DIVE BEHAVIOR OF A HARBOR SEAL (PHOCA VITULINA RICHARDII) IN THE PRESENCE OF TRANSIENT KILLER WHALES (ORCINUS ORCA) IN GLACIER BAY NATIONAL PARK, ALASKA

Predation can shape virtually every aspect of an organism's behavior, life history, and evolution (e.g., Dill 1987; Godin 1995; Reznick et al. 1996). Thus, understanding how organisms respond to perceived predation risk or the actual presence of predators is central to understanding their foraging decisions, habitat use, reproductive effort, and other proximate behaviors (e.g., Sih 1982; Ydenberg and Dill 1986; Gillam and Fraser 1987; Heithaus and Frid 2003). For harbor seals (Phoca vitulina richardii), predation (or the risk of predation) is thought to be common enough to influence habitat utilization (Calambokidis et al. 1987; Nordstrom 2002).

Harbor seals constitute one of the primary prey species of transient killer whales (Orcinus orca) in the North Pacific (Ford et al. 1998; Saulitis et al. 2000; Heise et al. 2003) and must balance foraging decisions with exposure to killer whales (Frid et al. 2006). There are numerous records of killer whales attacking and killing harbor seals; however, most observations have been analyzed in terms of the predatory behavior of killer whales (Baird and Dill 1996; Ford et al. 1998; Saulitis et al. 2000), with a few exceptions (Deecke et al. 2002). Herein we present data on dive and antipredator behavior by a harbor seal in the presence of transient killer whales. Our observation was unique in that the harbor seal was instrumented with both a VHF headmount transmitter (MM340, Advanced Telemetry Systems, Inc., Isanti, MN) and a time-depth recorder (TDR) (Mk9, Wildlife Computers, Redmond, WA) thus, providing us with dive and location data before, during, and after the harbor seal's encounter with the transient killer whales.

As part of a study examining the foraging ecology of harbor seals in Glacier Bay National Park (GBNP), we observed apparent antipredator behavior by a radio-tagged harbor seal when transient killer whales arrived in the area near the seal. GBNP has historically supported one of the largest breeding populations of harbor seals in Alaska (Calambokidis et al. 1987; Hoover-Miller 1994; Mathews 1995); however, from 1992 to 2002, the number of harbor seals declined by up to 75%, with predation identified as a possible factor contributing to the decline (Mathews and Pendleton 2006). Predators of harbor seals in GBNP include killer whales (Calambokidis et al. 1987), Steller sea lions (Eumetopias jubatus) (Mathews and Pendleton 2006; J. N. Womble, NPS unpublished data and possibly Pacific sleeper sharks (Somniosus pacificus) (Taggart et al. 2005).

On April 11, 2004, a subadult female harbor seal (45.1 kg) was captured at Kidney Stone Reef (58.528890N, 135.876429W), a terrestrial haulout site in the Beardslee Islands in lower GBNP. The seal was fitted with a VHF headmount transmitter and a TDR programmed to record depth, time, and temperature every 2 s. The resolution of the TDR was 0.5 m with an accuracy of ±1 m. The TDR was recovered during a boat survey in late July during the molt, and the data were downloaded. Hexadecimal files were displayed graphically and then corrected for changes in the pressure transducer calibration using the Zero Offset Correction (ZOC) program.

On June 11, 2004 at approximately 1355, we observed the harbor seal hauled out with approximately 40 other harbor seals on Kidney Stone Reef south of Kidney Island in the Beardslee Islands. At that time no killer whales were seen in the area. Later the same day (∼1517), we were slowly motoring near the south tip of Kidney Island on the research vessel, R/V Capelin, when we observed three killer whales traveling south along the western shore of Kidney Island, approximately 1 km away and heading toward our vessel. We turned off the engine to observe the killer whales. The killer whales, which were following the shoreline contour, maintained consistent dive/surface intervals and swam in a tight group. Approximately 1 min later (at 1518) we observed a harbor seal surface ∼20 m from our vessel. The seal was located mid-channel (depth ∼62 m) just south of Kidney Island (58.51319N, 135.876592W). We saw a blue VHF headmount (each seal had a unique color combination of radio tags) on the harbor seal and immediately began tracking its VHF signal to record dive duration. As we were not scanning for VHF signals when we first saw the killer whales, this was our first observation of the seal since locating it on Kidney Stone Reef ∼84 min earlier.

The group of killer whales accelerated, immediately changed course, and traveled directly to the area where the harbor seal was at the surface. At 1521, just as the killer whales reached the area, the harbor seal dove as indicated by the termination of the VHF signal (saltwater blocks radio signals). The killer whales began porpoising on the surface and breathing heavily, repeatedly traversing an area of approximately 0.10 km². For nearly 10 min the killer whales were active on the surface, with two of the killer whales repeatedly fluke slapping, a common method used to debilitate or immobilize prey (Ford and Ellis 1999). One of the killer whales breached on several occasions. The killer whales were photographed and subsequently identified by Dena Matkin (North Gulf Oceanic Society, Gustavus, Alaska) and Graeme Ellis (Pacific Biological Station, Department of Fisheries and Oceans, Nanaimo, British Columbia, Canada) as transients T087 (male), T088 (female), and T097 (male). At approximately 1533, about 12 min after the seal last was seen and VHF signal heard, the killer whales departed the area and began traveling toward the Beardslee Entrance.

Absence of the VHF signal (reception range for VHF signals was 7–9 km) indicated that the seal did not surface when the killer whales were present in the area between 1521 and 1533. Once the killer whales departed the area, the seal resurfaced at 1534 in a near-shore area in shallower water (∼18 m) approximately 0.65 km from the location of her last dive. When the transient killer whales were present in the area, the dive duration (13 min) of the harbor seal was considerably longer than after the killer whales left the area. The first surface interval after the extended dive lasted 2 min after which the seal began regular dive/surface intervals with average dive duration of 1.16 min until 1624 when the observation ended and we vacated the area.

Mid-channel depths where the seal initially submerged were 50–60 m, but TDR data indicate that instead of diving to the bottom, the seal dove to depths up to 3.5 m (Fig. 1). Dive depths before the arrival of the killer whales were up to 1.0 m, up to 3.5 m when the killer whales were present, and up to 0.5 m after they departed (Fig. 1). The greatest dive depth occurred between 1521 and 1527 and coincided with the height of the surface activity by the group of killer whales. Otherwise the harbor seal remained fairly close to the surface. Rather than devoting time and energy to a deep dive, it appears that the primary evasive tactic was to dive below the surface and move out of the immediate area to a near-shore area, effectively distancing herself from the surface activity and aggressive behavior of the killer whales.

On June 9, 2004, 2 d prior to the killer whale encounter the harbor seal was observed diving at about the same time of day (from 1257 to 1605) in the same immediate area where the killer whale encounter occurred with dive depths ranging from 1.5 to 40.0 m. In addition to differences in dive depths, the dive durations were much shorter (= 1.75 min ± SE 0.1; range 0.15–4.25 min) compared to the dive duration when killer whales were present (13 min). The seal was also observed diving on June 21 and June 26, 2004 and on both occasions it dove to at least 40 m. The subadult female harbor seal was subsequently relocated on nine more occasions, in and around the Beardslee Islands, after the interaction with transient killer whales.

We suggest that the activities of the killer whales constituted a potential predation attempt on the harbor seal because (1) the killer whales changed their direction of travel and proceeded directly toward the harbor seal (Ford and Ellis 1999; Saulitis et al. 2000) and (2) the harbor seal dove to depth, swam away from the killer whales, and traveled to a near-shore area. Similar antipredator behaviors have been reported for harbor seals after having been alerted to the presence of transient killer whales (Barrett-Lennard et al. 1996).

The change in behavior of the harbor seal upon the arrival of the transient killer whales indicated that the seal probably detected the killer whales, via visual and/or auditory cues, which then triggered the antipredator response of diving beneath the surface and moving away from the immediate area. Harbor seals can detect sound beyond 30 kHz (Møhl 1968), which is within the range of killer whale sonar clicks (Barrett-Lennard et al. 1996), and are able to respond to the calls of killer whales with antipredator behavior (Deecke et al. 2002). In a near-shore area with a highly convoluted shoreline like the Beardslee Islands, transient killer whales may need to echolocate more frequently to reduce the risk of stranding or collision while consequently increasing their risk of alerting potential prey (Barrett-Lennard et al. 1996).

The dive duration (13 min) of the subadult harbor seal when killer whales were present exceeded the estimated theoretical aerobic dive limit of 8.9 min for an 80-kg female harbor seal (Bowen et al. 1999). Allometrically, the theoretical aerobic dive limit for our 45-kg study animal would be even shorter; therefore, in an apparent evasion of predators, this harbor seal used an energetically more costly anaerobic dive not commonly observed in the majority of dives for harbor seals. For example, in the Gulf of Alaska most (70%–90%) harbor seal dives were 0–4 min in duration and less than 2% of all dives were >8 min in duration (Hastings et al. 2004).

Elucidating trade-offs between foraging and predation risk is important in furthering our understanding of dive behavior, habitat use, and prey selection by harbor seals and other pinnipeds. Real-time tracking studies that integrate movement, dive behavior, other behavioral data, prey availability, and indices of predation pressure provide a more complete understanding of the at-sea behavior of marine mammals.