
currents during their normal diurnal vertical migration and become trapped and are
369
unable to complete their downward migration as light levels increase (Rogers 1994).
370
Dive behavior.–With 7.9% of all dives to depths beyond 450 m and a maximum daily
371
dive depth of 492 m for dolphin 40001, it appears that bottlenose dolphins in offshore
372
waters near the Bermuda Pedestal are capable of making regular dives to deep depths
373
not previously recorded for free‐ranging bottlenose dolphins. Although fewer data
374
were received for dolphins 39999 and 40000, they also exhibited similar results for dive
375
durations and maximum dive depth. Along with the high hematocrit values (51 – 59%)
376
measured in the three dolphins during this study, these results support Hersh and
20 377
Duffield’s (1990) hypothesis that offshore dolphins may be capable of diving deeply
378
due to their increased oxygen‐binding capacity.
379
The maximum reported dive depths believed to be greater than 492 m for this study
380
are congruent with dive depth maximums previously recorded by trained free‐
381
swimming bottlenose dolphins. From unpublished data, Ridgway (1985) reported the
382
maximum‐recorded dive depth of 390 m with duration of 7.5 min by a trained coastal
383
Atlantic bottlenose dolphin, as well as an account of the maximum dive depth of a
384
trained Pacific bottlenose dolphin to 535 m lasting 8 min.
385
The results of this study also imply a pattern of diel dive behavior by dolphin 40001
386
over a 45‐day tracking period. The most frequent dives to depths greater than 450 m
387
occurred during night and to a smaller extent during dawn and dusk (Figure 8). This
388
relationship was supported by the time at depth and the duration of dives during these
389
periods (Figures 9 and 10). It is possible that the large mean number of total dives,
390
particularly to depths greater than 100 m during dusk, may reflect an increase in
391
“exploratory” dives in search of prey associated with the ascending, deep scattering
392
layer during its nightly migration into shallower waters. Studies by Gibbs and Roper
393
(1970) and Karnella (1987) have found a 12‐kHz scattering layer in Bermuda waters
394
during all seasons between 400 m and 600 m, with a less prominent layer between 0 m
395
and 150 m during the day. Shortly before sunset, vertically migrating species rise
21 396
within the upper 250 m and return back down to depth shortly after sunrise (Gibbs and
397
Roper 1970; Karnella 1987).
398
Dive depth data revealed consistent nighttime dives made to depths between 100 m
399
and 350 m, which may indicate nocturnal foraging at these depths. Little is known
400
about the diet of offshore dolphins in Bermuda waters, but stomach contents of
401
stranded offshore dolphins along the East Coast of the Unites States consist of
402
mesopelagic fish, primarily lanternfish and pelagic squid (Barros and Odell 1990; Mead
403
and Potter 1995). Several species of myctophids and cephalopods in Bermuda waters
404
make nightly vertical migrations associated with the deep scattering layer. Gibbs and
405
Roper (1970) discovered larger increases in nighttime catch per unit effort of lanternfish
406
compared to daytime catches at depths 50 m – 200 m and 350 m ‐500 m. During late
407
summer, Karnella (1987) found Bolinichthys indicus, Lampanyctus photonotus, and
408
Lampanyctus festivus exhibit the greatest abundance between 101 m ‐115 m, whereas,
409
Diaphus mollis, L. guentheri, and H. benoiti are most abundant down to 160 m. During the
410
day, myctophids exhibit a center of abundance from approximately 450 m ‐ 900 m, with
411
the most abundant species group found between 601 m ‐ 700 m during all seasons
412
(Gibbs and Roper 1970; Karnella 1987). One of the most abundant and commonly
413
caught squid in mid‐water trawls in Bermuda is Pyroteuthis margaritifera, which can be
414
found moving with the deep scattering layer from 300 m ‐ 500 m during the day to
415
between 100 m ‐ 200 m at night (Gibbs and Roper 1970). It is possible that the diurnal
22 416
dive behavior exhibited by the dolphin in this study is associated with the nightly
417
vertical migrations of potential prey species.
418
Nocturnal foraging on mesopelagic prey associated with the deep scattering layer
419
has been observed in several other delphinid species, such as pelagic spinner dolphins
420
(Stenella longirostris) in the eastern and western Pacific (Fitch and Brownell 1968; Perrin
421
et al. 1973; Perrin and Gilpatrick 1994), and common dolphins (Delphinus delphis) in the
422
eastern Pacific (Evans 1974). Based on radio‐tracking data and stomach content
423
analyses, Norris et al. (1994) reported that during the night, Hawaiian spinner dolphins
424
(Stenella longirostris) feed on the Hawaiian mesopelagic boundary community in deep
425
offshore waters (Norris and Dohl 1980). Recently, Benoit‐Bird et al. (2001) and Benoit‐
426
Bird and Au (2003) described an association between the Hawaiian spinner dolphins’
427
movements and the nightly vertical and horizontal movements of the Hawaiian
428
mesopelagic boundary community moving from offshore towards the shoreline.
429
Prior to the present study, our knowledge of the habitat use of bottlenose dolphins
430
in Bermuda waters in the summer was limited to occasional reports by fisherman and
431
other local residents of Bermuda. Although the location and dive data are not
432
temporally associated, meaning the dive data received during a transmission were
433
recorded during previous periods and not the current 6‐hour period, Tag 40001 was
434
programmed to transmit locations that occurred during the last two hours of the dusk
435
period and all six hours of the night period. Therefore, location estimates received from
23 436
this tag provide an idea of where the majority of deep dives were performed in the
437
study area.
438
Due to the small sample size, it is difficult to evaluate dolphin habitat use in this
439
area, but this pilot study does suggest that offshore dolphins prefer the steeply‐sided
440
banks to the southwest of the island, as well the northeastern edge of the island itself,
441
although there were locations received on all sides of the island. Additional movement
442
information is essential for determining if these dolphins should be managed as an
443
island‐associated population or part of a larger oceanic population. Clearly, more dive
444
data are necessary to confirm these findings and to further define the distribution and
445
diving capabilities of offshore bottlenose dolphins found in deep waters of the
446
northwest Atlantic Ocean and in close proximity to the Bermuda Pedestal. However,
447
the dive data presented here indicate the regular deep diving abilities of these dolphins,
448
and extends our understanding of this species’ diving capabilities and adaptations to a
449
wider variety of habitats than previously known.
450
451
Acknowledgments.–This study was funded and supported by Dolphin Quest and Quest
452
Global Management and would not have been possible with out the efforts of B.
453
Hurley, B. Balmer, J. Davidson, A. Allen and the unwavering support and talented
454
personnel of Dolphin Quest Bermuda. We are indebted to Captain S. Masters, the crew
455
of the Lady Gambler, and all of the Bermuda fishermen for their help and support
24 456
locating the dolphins. Thank you to R. Hayward for performing blood hematology and
457
D. Duffield for analyzing the hematological profiles. Thanks to M. Alexander, L.
458
McClenaghan, J. Franklin and D. Stow for their reviews and comments of this work.
459
Additional support for this project was provided by the Chicago Zoological Society and
460
the Bermuda Zoological Society. This research was performed under Special Permit
461
number 002/03 from the Bermuda Department of Environmental Protection and San
462
Diego State University APF# 03‐08‐027M.
463
464
465
LITERATURE CITED
466
Argos. 2003. User’s manual. www.argosinc.com, Service Argos, Landover, MD.
467
Asper, E.D. 1975. Techniques of live capture of smaller Cetacea. J. Fish. Res. Bd. Canada.
468
32:1191‐1196.
469
Barros, N.B. and D.K. Odell. 1990. Food habits of bottlenose dolphins in the
470
southeastern United States. Pp. 309‐328 in The Bottlenose Dolphin (S.
471
Leatherwood and R. Reeves, eds.). Academic Press, San Diego, CA.
472
Benoit‐Bird, K.J., W.W.L. Au., R.E. Brainard, and M.O. Lammers. 2001. Diel horizontal
473
migrations of the Hawaiian mesopelagic boundary community. Marine Ecology
474
Progress Series 217:1‐14.
25 475
Benoit‐Bird, K.J. and W.W.L. Au. 2003. Prey dynamics affect foraging by a pelagic
476
predator (Stenella longirostris) over a range of spatial and temporal scales.
477
Behavioral Ecological Sociobiology. 53:364‐373.
478
Curry, B.E., and J. Smith. 1997. Phylogeographic structure of the bottlenose dolphin
479
(Tursiops truncatus): stock identification and implications for management. Pp.
480
227‐247 in Molecular genetics of marine mammals (A. E. Dizon, S. J. Chivers, and
481
W. F. Perrin, eds.). Society for Marine Mammology.
482 483 484
Douglas, D. 2003. PC‐SAS Argos_Filter v6.2 Documentation. USGS Alaska Science Center, Juneau, AK. Dowling, T.E. and W.M. Brown. 1993. Population structure of the bottlenose dolphin
485
(Tursiops truncatus) as determined by restriction endonuclease analysis of
486
mitochondrial DNA. Marine Mammal Science 9:138‐155.
487
Duffield, D.A., S.H. Ridgway, and L.H. Cornell. 1983. Hematology distinguishes coastal
488
and offshore forms of dolphins (Tursiops). Canadian Journal of Zoology 61:930‐
489
933.
490 491
ESRI (1994) Using Arcview GIS. Environmental Systems Research Institute, Redlands, CA.
26 492
Estrada, E. L. and A. A. Hohn. 2003. Satellite monitored movements of bottlenose
493
dolphins along the Atlantic coast of the US. Abstract. 15th Biennial Conference on
494
the Biology of Marine Mammals, Greensboro, NC, December 14‐19, 2003.
495
Evans, W.E. 1974. Radio‐telemetric studies of two species of small odontocete cetaceans.
496
Pp. 385‐394 in The Whale Problem (W.E. Schevill, ed.). Harvard University Press,
497
Cambridge, MA.
498
Fielder, P. C. 2002. Ocean Environment. Pp. 824‐830 in Encyclopedia of Marine
499
Mammals (W. H. Perrin, B. Würsig, and J. G. M. Thewissen, eds.). Academic
500
Press, San Diego, CA.
501 502 503
Fitch, J.E. and R.L. Brownell, Jr. 1968. Fish otoliths and their importance in interpreting feeding habitats. Journal Fishery Research Bd. Canada, 25:2561‐2574. Gibbs, R.H., Jr. and C.F.E. Roper. 1970. Ocean Acre preliminary report on vertical
504
distribution of fishes and cephalopods. Pp. 119‐133 in Proceedings of an
505
international symposium on biological sound scattering in the ocean (G. B.
506
Farquhar, ed). Maury Center Report 005, Department of Navy, Washington D.C.
507
Gregr, E.J. and A.W. Trites. 2001. Predictions of critical habitat for five whale species in
508
the waters of coastal British Columbia. Canadian Journal of Fisheries and
509
Aquatic Science 58: 1265‐1285
27 510
Hersh, S.L. and D.A. Duffield. 1990. Distinguishing between northwest Atlantic
511
offshore and coastal bottlenose dolphins based on hemoglobin profile and
512
morphometry. Pp. 129‐139 in The Bottlenose Dolphin (S. Leatherwood and R.
513
Reeves, eds.). Academic Press, San Diego, CA.
514
Hoelzel, A. R., C. W. Potter, and P. B. Best. 1998. Genetic differentiation between
515
parapatric ‘nearshore’ and ‘offshore’ populations of bottlenose dolphin.
516
Proceedings of the Royal Society of London B 265: 1117‐1183.
517 518 519
Hooge, P.N. 1999. Animal Movement analysis ArcView extension. Alaska: Biological Science Centre, USGS‐BRD. Hooker, S. K., H. Whitehead, and S. Gowans. 1999. Marine protected area design and
520
the spatial and temporal distribution of cetaceans in a submarine canyon.
521
Conservation Biology 13(3):592‐602.
522
Irvine, A. B., M.D. Scott, R.S. Wells, J.H. Kaufmann. 1981. Movements and activities of
523
the Atlantic bottlenose dolphin, Tursiops truncatus, near Sarasota, Florida. Fishery
524
Bulletin 79:671‐688.
525
Karnella, C. 1987. Family Myctohidae, lanternfishes. Pp. 51‐168 in Biology of midwater
526
fishes of the Bermuda Ocean Acre (R.H. Gibbs, Jr. and W.H. Kreuger, eds.).
527
Smithsonian Contributions to Zoology 452.
28 528
Kenney, R.D. 1990. Bottlenose dolphins off the northeastern United States. Pp. 369‐386
529
in The Bottlenose Dolphin (S. Leatherwood and R. Reeves, eds.). Academic Press,
530
San Diego, CA.
531
Leatherwood, S. and R. Reeves. 1982. Bottlenose dolphins (Tursiops truncatus) and other
532
toothed cetaceans. Pp. 369‐414 in Wild Mammals of North America (J. Chapman
533
and G.A. Feldhamer, eds.). Johns Hopkins Univ. Press, Baltimore, MD.
534
Mate, B.R., K.A. Rossbach, S.L. Nieukirk, R.S. Wells, A.B. Blair, M.D. Scott, and A.J.
535
Read. 1995. Satellite‐monitored movements and dive behavior of a bottlenose
536
dolphin (Tursiops truncatus) in Tampa Bay, Florida. Marine Mammal Science,
537
11(4):452‐463.
538
Mead, J.G. and C. W. Potter. 1995. Recognizing two populations of the bottlenose
539
dolphin (Tursiops truncatus) off the Atlantic coast of North America:
540
morphological and ecological considerations. IBI Reports 5:31‐44.
541 542 543 544
Norris, K.S., B. Würsig, R.S. Wells, and M. Würsig. 1994. The Hawaiian spinner dolphin. University of California Press, Berkeley, CA. 408 pp. Norris, K.S. and T. P. Dohl. 1980. Behavior of Hawaiian spinner dolphin, Stenella longirostris. Fishery Bulletin. 77:821‐849.
29 545
Perrin, W.F., R.R. Warner, C.H. Fiscus, and D.B. Holts. 1973. Stomach contents of
546
porpoise, Stenella sp., and yellowfin tuna, Thunnus albacares, in mixed‐species
547
aggregations. Fishery Bulletin. Washington D.C., 71:1077‐1092.
548
Perrin, W. F. and J. W. Gilpatrick, Jr. 1994. Spinner dolphin Stenella longirostris (Gray,
549
1828). Pp. 99‐128 in Handbook of marine mammals, Vol. 5: The first book of
550
dolphins (S. H. Ridgway and R. Harrison, eds.). Academic Press, San Diego, CA.
551 552 553 554 555
Reynolds, J.E. III, R.S. Wells, and S.D. Eide. 2000. The Bottlenose Dolphin: Biology and Conservation. University Press of Florida. 289 pp. Rice, D.W. 1998. Marine Mammals of the World: Systematics and Distribution. Special Publication No. 4, Society for Marine Mammalogy, Allen Press, Lawrence, KS. Ridgway, S.H. 1985. Diving by Cetaceans. Pp. 33‐62 in Diving in Animals and Man: An
556
International Symposium, Kongsvoll. (Brubakk, A.O., J.W. Kanwisher, and G.
557
Sundnes eds.). The Royal Norwegian Society of Science and Letters, Trodheim,
558
Norway.
559
Rogers, A.D. 1994. The biology of seamounts. Advances in Marine Biology 30:305‐350.
560
Swihart, R.K. and N.A. Slade. 1985. Influence of sampling interval on estimates of
561 562 563
home‐range size. Journal of Wildlife Management 49:1019‐1025. Tanaka, S. 1987. Satellite radio tracking of bottlenose dolphins Tursiops truncatus. Nippon Suisan Gakkaishi 53:1327‐1338.
30 564
Waring, G.T., T. Hamazaki, D. Sheeman, G. Wood, and S. Baker. 2001. Characterization
565
of beaked whale and sperm whale summer habitat in shelf‐edge and deeper
566
waters off the northeast U.S. Marine Mammal Science 17(4): 703‐717.
567
Wells, R.S., H.L. Rhinehart, P. Cunningham, J. Whaley, M. Baran, C. Koberna, and D.
568
Costa. 1999. Long distance offshore movements of bottlenose dolphins. Marine
569
Mammal Science 15:1098‐1114.
570
Wells, R.S. and M. D. Scott. 1999. Bottlenose dolphin, Tursiops truncatus, (Montagu,
571
1821). Pp. 137‐182 in The Handbook of Marine Mammals Vol. 6: The Second
572
Book of the Dolphins and the Porpoises (Ridgway, S.H. & Harrison, S.R. Eds.).
573
Academic Press, San Diego:CA.
574 575 576
Würsig, B. and M. Würsig. 1979. Behavior and ecology of the bottlenose dolphin, Tursiops truncatus. Science 198:755‐756. Zar, J.H. 1999. Biostatistical Analysis. 4th ed. Prentice Hall, Saddlebrook, NJ.
31 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
Figure legends Fig. 1.–Study Area with 200 m contours Fig. 2.–Dolphin movements a) dolphin 40000, b) dolphin 39999, c) dolphin 40001 (200 m depth contours illustrated) Fig. 3.–Mean percentage of depth of dives recorded in six‐hour periods for dolphin 40001. Day, n=29; dusk, n=33; night, n=35; dawn, n=37 (n=number of 6‐hour time periods). Fig. 4.–Mean percentage of time spent at depth recorded in six‐hour periods for dolphin 40001. Day, n=30; dusk, n=30; night, n=39; dawn, n=33 (n=number of 6‐hour time periods). Fig. 5.–Mean percentage of dive durations recorded in six‐hour periods for dolphin 40001. Day, n=32; dusk, n=34; night, n=32; dawn, n=32 (n=number of 6‐hour time periods).