- Email: [email protected]
How harbor seals (Phoca vitulina) pursue schooling herring
Mammalian Biology 80 (2015) 385–389
Contents lists available at ScienceDirect
Mammalian Biology journal homepage: www.elsevier.com/locate/mambio
Short Communication
How harbor seals (Phoca vitulina) pursue schooling herring Meike Kilian, Guido Dehnhardt, Frederike D. Hanke ∗ University of Rostock, Institute for Biosciences, Sensory and Cognitive Ecology, Albert-Einstein-Str. 3, 18059 Rostock, Germany
a r t i c l e
i n f o
Article history: Received 11 December 2014 Accepted 25 April 2015 Handled by Luca Corlatti Available online 2 May 2015 Keywords: Hunting strategy Visual hunting Prey Foraging Pinniped
a b s t r a c t Only a few reports have described how pinnipeds hunt schooling fish. However, we had the opportunity to systematically observe captive harbor seals hunting a school of herring in a shallow enclosure during daylight. The seals actively pursued the fish, mostly changing from one side of the school to the other, swimming in a supine position close to the water surface. They were often found swimming rapidly down to the fish out of this position. When hunting in the vertical, the seals mostly adopted a dorsal body posture relative to the school. They swam and attacked the school in a supine orientation when approaching from the water surface and swam in a prone orientation when approaching from below. They even maintained this relative body position during turning movements. These phenomena suggest that the seals were constantly keeping the school in their large dorsal visual field, which favors visual hunting during daylight and in clear waters. When interacting with the school, the school mostly split asymmetrically, and the seals were following preferably a smaller number of fish afterwards. Successful prey catch was only observed, when a small group or a single herring had been separated, thus the seals probably avoided the confusion effect of the school. In conclusion, we provide detailed insight in seals pursuing schooling prey, thereby extending previous reports and confirming speculations as well as brief observations. © 2015 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
Harbor seals hunt benthic as well as pelagic fish individually and in schools (Bowen et al., 2002; Marshall, 1998). From the few studies available, we know that, when hunting schooling fish, harbor seals attack the edges of the school most likely in an attempt to separate single fish or small subunits (Bowen et al., 2002; Zamon, 2001). Seals often seem to be engaged in “active pursuit”, during which they constantly keep contact with the school and shift from one side of the school to the other (Bowen et al., 2002). Zamon (Zamon, 2001) document behavioral differences of adults and pups; adult harbor seals ingest a mouthful of fish from the edges of a school whereas pups swim through the school and are seen darting on single fish that were separated from the school. Harbor seals also hunt schooling fish close to the sea floor which seems to constrain fish to escape (Olsen and Bjorge, 1995). The same effect could be achieved by chasing fish towards the water surface as described for separated sand lance (Ammodytes dubius) (Bowen et al., 2002). These few insights in harbor seal foraging behavior were obtained with the help of animal-borne video systems (Bowen and Harrison, 1996; Bowen et al., 2002; Marshall, 1998), developed in
∗ Corresponding author. Tel.: +49 381 666971914; fax: +49 381 666971935. E-mail addresses: meike [email protected] (M. Kilian), [email protected] (G. Dehnhardt), [email protected] (F.D. Hanke).
the 1990s (Davis et al., 1992, 1999; Marshall, 1998), or by observations from vessels and land (Middlemas et al., 2005; Wright et al., 2007; Yurk and Trites, 2000; Zamon, 2001). However, these documents are rare and allow only glimpses on the foraging behavior. This is due to two main challenges associated with studying foraging in marine mammals. One challenge is that seals hunt under water, an environment, which is still relatively inaccessible to humans despite advances in modern technology. The other challenge refers to seals being highly mobile species. To successfully monitor their foraging behavior thus requires techniques that do not impose spatial constraints and preferably allows observing the seal-school-unit in total. In the present study, we had the unique opportunity to characterize the foraging behavior of harbor seals hunting schooling fish in greater detail. We could systematically observe a group of nine captive harbor seals (3–27 years), which are kept for scientific reasons under semi-natural conditions in a large seawater enclosure (3–6 m depth, 60 m in length and 30 m in width) separated from the open sea only by a net (mesh size 5 × 5 cm) at the Marine Science Center, Rostock, Germany. All seals were born in zoos and had only come into contact with living fish such as flatfish, eel and herring after moving into the seawater enclosure in summer 2008. In spring 2009, a school of herring (Clupea harengus) with more than 1000 fish entered the seals’ enclosure and stayed for four weeks from mid-April to May. During this time, the seals were observed
http://dx.doi.org/10.1016/j.mambio.2015.04.004 1616-5047/© 2015 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
386
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389
Table 1 Ethogram used to analyze the video recordings listing all behaviors documented and their definitions as well as the results listing the frequency with which the behavior occurred as number of total events N and in % (within a category).
Behavioral components of prey pursuit Absolute body position Prone Supine Side Relative body position Dorsal Ventral Lateral Turning movements Unchanged Changed Effect of an approaching seal on school Splitting Symmetrical Asymmetrical Following Following smaller subunit Following larger subunit Following a single fish Not following Acceleration Acceleration on small subunit Acceleration in large subunit Acceleration on single fish Acceleration when others accelerated Acceleration with no apparent reason Prey capture
Definition Body position absolute in the water column Back pointing towards the water surface Belly pointing towards the water surface A flipper pointing towards the water surface Body position relative to the school A seal’s back pointing towards the school A seal’s belly pointing towards the school A seal pointing with one side, tail or snout towards the school A change of absolute and/or relative body position during ascent and descent A seal keeping its relative body position during turning A seal changing its relative body position during turning
The two resulting parts of the school were approximately the same size A different number of individuals was estimated in two or more subunits A seal following fish after splitting A seal following the smaller fish subunit after splitting A seal following the larger fish subunit after splitting A seal following a single fish after splitting A seal did not follow fish after splitting A seal was accelerating A seal was accelerating when following a small fish subunit A seal was accelerating when following a large fish subunit A seal was accelerating when following a single fish A seal was accelerating when another seal accelerated A seal was accelerating with no apparent reason A seal was successfully catching a fish
while hunting throughout the day. Two of the seals exclusively fed on herring captured from the school. All other seals occasionally hunted but also participated in the scientific experiments and routine training activity, during which they received fish rewards. The behavior of the hunting seals during daylight hours was recorded from a top view video camcorder (Canon XL1S; Canon Deutschland GmbH, Krefeld, Germany) that captured the seal as well as the school. On the basis of an ethogram (Table 1) that was developed during preparatory observations and that focused on behavioral components of pursuing the fish and on the effect an approaching seal has on the school, we analyzed 75 min of high quality video footage during which six of the altogether nine seals and the school of fish could be observed throughout the entire water column of the enclosure. The depth within the water column, at which a specific behavior occurred, was documented and categorized as “at the water surface” in cases where at least one part of the seal’s body was seen out of the water, “down to approximately 1 m depth” when the seal was below the water surface, and “deeper than approximately 1 m”. The 1 m transition point was estimated on the basis of the height of a seal and took some uncertainty in the determination of the water depth from above into account. In contrast to previous studies, we were able to continuously record six of the nine hunting seals during daylight hours and to systematically document the foraging behavior over a large area with full view on the seals and the school together. Our goal was to generally extend the existing knowledge of harbor seals hunting schooling fish and to specifically describe the behavioral sequence occurring during the pursuit of a school of herring. We were interested in how the seals were positioning themselves absolutely in the water column and relatively to the fish, which might allow assessing the sensory modality the seals were predominantly using. Our hypothesis was that the seals mainly hunt visually in the shallow enclosure in which light penetrates up to the bottom. Furthermore we wanted to test the hypothesis that the school breaks into subunits by an approaching seal and that the seals are predominantly following a smaller number of fish afterwards. In this context, we hypothesized that attacks occur mostly on small
N
In %
609 1066 179
32.8 57.5 9.7
329 41 1484
17.8 2.2 80.0
327 138
70.3 29.7
91 318
19.3 80.7
176 54 29 59
55.0 17.0 9.0 19.0
87 43 22 9 48 0
41.2 20.4 10.4 4.3 22.7 0
subunits and single fishes, which would document the adaptive value of breaking a school into subunits or separating single fish. Our analysis revealed that, during hunting, the six recorded seals were moving through or with the school and only rarely remained stationary. Their absolute body position was significantly correlated with water depth (2 = 11,770.5, df = 8, p < 0.0001, Fig. 1). Overall, the seals were swimming in a “supine” orientation in 57.7% of the time (n = 1066, 2 = 636.7, df = 2, p < 0.0001). However, when 100 90 80
N=303 N=709
70 N=294
frequency (%)
Documented behaviors
60 50 40 30
N=123 N=183
20 10
N=63 N=100
N=42
N=37
0 at surface
< 1m
> 1m
absolute body position depending on depth
Fig. 1. Orientation of the seals in the water. Absolute body positions adopted by the seals in the water column depending on water depth. The absolute body position (“prone”, “supine”, or “side”) was documented depending on the depth (“at the water surface”, “down to approximately 1 m” and “deeper than approximately 1 m”), at which it was performed within the water column. Frequency is plotted as percentage of total events, absolute numbers are indicated at the respective bars. Black bars indicate the frequency of adopting the body position “prone”, gray bars the frequency of the position “supine” and white bars the frequency of the position “side”.
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389
387
N=914
50 45 40
frequency (%)
35 30 N=460
25 20 15 10
N=146
N=126
N=110
5
N=57 N=23
N=12
N=6
0 prone
supine
side
absolute and relative body position to school Fig. 2. Orientation of the seals in the water and relative to the school. The absolute body position of the seal within the water column as well as its orientation relative to the school are documented. Concerning the relative body position the school could be either “ventral” (black bars), “dorsal” (gray bars) or “lateral” (white bars) of the seal. Absolute body positions are as in Fig. 1. Insets give examples of the respective absolute and relative body positions.
the seals were swimming deeper than approximately 1 m within the water column, they changed their absolute body position into a “prone” orientation in 75.2% of the time (n = 303). If the body position of the seals relative to the school was analyzed, it appeared that in 80% of the cases the seals swam slightly “lateral” orientated to the school (n = 1484, 2 = 1887.4, df = 2, p < 0.0001, Table 1). In the majority of cases being “laterally” orientated, the snout was pointing towards the school. When the absolute as well as the relative body position were taken into consideration, the seals swam in 49.3% (n = 914, 2 = 3487, df = 8, p < 0.0001, Fig. 2B) of the cases in a “supine” orientation slightly “lateral” relative to the school. When interacting with the school, the seals swam constantly ascending and descending but always kept contact with the school during these turning movements. In 70.3% of the time, the seals kept their relative body positions unchanged towards the school of herring by changing their absolute body positions (n = 327, 2 = 76.8, df = 1, p < 0.0001, Table 1). Analyzing the turning movements, during which the seals did not change their body position relative to the school, in detail, the seals mostly turned from a “supine” to a “prone” position during descent (98.3%, n = 232) and from a “prone” to a “supine” position during ascent (98.9%, n = 90), indicating that the seal was orientated “dorsal” relatively towards the school of fish. An approaching seal initiated a splitting of the school (Table 1). We documented an asymmetrical split of the school in 78% of the cases (n = 318), thus significantly (2 = 126, df = 1, p < 0.0001) more often than a symmetrical splitting. When the school split asymmetrically, a seal was significantly more likely to follow a smaller sized group (55.0%, n = 176) than a bigger sized group (17.0%, n = 54) or a single fish (9.0%, n = 29; 2 = 143.3, df = 2, p < 0.0001). Only on a few occasions, the seal stopped interacting with the school after it had split (19%, n = 59). We could observe the seals accelerate while interacting with the school of herring. They started accelerating from right below the water surface (59.7%, n = 126) and from deep in the water (38.9%, n = 82). In 41.2% (n = 87) of all documented acceleration instances, a seal accelerated when
following a small group of fish, thus significantly (2 = 89, df = 4, p < 0.0001) more often than when following a big fish unit, a single fish, when another seal accelerated or without apparent reason (Table 1). On the video footage, we did not record a single successful prey catch event. From observations that took place besides recording we know, however, that the seals were catching prey, especially when a small group of fish or a single herring had been separated. Obvious systematic cooperative feeding such as prey herding was not documented; however, we recorded opportunistic behaviors such as kleptoparasitism, stealing fish caught by conspecifics. The analysis of the seals’ foraging behavior revealed that, when the seals were interacting with the school of herring, they were in most instances in motion and actively pursued the fish. They always kept contact with the fish and moved from one side to the other or from below the school to above, comparable to what has been briefly reported for wild seals hunting sand lance before (Bowen et al., 2002). Our result of the seals being mostly “lateral” of the school illustrates that they most often shifted from one side to the other during active prey pursuit. During these movements, a seal adopted a “supine” absolute body position in the majority of the cases and was found mostly slightly below the water surface. We often observed the seals attacking the school out of this position at the water surface. Yurk and Trites (Yurk and Trites, 2000) also observed seals to wait for salmon at the water surface in a “supine” position, however, they did not describe how the seals behaved when attacking prey. In contrast, when hunting sand lance (Bowen et al., 2002), seals were actively ascending first and then descending on the school. Active pursuit also took place in the vertical with the seals ascending and descending continuously. During these instances, the majority of the fish was mostly “dorsal” of the seal irrespective if the seals were residing at or slightly below the water surface or deeper in the water column. This was achieved by a change of the absolute body position in correlation with water depth; while swimming close to the water surface, the seals swam in a “supine” position whereas they adopted a “prone”
388
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389
position at several meters depth. The relative body position “dorsal” of the school was maintained during turning movements; the seals mostly kept their body position unchanged relative to the school irrespective of whether they were ascending or descending. The way the seals were pursuing the fish in the horizontal or vertical harmonizes with visual prey detection and pursuit as hypothesized. Harbor seals possess a large dorsal visual field (Hanke et al., 2006). Consequently, the seals can permanently observe the fish when being “supine” close to or at the water surface or when generally keeping the fish “dorsal” of the body. Visual prey pursuit in seals has been proposed by various authors before (Bonnot, 1932; Davis et al., 1992, 1999; Hobson, 1966; Lavigne et al., 1977; Levenson and Schusterman, 1999), which was, however, solely based on the observation of seals attacking prey from below when hunting pelagic fish (Bonnot, 1932; Davis et al., 1999; Hobson, 1966), a behavior that we could also document in our study. This way, the seals seem to silhouette their prey against the bright water or under-ice surface. It is important to mention that in all the studies we have referred to, including our own, light was richly available enabling the seals to rely on vision while hunting. However, seals often encounter conditions when diving deep, in turbid waters or at night, which do not enable them to use their visual system. This lead many authors to question visual hunting. Especially as it was previously shown that visual acuity is drastically reduced by the presence of dissolved particles in the water (Weiffen et al., 2006) or by decreased ambient luminance (Hanke and Dehnhardt, 2009; Hanke et al., 2009; Schusterman and Balliet, 1971). Under these circumstances, harbor seals might rely on their hydrodynamic or acoustic sense (Dehnhardt et al., 1998). Generally, hunting has to be considered a multimodal process, but our results underline, that vision is indeed involved in this process. The seals, we observed hunting herring, were constantly moving with or diving through the school and changed their behavior between breaking the school and following the trailing edge continuously. In contrast, Bowen et al. (2002) reported two tactics that seals used while foraging on sand lance: in most cases, the seals seem to have attempted at separating small number of fishes and were attacking isolated fishes. In a few instances, the seals moved with the school and tried to catch fish from the trailing edge. In our study, we could not clearly distinguish between these two tactics. We do not think that this difference is due to the different fish species being hunted as overall the seals’ behaviors hunting sand lance or herring are similar. Instead this difference might be explained by the fact, that we could oversee a large area, including the whole school and the seal, in contrast to the more restricted visual field of a critter cam, which might render the interpretation of behavioral sequences more difficult. Recording the school-seal-unit in total allowed us to obtain data showing that the school indeed splits mostly asymmetrically by an approaching seal. And, moreover, the seals in return followed the smaller subunit in most instances. Accelerations occurred mainly when following a small group of fish in line with our hypothesis. By decreasing the number of individuals, the seals avoid or at least reduce the confusion effect of the whole school (Krause and Ruxton, 2002), which is probably the prerequisite for successful prey capture. And indeed, only when chasing a small number of fish, the seals were capturing fish. Unfortunately, we could not record prey capture events but we observed the seals catching fish in various instances especially in a small enclosure in which a subunit of the school used to settle. The reasons why the seals could not be recorded catching fish in the main foraging area remain unknown. The fact that most seals, except for two individuals, continued to work and thus were fed besides hunting might have reduced their motivation to actually catch a fish. However, we could not find any systematic differences between the seals that got additional fish and those exclusively feeding on wild herring. In this context, it
might also be relevant to recall the fact that the seals observed during the course of this study had only come into contact with living fish the year before the school entered the enclosure and thus might not have been expert hunters although again two seals seem to have been efficient enough to rely solely on fish intake from the school. The question arising now is whether our results are representative for seals in general, meaning if the behaviors we documented could also be observed in seals in the wild. There are several apparent differences between the conditions encountered in our study and wild seals hunting in their natural habitat. First, the seals we observed in this study had limited experience in catching living fish as just mentioned. Thus, they might be less experienced in hunting than wild seals of the same age, which could result in different foraging strategies (Zamon 2001). Experience and learning might to some extent be involved in hunting as Bowen et al. (1999) observed that harbor seal pups were accompanying their mothers on foraging dives. The authors speculated that the pups learnt where and what the mother was foraging and thus maybe also learnt how to pursue and capture fish. Second, the seals hunted in very shallow waters of less than 6 m depth thus almost all behaviors can be considered as surface behaviors. Wild harbor seals also perform shallow dives during feeding, sometimes even shallower than 4 m (Lesage et al., 1999), however, in most foraging dives, they reach greater depths (Bjorge et al., 1995; Boness et al., 1994; Gjertz et al., 2001; Tollit et al., 1998). However, harbor seals have to hunt under diverse conditions, and we consider the conditions encountered in our facility as one specific example out of many. It is evident that every condition poses different demands on the seals. As an example, the seals might use different senses for prey detection and capture depending on the foraging depth that correlates with visibility in most waters. In return, the usage of different senses could lead to different hunting strategies. Nevertheless we are confident that the behaviors we observed belong to the natural repertoire of the foraging behavior of harbor seals as the seals we observed showed behaviors that have already been documented for wild seals, however hunting different prey or only reported anecdotically, before. It is generally remarkable that the foraging behavior of seals hunting in shallow waters shows some similarities to when they hunt in deeper water. In conclusion, we were able to obtain valuable information about the foraging behavior in shallow waters of a group of harbor seals living under semi-natural conditions. By continuously filming seals hunting herring from a global perspective, we could extend existing knowledge of harbor seal hunting behavior. Two aspects, the seals’ relative body position to the school and their preferred position before attacking, could be added in favor of visual hunting under light rich conditions. Moreover, we showed clearly that a school indeed breaks into subunits by an approaching seal and that the seals are more likely to follow a smaller subunit afterwards. Acknowledgements We express our gratitude to T. Coppack and J. Krieger commenting on a draft of this paper. This study was supported by a grant of the VolkswagenStiftung to GD. The experiments are in line with the current German law on the protection of animals. We thank the subject editor and an anonymous reviewer for their helpful reviews. The funding sources were not involved in any stage of the study starting with study design, collection, analysis and interpretation of the data up to the writing of the manuscript. References Bjorge, A., Thompson, D., Hammond, P., Fedak, M., Bryant, E., Aarefjord, H., Roen, R., Olsen, M., 1995. Habitat use and diving behaviour of harbour seals in a coastal
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389 archipelago in Norway. In: Blix, A.S., Walloe, L., Ulltang, O. (Eds.), Whales, Seals, Fish and Man. Elsevier Science, Amsterdam, pp. 211–223. Boness, D.J., Bowen, W.D., Oftedal, O.T., 1994. Evidence of a maternal foraging cycle resembling that of otariid seals in a small phocid, the harbor seal. Behav. Ecol. Sociobiol. 34, 95–104. Bonnot, P., 1932. A note on the fishing of the California sea lion. Calif. Fish Game 18, 98–99. Bowen, W.D., Harrison, G.D., 1996. Comparison of harbour seal diets in two inshore habitats of Atlantic Canada. Can. J. Zool. 74, 125–135. Bowen, W.D., Boness, D.J., et al., 1999. Diving behaviour of lactating harbour seals and their pups during maternal foraging trips. Can. J. Zool. 77, 978–988. Bowen, W.D., Tully, D., Boness, D.J., Bulheier, B.M., Marshall, G.J., 2002. Preydependent foraging tactics and prey profitability in a marine mammal. Mar. Ecol. Prog. Ser. 244, 235–245. Davis, R.W., Fiuman, L.A., Williams, T.M., Collier, S.O., Hagey, W.P., Kanatous, S.B., Kohin, S., Horning, M., 1999. Hunting behaviour of a marine mammal beneath the Antarctic fast ice. Science 283, 993–996. Davis, R.W., Wartzok, D., Elsner, R., Stone, H., 1992. A small video camera attached to a Weddell seal: a new way to observe diving behavior. In: Thomas, J., Kastelein, R. (Eds.), Marine Mammal Sensory Systems. Plenum, New York, pp. 631–642. Dehnhardt, G., Mauck, B., Bleckmann, H., 1998. Seal whiskers detect water movements. Nature 394, 235–236. Gjertz, I., Lydersen, C., Wiig, O., 2001. Distribution and diving of harbour seals (Phoca vitulina) in Svalbard. Polar Biol. 24, 209–214. Hanke, F.D., Dehnhardt, G., 2009. Aerial visual acuity in harbor seals (Phoca vitulina) as a function of luminance. J. Comp. Physiol. A 195, 643–650. Hanke, F.D., Hanke, W., Scholtyssek, C., Dehnhardt, G., 2009. Basic mechanisms in pinniped vision. Exp. Brain Res. 199, 299–311. Hanke, W., Römer, R., Dehnhardt, G., 2006. Visual fields and eye movements in a harbor seal (Phoca vitulina). Vis. Res. 46, 2804–2814. Hobson, E.S., 1966. Visual orientation and feeding in seals and sea lions. Nature 210, 326–327. Krause, J., Ruxton, G.D., 2002. Living in Groups. Oxford University Press, Oxford. Lavigne, D.M., Bernholz, G.D., Ronald, K., 1977. Functional aspects of pinniped vision.
389
In: Functional Anatomy of Marine Mammals. Academic Press, London, New York, pp. 135–173. Lesage, V., Hammil, M.O., Kovacs, K.M., 1999. Functional classification of harbor seal (Phoca vitulina) dives using depth profiles, swimming velocity, and an index of foraging success. Can. J. Zool. 77, 74–87. Levenson, D.H., Schusterman, R.J., 1999. Dark adaptation and visual sensitivity in shallow and deep-diving pinnipeds. Mar. Mammal. Sci. 15, 1303–1313. Marshall, G.J., 1998. Crittercam: an animal-borne imaging and data logging system. Mar. Technol. Soc. J. 32, 11–17. Middlemas, S.J., Barton, T.R., Armstrong, J.D., Thompson, P.M., 2005. Functional and aggregative responses of harbour seals to changes in salmonid abundance. Proc. R. Soc. B 273, 193–198. Olsen, M., Bjorge, A., 1995. Seasonal and regional variations in the diet of harbour seal in Norwegian waters. In: Blix, A.S., Walloe, L., Ulltang, O. (Eds.), Whales, Seals, Fish and Man. Elsevier Science, Amsterdam, pp. 271–285. Schusterman, R.J., Balliet, R.F., 1971. Aerial and underwater visual acuity in the California sea lion (Zalophus californianus) as a function of luminance. Ann. N.Y. Acad. Sci. 188, 37–46. Tollit, D.J., Black, A.D., Thompson, P.M., Mackay, A., Corpe, H.M., Wilson, B., van Parijs, S.M., Grellier, K., Parlane, S., 1998. Variations in harbour seal Phoca vitulina diet and dive-depths in relation to foraging habitat. J. Zool. 244, 209–222. Weiffen, M., Möller, B., Mauck, B., Dehnhardt, G., 2006. Effect of water turbidity on the visual acuity of harbor seals (Phoca vitulina). Vis. Res. 46, 1777– 1783. Wright, B.E., Riemer, S.D., Brown, R.F., Ougzin, A.M., Bucklin, K.A., 2007. Assessment of harbor seal predation on adult salmonids in a Pacific Northwest estuary. Ecol. Appl. 17, 338–351. Yurk, H., Trites, A.W., 2000. Experimental attempts to reduce predation by harbor seals on out-migrating juvenile salmonids. Trans. Am. Fish. Soc. 129, 1560– 1566. Zamon, J.E., 2001. Seal predation on salmon and forage fish schools as a function of tidal currents in the San Juan Islands, Washington, USA. Fish. Oceanogr. 10, 353–366.
Contents lists available at ScienceDirect
Mammalian Biology journal homepage: www.elsevier.com/locate/mambio
Short Communication
How harbor seals (Phoca vitulina) pursue schooling herring Meike Kilian, Guido Dehnhardt, Frederike D. Hanke ∗ University of Rostock, Institute for Biosciences, Sensory and Cognitive Ecology, Albert-Einstein-Str. 3, 18059 Rostock, Germany
a r t i c l e
i n f o
Article history: Received 11 December 2014 Accepted 25 April 2015 Handled by Luca Corlatti Available online 2 May 2015 Keywords: Hunting strategy Visual hunting Prey Foraging Pinniped
a b s t r a c t Only a few reports have described how pinnipeds hunt schooling fish. However, we had the opportunity to systematically observe captive harbor seals hunting a school of herring in a shallow enclosure during daylight. The seals actively pursued the fish, mostly changing from one side of the school to the other, swimming in a supine position close to the water surface. They were often found swimming rapidly down to the fish out of this position. When hunting in the vertical, the seals mostly adopted a dorsal body posture relative to the school. They swam and attacked the school in a supine orientation when approaching from the water surface and swam in a prone orientation when approaching from below. They even maintained this relative body position during turning movements. These phenomena suggest that the seals were constantly keeping the school in their large dorsal visual field, which favors visual hunting during daylight and in clear waters. When interacting with the school, the school mostly split asymmetrically, and the seals were following preferably a smaller number of fish afterwards. Successful prey catch was only observed, when a small group or a single herring had been separated, thus the seals probably avoided the confusion effect of the school. In conclusion, we provide detailed insight in seals pursuing schooling prey, thereby extending previous reports and confirming speculations as well as brief observations. © 2015 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
Harbor seals hunt benthic as well as pelagic fish individually and in schools (Bowen et al., 2002; Marshall, 1998). From the few studies available, we know that, when hunting schooling fish, harbor seals attack the edges of the school most likely in an attempt to separate single fish or small subunits (Bowen et al., 2002; Zamon, 2001). Seals often seem to be engaged in “active pursuit”, during which they constantly keep contact with the school and shift from one side of the school to the other (Bowen et al., 2002). Zamon (Zamon, 2001) document behavioral differences of adults and pups; adult harbor seals ingest a mouthful of fish from the edges of a school whereas pups swim through the school and are seen darting on single fish that were separated from the school. Harbor seals also hunt schooling fish close to the sea floor which seems to constrain fish to escape (Olsen and Bjorge, 1995). The same effect could be achieved by chasing fish towards the water surface as described for separated sand lance (Ammodytes dubius) (Bowen et al., 2002). These few insights in harbor seal foraging behavior were obtained with the help of animal-borne video systems (Bowen and Harrison, 1996; Bowen et al., 2002; Marshall, 1998), developed in
∗ Corresponding author. Tel.: +49 381 666971914; fax: +49 381 666971935. E-mail addresses: meike [email protected] (M. Kilian), [email protected] (G. Dehnhardt), [email protected] (F.D. Hanke).
the 1990s (Davis et al., 1992, 1999; Marshall, 1998), or by observations from vessels and land (Middlemas et al., 2005; Wright et al., 2007; Yurk and Trites, 2000; Zamon, 2001). However, these documents are rare and allow only glimpses on the foraging behavior. This is due to two main challenges associated with studying foraging in marine mammals. One challenge is that seals hunt under water, an environment, which is still relatively inaccessible to humans despite advances in modern technology. The other challenge refers to seals being highly mobile species. To successfully monitor their foraging behavior thus requires techniques that do not impose spatial constraints and preferably allows observing the seal-school-unit in total. In the present study, we had the unique opportunity to characterize the foraging behavior of harbor seals hunting schooling fish in greater detail. We could systematically observe a group of nine captive harbor seals (3–27 years), which are kept for scientific reasons under semi-natural conditions in a large seawater enclosure (3–6 m depth, 60 m in length and 30 m in width) separated from the open sea only by a net (mesh size 5 × 5 cm) at the Marine Science Center, Rostock, Germany. All seals were born in zoos and had only come into contact with living fish such as flatfish, eel and herring after moving into the seawater enclosure in summer 2008. In spring 2009, a school of herring (Clupea harengus) with more than 1000 fish entered the seals’ enclosure and stayed for four weeks from mid-April to May. During this time, the seals were observed
http://dx.doi.org/10.1016/j.mambio.2015.04.004 1616-5047/© 2015 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved.
386
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389
Table 1 Ethogram used to analyze the video recordings listing all behaviors documented and their definitions as well as the results listing the frequency with which the behavior occurred as number of total events N and in % (within a category).
Behavioral components of prey pursuit Absolute body position Prone Supine Side Relative body position Dorsal Ventral Lateral Turning movements Unchanged Changed Effect of an approaching seal on school Splitting Symmetrical Asymmetrical Following Following smaller subunit Following larger subunit Following a single fish Not following Acceleration Acceleration on small subunit Acceleration in large subunit Acceleration on single fish Acceleration when others accelerated Acceleration with no apparent reason Prey capture
Definition Body position absolute in the water column Back pointing towards the water surface Belly pointing towards the water surface A flipper pointing towards the water surface Body position relative to the school A seal’s back pointing towards the school A seal’s belly pointing towards the school A seal pointing with one side, tail or snout towards the school A change of absolute and/or relative body position during ascent and descent A seal keeping its relative body position during turning A seal changing its relative body position during turning
The two resulting parts of the school were approximately the same size A different number of individuals was estimated in two or more subunits A seal following fish after splitting A seal following the smaller fish subunit after splitting A seal following the larger fish subunit after splitting A seal following a single fish after splitting A seal did not follow fish after splitting A seal was accelerating A seal was accelerating when following a small fish subunit A seal was accelerating when following a large fish subunit A seal was accelerating when following a single fish A seal was accelerating when another seal accelerated A seal was accelerating with no apparent reason A seal was successfully catching a fish
while hunting throughout the day. Two of the seals exclusively fed on herring captured from the school. All other seals occasionally hunted but also participated in the scientific experiments and routine training activity, during which they received fish rewards. The behavior of the hunting seals during daylight hours was recorded from a top view video camcorder (Canon XL1S; Canon Deutschland GmbH, Krefeld, Germany) that captured the seal as well as the school. On the basis of an ethogram (Table 1) that was developed during preparatory observations and that focused on behavioral components of pursuing the fish and on the effect an approaching seal has on the school, we analyzed 75 min of high quality video footage during which six of the altogether nine seals and the school of fish could be observed throughout the entire water column of the enclosure. The depth within the water column, at which a specific behavior occurred, was documented and categorized as “at the water surface” in cases where at least one part of the seal’s body was seen out of the water, “down to approximately 1 m depth” when the seal was below the water surface, and “deeper than approximately 1 m”. The 1 m transition point was estimated on the basis of the height of a seal and took some uncertainty in the determination of the water depth from above into account. In contrast to previous studies, we were able to continuously record six of the nine hunting seals during daylight hours and to systematically document the foraging behavior over a large area with full view on the seals and the school together. Our goal was to generally extend the existing knowledge of harbor seals hunting schooling fish and to specifically describe the behavioral sequence occurring during the pursuit of a school of herring. We were interested in how the seals were positioning themselves absolutely in the water column and relatively to the fish, which might allow assessing the sensory modality the seals were predominantly using. Our hypothesis was that the seals mainly hunt visually in the shallow enclosure in which light penetrates up to the bottom. Furthermore we wanted to test the hypothesis that the school breaks into subunits by an approaching seal and that the seals are predominantly following a smaller number of fish afterwards. In this context, we hypothesized that attacks occur mostly on small
N
In %
609 1066 179
32.8 57.5 9.7
329 41 1484
17.8 2.2 80.0
327 138
70.3 29.7
91 318
19.3 80.7
176 54 29 59
55.0 17.0 9.0 19.0
87 43 22 9 48 0
41.2 20.4 10.4 4.3 22.7 0
subunits and single fishes, which would document the adaptive value of breaking a school into subunits or separating single fish. Our analysis revealed that, during hunting, the six recorded seals were moving through or with the school and only rarely remained stationary. Their absolute body position was significantly correlated with water depth (2 = 11,770.5, df = 8, p < 0.0001, Fig. 1). Overall, the seals were swimming in a “supine” orientation in 57.7% of the time (n = 1066, 2 = 636.7, df = 2, p < 0.0001). However, when 100 90 80
N=303 N=709
70 N=294
frequency (%)
Documented behaviors
60 50 40 30
N=123 N=183
20 10
N=63 N=100
N=42
N=37
0 at surface
< 1m
> 1m
absolute body position depending on depth
Fig. 1. Orientation of the seals in the water. Absolute body positions adopted by the seals in the water column depending on water depth. The absolute body position (“prone”, “supine”, or “side”) was documented depending on the depth (“at the water surface”, “down to approximately 1 m” and “deeper than approximately 1 m”), at which it was performed within the water column. Frequency is plotted as percentage of total events, absolute numbers are indicated at the respective bars. Black bars indicate the frequency of adopting the body position “prone”, gray bars the frequency of the position “supine” and white bars the frequency of the position “side”.
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389
387
N=914
50 45 40
frequency (%)
35 30 N=460
25 20 15 10
N=146
N=126
N=110
5
N=57 N=23
N=12
N=6
0 prone
supine
side
absolute and relative body position to school Fig. 2. Orientation of the seals in the water and relative to the school. The absolute body position of the seal within the water column as well as its orientation relative to the school are documented. Concerning the relative body position the school could be either “ventral” (black bars), “dorsal” (gray bars) or “lateral” (white bars) of the seal. Absolute body positions are as in Fig. 1. Insets give examples of the respective absolute and relative body positions.
the seals were swimming deeper than approximately 1 m within the water column, they changed their absolute body position into a “prone” orientation in 75.2% of the time (n = 303). If the body position of the seals relative to the school was analyzed, it appeared that in 80% of the cases the seals swam slightly “lateral” orientated to the school (n = 1484, 2 = 1887.4, df = 2, p < 0.0001, Table 1). In the majority of cases being “laterally” orientated, the snout was pointing towards the school. When the absolute as well as the relative body position were taken into consideration, the seals swam in 49.3% (n = 914, 2 = 3487, df = 8, p < 0.0001, Fig. 2B) of the cases in a “supine” orientation slightly “lateral” relative to the school. When interacting with the school, the seals swam constantly ascending and descending but always kept contact with the school during these turning movements. In 70.3% of the time, the seals kept their relative body positions unchanged towards the school of herring by changing their absolute body positions (n = 327, 2 = 76.8, df = 1, p < 0.0001, Table 1). Analyzing the turning movements, during which the seals did not change their body position relative to the school, in detail, the seals mostly turned from a “supine” to a “prone” position during descent (98.3%, n = 232) and from a “prone” to a “supine” position during ascent (98.9%, n = 90), indicating that the seal was orientated “dorsal” relatively towards the school of fish. An approaching seal initiated a splitting of the school (Table 1). We documented an asymmetrical split of the school in 78% of the cases (n = 318), thus significantly (2 = 126, df = 1, p < 0.0001) more often than a symmetrical splitting. When the school split asymmetrically, a seal was significantly more likely to follow a smaller sized group (55.0%, n = 176) than a bigger sized group (17.0%, n = 54) or a single fish (9.0%, n = 29; 2 = 143.3, df = 2, p < 0.0001). Only on a few occasions, the seal stopped interacting with the school after it had split (19%, n = 59). We could observe the seals accelerate while interacting with the school of herring. They started accelerating from right below the water surface (59.7%, n = 126) and from deep in the water (38.9%, n = 82). In 41.2% (n = 87) of all documented acceleration instances, a seal accelerated when
following a small group of fish, thus significantly (2 = 89, df = 4, p < 0.0001) more often than when following a big fish unit, a single fish, when another seal accelerated or without apparent reason (Table 1). On the video footage, we did not record a single successful prey catch event. From observations that took place besides recording we know, however, that the seals were catching prey, especially when a small group of fish or a single herring had been separated. Obvious systematic cooperative feeding such as prey herding was not documented; however, we recorded opportunistic behaviors such as kleptoparasitism, stealing fish caught by conspecifics. The analysis of the seals’ foraging behavior revealed that, when the seals were interacting with the school of herring, they were in most instances in motion and actively pursued the fish. They always kept contact with the fish and moved from one side to the other or from below the school to above, comparable to what has been briefly reported for wild seals hunting sand lance before (Bowen et al., 2002). Our result of the seals being mostly “lateral” of the school illustrates that they most often shifted from one side to the other during active prey pursuit. During these movements, a seal adopted a “supine” absolute body position in the majority of the cases and was found mostly slightly below the water surface. We often observed the seals attacking the school out of this position at the water surface. Yurk and Trites (Yurk and Trites, 2000) also observed seals to wait for salmon at the water surface in a “supine” position, however, they did not describe how the seals behaved when attacking prey. In contrast, when hunting sand lance (Bowen et al., 2002), seals were actively ascending first and then descending on the school. Active pursuit also took place in the vertical with the seals ascending and descending continuously. During these instances, the majority of the fish was mostly “dorsal” of the seal irrespective if the seals were residing at or slightly below the water surface or deeper in the water column. This was achieved by a change of the absolute body position in correlation with water depth; while swimming close to the water surface, the seals swam in a “supine” position whereas they adopted a “prone”
388
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389
position at several meters depth. The relative body position “dorsal” of the school was maintained during turning movements; the seals mostly kept their body position unchanged relative to the school irrespective of whether they were ascending or descending. The way the seals were pursuing the fish in the horizontal or vertical harmonizes with visual prey detection and pursuit as hypothesized. Harbor seals possess a large dorsal visual field (Hanke et al., 2006). Consequently, the seals can permanently observe the fish when being “supine” close to or at the water surface or when generally keeping the fish “dorsal” of the body. Visual prey pursuit in seals has been proposed by various authors before (Bonnot, 1932; Davis et al., 1992, 1999; Hobson, 1966; Lavigne et al., 1977; Levenson and Schusterman, 1999), which was, however, solely based on the observation of seals attacking prey from below when hunting pelagic fish (Bonnot, 1932; Davis et al., 1999; Hobson, 1966), a behavior that we could also document in our study. This way, the seals seem to silhouette their prey against the bright water or under-ice surface. It is important to mention that in all the studies we have referred to, including our own, light was richly available enabling the seals to rely on vision while hunting. However, seals often encounter conditions when diving deep, in turbid waters or at night, which do not enable them to use their visual system. This lead many authors to question visual hunting. Especially as it was previously shown that visual acuity is drastically reduced by the presence of dissolved particles in the water (Weiffen et al., 2006) or by decreased ambient luminance (Hanke and Dehnhardt, 2009; Hanke et al., 2009; Schusterman and Balliet, 1971). Under these circumstances, harbor seals might rely on their hydrodynamic or acoustic sense (Dehnhardt et al., 1998). Generally, hunting has to be considered a multimodal process, but our results underline, that vision is indeed involved in this process. The seals, we observed hunting herring, were constantly moving with or diving through the school and changed their behavior between breaking the school and following the trailing edge continuously. In contrast, Bowen et al. (2002) reported two tactics that seals used while foraging on sand lance: in most cases, the seals seem to have attempted at separating small number of fishes and were attacking isolated fishes. In a few instances, the seals moved with the school and tried to catch fish from the trailing edge. In our study, we could not clearly distinguish between these two tactics. We do not think that this difference is due to the different fish species being hunted as overall the seals’ behaviors hunting sand lance or herring are similar. Instead this difference might be explained by the fact, that we could oversee a large area, including the whole school and the seal, in contrast to the more restricted visual field of a critter cam, which might render the interpretation of behavioral sequences more difficult. Recording the school-seal-unit in total allowed us to obtain data showing that the school indeed splits mostly asymmetrically by an approaching seal. And, moreover, the seals in return followed the smaller subunit in most instances. Accelerations occurred mainly when following a small group of fish in line with our hypothesis. By decreasing the number of individuals, the seals avoid or at least reduce the confusion effect of the whole school (Krause and Ruxton, 2002), which is probably the prerequisite for successful prey capture. And indeed, only when chasing a small number of fish, the seals were capturing fish. Unfortunately, we could not record prey capture events but we observed the seals catching fish in various instances especially in a small enclosure in which a subunit of the school used to settle. The reasons why the seals could not be recorded catching fish in the main foraging area remain unknown. The fact that most seals, except for two individuals, continued to work and thus were fed besides hunting might have reduced their motivation to actually catch a fish. However, we could not find any systematic differences between the seals that got additional fish and those exclusively feeding on wild herring. In this context, it
might also be relevant to recall the fact that the seals observed during the course of this study had only come into contact with living fish the year before the school entered the enclosure and thus might not have been expert hunters although again two seals seem to have been efficient enough to rely solely on fish intake from the school. The question arising now is whether our results are representative for seals in general, meaning if the behaviors we documented could also be observed in seals in the wild. There are several apparent differences between the conditions encountered in our study and wild seals hunting in their natural habitat. First, the seals we observed in this study had limited experience in catching living fish as just mentioned. Thus, they might be less experienced in hunting than wild seals of the same age, which could result in different foraging strategies (Zamon 2001). Experience and learning might to some extent be involved in hunting as Bowen et al. (1999) observed that harbor seal pups were accompanying their mothers on foraging dives. The authors speculated that the pups learnt where and what the mother was foraging and thus maybe also learnt how to pursue and capture fish. Second, the seals hunted in very shallow waters of less than 6 m depth thus almost all behaviors can be considered as surface behaviors. Wild harbor seals also perform shallow dives during feeding, sometimes even shallower than 4 m (Lesage et al., 1999), however, in most foraging dives, they reach greater depths (Bjorge et al., 1995; Boness et al., 1994; Gjertz et al., 2001; Tollit et al., 1998). However, harbor seals have to hunt under diverse conditions, and we consider the conditions encountered in our facility as one specific example out of many. It is evident that every condition poses different demands on the seals. As an example, the seals might use different senses for prey detection and capture depending on the foraging depth that correlates with visibility in most waters. In return, the usage of different senses could lead to different hunting strategies. Nevertheless we are confident that the behaviors we observed belong to the natural repertoire of the foraging behavior of harbor seals as the seals we observed showed behaviors that have already been documented for wild seals, however hunting different prey or only reported anecdotically, before. It is generally remarkable that the foraging behavior of seals hunting in shallow waters shows some similarities to when they hunt in deeper water. In conclusion, we were able to obtain valuable information about the foraging behavior in shallow waters of a group of harbor seals living under semi-natural conditions. By continuously filming seals hunting herring from a global perspective, we could extend existing knowledge of harbor seal hunting behavior. Two aspects, the seals’ relative body position to the school and their preferred position before attacking, could be added in favor of visual hunting under light rich conditions. Moreover, we showed clearly that a school indeed breaks into subunits by an approaching seal and that the seals are more likely to follow a smaller subunit afterwards. Acknowledgements We express our gratitude to T. Coppack and J. Krieger commenting on a draft of this paper. This study was supported by a grant of the VolkswagenStiftung to GD. The experiments are in line with the current German law on the protection of animals. We thank the subject editor and an anonymous reviewer for their helpful reviews. The funding sources were not involved in any stage of the study starting with study design, collection, analysis and interpretation of the data up to the writing of the manuscript. References Bjorge, A., Thompson, D., Hammond, P., Fedak, M., Bryant, E., Aarefjord, H., Roen, R., Olsen, M., 1995. Habitat use and diving behaviour of harbour seals in a coastal
M. Kilian et al. / Mammalian Biology 80 (2015) 385–389 archipelago in Norway. In: Blix, A.S., Walloe, L., Ulltang, O. (Eds.), Whales, Seals, Fish and Man. Elsevier Science, Amsterdam, pp. 211–223. Boness, D.J., Bowen, W.D., Oftedal, O.T., 1994. Evidence of a maternal foraging cycle resembling that of otariid seals in a small phocid, the harbor seal. Behav. Ecol. Sociobiol. 34, 95–104. Bonnot, P., 1932. A note on the fishing of the California sea lion. Calif. Fish Game 18, 98–99. Bowen, W.D., Harrison, G.D., 1996. Comparison of harbour seal diets in two inshore habitats of Atlantic Canada. Can. J. Zool. 74, 125–135. Bowen, W.D., Boness, D.J., et al., 1999. Diving behaviour of lactating harbour seals and their pups during maternal foraging trips. Can. J. Zool. 77, 978–988. Bowen, W.D., Tully, D., Boness, D.J., Bulheier, B.M., Marshall, G.J., 2002. Preydependent foraging tactics and prey profitability in a marine mammal. Mar. Ecol. Prog. Ser. 244, 235–245. Davis, R.W., Fiuman, L.A., Williams, T.M., Collier, S.O., Hagey, W.P., Kanatous, S.B., Kohin, S., Horning, M., 1999. Hunting behaviour of a marine mammal beneath the Antarctic fast ice. Science 283, 993–996. Davis, R.W., Wartzok, D., Elsner, R., Stone, H., 1992. A small video camera attached to a Weddell seal: a new way to observe diving behavior. In: Thomas, J., Kastelein, R. (Eds.), Marine Mammal Sensory Systems. Plenum, New York, pp. 631–642. Dehnhardt, G., Mauck, B., Bleckmann, H., 1998. Seal whiskers detect water movements. Nature 394, 235–236. Gjertz, I., Lydersen, C., Wiig, O., 2001. Distribution and diving of harbour seals (Phoca vitulina) in Svalbard. Polar Biol. 24, 209–214. Hanke, F.D., Dehnhardt, G., 2009. Aerial visual acuity in harbor seals (Phoca vitulina) as a function of luminance. J. Comp. Physiol. A 195, 643–650. Hanke, F.D., Hanke, W., Scholtyssek, C., Dehnhardt, G., 2009. Basic mechanisms in pinniped vision. Exp. Brain Res. 199, 299–311. Hanke, W., Römer, R., Dehnhardt, G., 2006. Visual fields and eye movements in a harbor seal (Phoca vitulina). Vis. Res. 46, 2804–2814. Hobson, E.S., 1966. Visual orientation and feeding in seals and sea lions. Nature 210, 326–327. Krause, J., Ruxton, G.D., 2002. Living in Groups. Oxford University Press, Oxford. Lavigne, D.M., Bernholz, G.D., Ronald, K., 1977. Functional aspects of pinniped vision.
389
In: Functional Anatomy of Marine Mammals. Academic Press, London, New York, pp. 135–173. Lesage, V., Hammil, M.O., Kovacs, K.M., 1999. Functional classification of harbor seal (Phoca vitulina) dives using depth profiles, swimming velocity, and an index of foraging success. Can. J. Zool. 77, 74–87. Levenson, D.H., Schusterman, R.J., 1999. Dark adaptation and visual sensitivity in shallow and deep-diving pinnipeds. Mar. Mammal. Sci. 15, 1303–1313. Marshall, G.J., 1998. Crittercam: an animal-borne imaging and data logging system. Mar. Technol. Soc. J. 32, 11–17. Middlemas, S.J., Barton, T.R., Armstrong, J.D., Thompson, P.M., 2005. Functional and aggregative responses of harbour seals to changes in salmonid abundance. Proc. R. Soc. B 273, 193–198. Olsen, M., Bjorge, A., 1995. Seasonal and regional variations in the diet of harbour seal in Norwegian waters. In: Blix, A.S., Walloe, L., Ulltang, O. (Eds.), Whales, Seals, Fish and Man. Elsevier Science, Amsterdam, pp. 271–285. Schusterman, R.J., Balliet, R.F., 1971. Aerial and underwater visual acuity in the California sea lion (Zalophus californianus) as a function of luminance. Ann. N.Y. Acad. Sci. 188, 37–46. Tollit, D.J., Black, A.D., Thompson, P.M., Mackay, A., Corpe, H.M., Wilson, B., van Parijs, S.M., Grellier, K., Parlane, S., 1998. Variations in harbour seal Phoca vitulina diet and dive-depths in relation to foraging habitat. J. Zool. 244, 209–222. Weiffen, M., Möller, B., Mauck, B., Dehnhardt, G., 2006. Effect of water turbidity on the visual acuity of harbor seals (Phoca vitulina). Vis. Res. 46, 1777– 1783. Wright, B.E., Riemer, S.D., Brown, R.F., Ougzin, A.M., Bucklin, K.A., 2007. Assessment of harbor seal predation on adult salmonids in a Pacific Northwest estuary. Ecol. Appl. 17, 338–351. Yurk, H., Trites, A.W., 2000. Experimental attempts to reduce predation by harbor seals on out-migrating juvenile salmonids. Trans. Am. Fish. Soc. 129, 1560– 1566. Zamon, J.E., 2001. Seal predation on salmon and forage fish schools as a function of tidal currents in the San Juan Islands, Washington, USA. Fish. Oceanogr. 10, 353–366.