Tourism in Marine Environments, Vol. 15, No. 3–4, pp. 159-171 1544-273X/20 $60.00 + .00 Printed in the USA. All rights reserved. DOI: https://doi.org/10.3727/154427320X15945013137030 Copyright © 2020 Cognizant, LLC. E-ISSN 2169-0197 www.cognizantcommunication.com

BEHAVIORS OF GREY SEALS (HALICHOERUS GRYPUS) ADDRESSED TOWARDS HUMAN SWIMMERS DURING EXPERIMENTAL OPEN WATER ENCOUNTERS OFF (GERMAN BIGHT, )

MICHAEL SCHEER

Independent Scholar, Bremen, Germany

Risks arising for humans during swim encounters with seals are poorly understood. This study was initiated to examine behaviors of unhabituated grey seals addressed towards humans during experi- mental, noncommercial seal-swim activities off Heligoland. In total, 26 in-water encounters were conducted. Behavioral classes and the number of seals simultaneously approaching swimmers were time sampled. A set of risky and nonrisky interactive behaviors was continuously sampled. Seals spent approximately the same amount of time interacting with swimmers (53%) as they did ignoring them (47%). Seals displayed higher rates of nonrisky behaviors than risky ones, but risky behaviors occurred during 73% of all seal-swims. Seals remained ≤20 m near swimmers for 51% and ≤1 m for 13% of the time. A mean number of 0.65 and 0.18 seals approached swimmers per minute within a range of ≤20 m and ≤1 m, respectively. Behavioral classes, interactive behaviors, and the number of seals approaching ≤20 m did not vary significantly throughout seal-swims but the number of seals approaching ≤1 m moderately decreased. Due to high rates of risky behaviors, it is recommended to promote public awareness on site and to regulate seal-swims before commercial operations emerge.

Key words: Grey seals; Halichoerus grypus; Tourism; Swim-with seals; Behavior

Introduction approaches, swim-with-seals activities in wild settings have become a popular and growing segment of this -focused tourism is part of the marine industry (Constantine, 1999; Curtin & Garrod, 2008; wildlife tourism industry that has increased rapidly Kirkwood et al., 2003; Muir, Barnes, & Reid, 2006; (Cowling, Kirkwood, Boren, & Scarpaci, 2014; Newsome & Rodger, 2008; Scarpaci, Nugegoda, & O’Connor, Campbell, Cortez & Knowles, 2009; Corkeron, 2005). Past research has demonstrated Stafford-Bell, Scarr & Scarpaci, 2012; Strong negative impacts of land- and boat-based approaches & Morris, 2010). Next to land- and boat-based by tourists for seals during commercial activities

Address correspondence to Michael Scheer, Brunnenstrasse 15-16, 28203 Bremen, Germany. Tel: +49(0)173 238 61 56; E-mail: [email protected] Delivered 159by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. 160 SCHEER

(e.g., Boren, Gemmell, & Barton, 2002; Cassini, Salgado Kent, & Robinson, 2017; Stafford-Bell et 2001; Henry & Hammill, 2001; Holcomb, Young, al., 2012). During one extreme case, a diver was & Gerber, 2009; Kovacs & Inness, 1990). However, drowned by a seal (Hydrurga leptonyx) seal-swim activities remain understudied (Curtin & (Muir et al., 2006). Muir et al. (2006) pointed out Garrod, 2008; Kirkwood et al., 2003). that people who want to encounter free-ranging Commercial swim-with programs targeting wild seals are at risk, and they need appropriate infor- have become critical, and research is mation about potential hazards. needed on behavioral responses initiated by the From a management perspective, understand- . Although swimmers often try to attract or ing close encounters with seals can inform recom- to closely approach seals in the water (Boren, Gem- mendations to reduce the likelihood of aggressive mell, & Barton, 2009), it generally depends on the and dangerous interactions. Kirkwood et al. (2003) animals to initiate or terminate contact, or to avoid stated that commercial seal-swim activities in humans and stay out of sight. However, the level of wild settings are one of the most critical areas habituation, swimmer behavior during encounters, where research is needed to foster regulations. It as well as the presence or absence of guidelines was proposed that baseline and longitudinal stud- and guides may influence the nature of encoun- ies be conducted before commercial programs are ters. During commercial and unguided swim-with implemented and to introduce species-specific activities with habituated animals, Australian fur regulations (Curtin & Garrod, 2008; Kirkwood et seals ( pusillus doriferus) were al., 2003; Öqvist, Granquist, Burns, & Angerbjörn, found to increase haul-out events in response to 2018). While the responses of animals being habit- human swimmers (Stafford-Bell et al., 2012) and uated to commercial swim activities have been New Zealand fur seals () studied for some pinniped species and regions interacted with swimmers but mostly ignored them (Boren et al., 2009; Cowling et al., 2014; Dans et (Cowling et al., 2014). Boren et al. (2009) showed al., 2017; Osterrieder et al., 2017; Scarpaci et al., that the presence of guides during commercial 2005; Stafford-Bell et al., 2012), behavioral studies swims reduces the number of avoidance responses on in-water encounters with unhabituated animals among New Zealand fur seals because guides man- are absent. Such insights are essential to outline aged inappropriate swimmer behaviors. In many and understand natural responses before commer- areas where seal swims occur, there are no regu- cial operations are established. lations or guidelines present (Öqvist, Granquist, In the North Sea, growing populations of grey Burns, & Angerbjörn, 2018). seals (Halichoerus grypus) are increasingly inter- Next to negative impacts for seals, humans can acting with humans. For example, at the South experience health risks as well. As pointed out by Devon coastline (UK), 70,000 tourists are annually Moscardo, Taverner, and Woods (2006), the poten- engaged in watching and swimming with grey seals tial effects of wildlife on tourist safety remain (Curtin, Richards, & Westcott, 2009), and such understudied. Swimmers and divers often are in activities represent a significant economic potential close contact with the animals in their open water (Bosetti & Pearce, 2003). The most economically environments and thus are exposed to unpredict- productive seal watching sites are those that can able behaviors. Seals were reported to address be easily approached, are close to human popula- inquisitive behaviors towards swimmers such as tion centers, and have spatial and temporal predict- encircling, close approaches, mimicking of human ability for viewing the animals (Curtin & Garrod, postures, and establishing physical contact in a 2008; Kirkwood et al., 2003). The island of Heli- neutral way. However, behaviors of the seals also goland has several ferry and flight connections to included potentially threatening displays: they the mainland and provides a sound and relatively were shown to chase, duck, pull, and bite swim- cheap touristic infrastructure. However, there is a mers and divers, and they showed threat and sexual lack of regulations regarding human contact with behaviors (Boren, 2001; Constantine, 1999; Dans, pinnipeds. From 2008 to 2011, more than 300,000 Crespo, & Coscarella, 2017; Kirkwood et al., 2003; tourists annually visited the island and about 50% Martinez, 2003; Muir et al., 2006; Osterrieder, of them came to Dune islet (Gemeinde Heligoland, Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. GREY SEAL BEHAVIOR TOWARD SWIMMERS 161

2011). Grey seals on Dune islet have become a by them during their 20–30 min swims (personal popular wildlife attraction in recent years with observation). Though scratches were reported to the animals being annually confronted with land- happen incidentally by seals during rapid underwa- based approaches mainly during the pupping sea- ter approaches, swimmers chose to wear wetsuits son in wintertime. Habituation effects are unknown to protect themselves. Regulations, guidelines, or because residency and migration patterns remain educational signs do not exist on site to this date. unstudied so far for local animals. Thus, there is a growing concern among wildlife For breeding, females at UK breeding sites managers about the potential negative impacts for typically return to the site where they were born human bathers, snorkelers, and divers. (Pomeroy, Anderson, Twiss, & McConnell, 1994). Behaviors displayed by seals and addressed By extrapolating this to Dune islet, the chance of towards swimmers and divers during open water successive contacts with humans across seasons encounters have received little attention by research- is quite probable. For the summer months, an ers. Thus, their structure and function mostly unknown fraction of tourists uses local beaches for remain unclear. Furthermore, seal-swim activities recreational activities such as bathing, diving, and with unhabituated animals remain understudied so snorkeling. Consequently, humans have had occa- far. This study was conducted to experimentally sional and incidental contact with grey seals. How- investigate responses of grey seals during controlled ever, in contrast to commercial seal-swim sites, on seal-swim activities with humans by examining self- Dune islet just a few snorkelers and divers are seek- initiated behaviors addressed towards swimmers. ing direct contact with them. On site, commercial operations have been absent so far. Beach use on site has been spatially separated into bathing and Methodology nonbathing zones for safety reasons and to provide Study Site space for seals to avoid humans. However, seals do enter the bathing zones and interact with swim- Data were collected in June 2014 on the south- mers and waders. Swimmers reported that they ern beach of Dune islet, Heligoland (German were sometimes touched, grabbed, and scratched Bight, North Sea; Fig. 1). While a research permit

Figure 1. Map of the study area. Left: location of the island of Heligoland in the North Sea. Right: the gray area and arrow indicate the southern beach of the Dune islet where swim encounters were conducted. Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. 162 SCHEER

was not needed, the study was conducted in agree- their own but waited for reactions by the animals ment with local stakeholders. Dune islet, located while passively floating at the water surface and in close to the main island of Heligoland, is used as a stationary position. They always remained within a haul-out and breeding site by grey seals. The ani- the bathing zone for safety reasons and to provide mals were detected for the first time in the area in space for seals to avoid them. 1975 (Vauk, 1978). Since then, the population has Data recording started from the moment the increased (Abt & Engler, 2009). In the 2014/2015 swimmers entered the water and was terminated pupping season the number of newborns reached after 30 min, which corresponds with the maxi- 244 individuals. Though the animals use beaches mum amount of time usual bathers remain in the year-round, their numbers concentrate in winter- water. It was noted whether additional bathers time (November–February) to give birth and to entered the water during the swims. Behavioral mate. During this period, all age and sex classes observations were made from the beach by a sin- are present (Abt & Engler, 2009; Härkönen et al., gle observer and documented on data sheets. Once 2007). During the summer months, the abundance swimmers entered the water, the behavior of each of grey seals decreases on site (Rolf Blädel, per- seal in the water was scan sampled at 1-min inter- sonal communication) and might correspond with vals (Altmann, 1974; Martin & Bateson, 1993) trends in adjacent areas (Abt, Hoyer, Koch, & using behavioral classes adopted from Boren et al. Adelung, 2002). (2009). These were “Interaction” [seal(s) actively swimming towards human swimmer(s)] and “Neu- tral” [seal(s) are swimming in the bathing zone or Sampling Protocol and Code of Conduct are lying along its beach but neither showing an Experimental swims were initiated when one or apparent change in swim direction nor entering the more seals were swimming inside/outside the bath- water, respectively]. The behavioral class “Avoid- ing zone (300 m × 150 m in size) or were lying ance” [seal(s) porpoised away from the stimulus along the stretch of the beach (see Fig. 1). They and did not return to it] was later excluded because were conducted during daylight hours and calm it did not occur. The dominant response of seals in sea states (≤ Beaufort 3). To imitate typical bathers’ the water (when more than 75% of individuals are behavior and to prevent from the unusual attrac- engaged in the same behavioral activity) was used tion, swimmers adhered to the following code-of- for behavioral analysis. Furthermore, the number conduct: they did not make any loud noise, and of seals swimming within a range of ≤20 m and they did not initiate physical contact with seals. ≤1 m was recorded for each scan sample. However, when seals started contact, swimmers Species-specific behavioral repertoires can be did not move away from the animals but tolerated described and catalogued with an ethogram required touch. Though physical contact is undesirable, to pursue further behavioral analysis (Lehner, it has been reported to happen regularly between 1987). By using an a priori ethogram (Table 1), bathers and seals. Every few minutes, the experi- interactive behaviors were continuously sampled mental swimmers swam 10–20 m to provide some using the ad libitum method (Altmann, 1974; Mar- movement similarly to usual bathers. Swimmers tin & Bateson, 1993). Interactive behaviors were used black colored wetsuits so that they all looked defined as behaviors initiated by a seal and directed the same and to protect themselves from the low towards a swimmer within a 20-m range. Behav- water temperatures of 13.5–15.3°C. To ensure a iors were categorized as “risky” when a seal made moderate water depth (1–3 m; this is the standard abrupt movements at close range (≤1 m) to a swim- water depth that bathers are in at this location) in mer, seal and swimmer had physical contact, when the bathing zone so that seals can easily maneuver behaviors had the potential to cause physical harm through the water column, 1–3 experimental swim- to humans, or had the potential for disease transmis- mers entered the water around high tide (±2 h). sion. They were categorized as “nonrisky” when They entered the water individually or together and the distance between seal and swimmer was ≥1 m, moved slowly to 20 m offshore with a water depth behaviors were inquisitive but nonaggressive, and of at least 1 m. Swimmers did not approach seals on when there were no signs of threat by the animals Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. GREY SEAL BEHAVIOR TOWARD SWIMMERS 163

Table 1 Ethogram of Behaviors Displayed by Grey Seals During Seal-Swim Activities off Heligoland

Behavior Code Description

Nonrisky Close approach CA approaches human swimmer to ≥1 m distance, at the water surface or from underwater. Head-up stare HUS Animal lifts its head vertically out of the water while floating stationary or swimming towards swimmer (<20 m). The eyes are orientated towards the swimmer for one to several seconds. Risky Initiate contact IC Animal initiates single or consecutive gentle physical contact/s with rostrum, mouth, whis- kers, or ventral/dorsal body parts in a non-aggressive way while its mouth is closed. Contact is initiated at the water surface or from underwater. Mouth grab MG Animal opens its mouth and grabs a swimmer’s feet, leg, hand or arm, mostly in a non- forceful way. Sometimes the contact was more forceful (i.e., pulling off wetsuit sock liners and gloves). Foreflipper scratch FS Animal scratches swimmer with its foreflipper claws, mainly during erratic interactions. Foreflipper hug FH Animal is positioned below a human swimmer in a belly up position and hugs a swimmer’s leg, foot, arm or hand with its foreflippers. Tolerate touch TT A seal is being touched by a human and tolerates this contact, i.e. touching does not elicit an avoidance reaction. Rapid submerge RS Animal rapidly submerges with water splashes while swimming close (≤1 m) to human swim- mers. The animal disappears from the surface with an abrupt body movement. After a few seconds it re-appears at the surface close to the swimmers (< 20 m).

(Dans et al., 2017; Osterrieder et al., 2017; Waltzek, by the total number of samples, and are expressed ± Cortés-Hinojosa, Wellehan, & Gray, 2012). While standard deviations. Significant differences in seal the observer could detect most behaviors, some responses during swims were assessed by applying occurred underwater and were only identified by ANOVA, Student’s t-test (based on a statistical tol- the experimental swimmers themselves (i.e., a erance of p ≤ 0.05), and linear regression analysis physical contact made from underwater). A set of (R2) using R software (R Foundation for Statistical hand gestures signaling five distinctive behaviors Computing, Vienna, Austria). (initiate contact, mouth grab, foreflipper hug, fore- flipper scratch, and close approach; see Table 1) Results was agreed between observer and swimmers and signaled to the observer immediately when one of A total of 26 seal-swims were conducted between these behaviors occurred. June 16 and 25, 2014, and 780 1-min scan samples During interactions, seals and their behaviors were obtained. During five swim encounters, a were opportunistically photographed from the maximum of four further unknown bathers without beach using a digital camera with a 100–400-mm wetsuits entered the bathing zone simultaneously lens. The photo material was used for the analy- by chance. However, they all left the water after sis of age classes of grey seals being involved in a few minutes, apparently due to the low water interactions by two independent observers. These temperatures. Thus, the number of swimmers in were yearlings (<1 year), juveniles (1–4 years), the water was mostly 1–3 during seal-swim activi- and adults (>4 years) (Hall & Thompson, 2009; ties and consisted of experimental swimmers. All McLaren, 1993). 26 seal-swims resulted in interactions with seals to some extent. For 53% of the time seals interacted with swimmers and for 47% they did not. Statistical Analysis Overall, grey seals showed a higher variability Mean values of relative occurrences of nonrisky of risky (N = 6) than nonrisky behaviors (N = 2) and risky behaviors were calculated by dividing the (Table 1) but nonrisky behaviors occurred at higher number of scan samples with/without occurrences rates than risky ones (Table 2). Risky behaviors Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. 164 SCHEER

Table 2 Absolute (N total), Relative (M, SD), and Ranging (per min) Occurrence of Nonrisky and Risky Behaviors of Grey Seals During 26 Seal-Swim Activities off Heligoland

Nonrisky Risky

HUS CA IC MG RS TT FS FH N total 1,039 239 175 65 47 8 7 6 M/min 1.33 0.31 0.22 0.08 0.06 0.01 0.01 0.01 SD 0.20 0.08 0.10 0.05 0.03 0.02 0.01 0.01 Range/min 0–9 0–3 0–6 0–3 0–2 0–2 0–1 0–1 Occurrence 26 25 13 8 19 2 2 3 Note. See Table 1 for behavioral codes. Occurrence shows the number of 30-min seal-swims during which a behavior occurred.

were observed during 73% of all encounters (N = although once several holes were put in a wetsuit. 19) (see Table 2 for one/zero occurrences). “Head- During three instances, seals tried to pull off wet- up stare” was the most frequent behavior (N = suit sock liners and gloves. 1,039; M = 1.33/min, SD = 0.2) followed by “close Grey seal response did not vary throughout approach” (N = 239; M = 0.31/min, SD = 0.08). seal-swims. Neither behavioral classes (one-way “Initiate contact” (N = 175; M = 0.22/min, SD = ANOVA: F = 8.29, p < 0.006) (Fig. 2) nor inter- 0.1) and “mouth grab” (N = 65; M = 0.08/min, active behaviors changed significantly (one-way SD = 0.05) were the most frequent risky behaviors. ANOVA: F = 799.24, p < 0.001) (Fig. 3) for the “Mouth grab” was expressed at different levels. most common behaviors. Interactive behaviors Mostly it were gentle and bite strength was low, were already observed during the first minutes after

Figure 2. Grey seal response (behavioral classes) over time during seal-swim activities (N = 26). Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. GREY SEAL BEHAVIOR TOWARD SWIMMERS 165

Figure 3. Grey seal response over time during seal-swims (N = 26). See Table 1 for behavioral codes. Lines are linear regres- sions. (a) Nonrisky behaviors “head-up stare” (gray) and “close approach” (black), and (b) the most common risky behaviors “initiate contact” (gray) and “mouth grab” (black).

Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. 166 SCHEER

the onset of an encounter (Fig. 3). Though in close during 26 seal-swims (range 1–4). During 18 range to the swimmers and attracted by them, seals seal-swims, the age class of 27 animals could be showed “rapid submerge” as an apparent form of determined. These were 11 yearlings, 14 juveniles, startle during approaches. This behavior mostly and 2 adults. Due to distance, incomplete visibil- occurred when swimmers stopped floating and lifted ity, or underwater approaches on part of the seals, their bodies out of shallow water or when seals sur- the classification of all individuals involved was faced close to swimmers, apparently due to the low not possible. Thus, individuals were not individu- underwater visibility (mostly <5 m). However, seals ally identifiable and it remained unclear whether always reapproached swimmers after a few seconds. single animals approached swimmers during con- On average, seals were ≤20 m from swimmers secutive encounters. However, up to four animals for 51% of the time and ≤1 m for 13% of the time. approached swimmers simultaneously, and interac- It rarely occurred (1% of the time) that seals did not tive animals were found among all age classes. interact with swimmers while being within a 20-m distance. On average, a seal approached a swim- Discussion mer every 1.5 min ≤20 m and every 5.6 min ≤1 m. A mean number of 0.65 (SD = 0.13) and 0.18 This study examined for the first time the behav- (SD = 0.06) seals approached swimmer(s) per min- ior of grey seals in response to swimmers during ute within a range of ≤20 m and ≤1 m, respectively. open water encounters. It adds another seal spe- While the number of seals ≤20 m remained rela- cies to the growing literature on characterizing seal tively constant (regression analysis: R2 = 0.02; one- responses to tourism that might be helpful to gen- way ANOVA: F = 368.17, p < 0.001) throughout eralize across species and to determine regulations the course of seal-swims, the number of seals ≤1 m in situations that display a lack of data. Grey seals moderately decreased (R2 = 0.42) (Fig. 4). interacted with swimmers for 53% and behaved A mean maximum of 2.5 (SD = 0.69) animals neutral for 47% of the time. Cowling et al. (2014) simultaneously approached swimmers ≤20 m found that New Zealand fur seals at a small colony

Figure 4. Numbers of grey seals approaching swimmer(s) per min ≤20 m (gray) and ≤1 m (black) during 30 min seal-swim activities (N = 26). Lines are linear regressions. Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. GREY SEAL BEHAVIOR TOWARD SWIMMERS 167

interacted with swimmers 41% of the time, and data preliminary indicate that swimmers were pref- for 54% they ignored them. Boren et al. (2009) erably approached by subadult grey seals and less observed that New Zealand fur seals at a larger col- by adults. Approaches by subadult animals during ony mostly showed no response (64%), followed this study might be explained as play or explora- by exhibiting a response (24%) to swimmers. tion and curiosity. Yearlings and subadult grey seals Thus, grey seals during this study interacted with during this study were presumably born on Dune swimmers at a higher rate than did New Zealand islet and they have become habituated to land-based fur seals. Cowling et al. (2014) further found for approaches by humans since birth. They might have fur seals that they remained ≤20 m near swimmers learned that humans do not pose a danger to them during an entire encounter on 100% of seal-swims, (see also Lovasz, Croft, & Banks, 2008; Orsini, while during this study 51% was measured. Fur Shaughnessy, & Newsome, 2006). However, the seals often behaved neutrally despite ranging close age class of less than 50% of all animals during this to swimmers while grey seals rarely did when close study could be identified. Furthermore, this study to humans in the water. Cowling et al. (2014) found was conducted for 10 days. It might be that the an avoidance rate of 5% and Boren et al. (2009) a same seal individuals approached swimmers dur- rate of 6%. During this study, seals did not avoid ing consecutive days, which might bias age-related swimmers entering the water. However, it remains data obtained here. Thus, interpretations have to be questionable whether both species and settings are regarded with caution. However, the simultaneous comparable. Both studies on New Zealand fur seals approach of up to four animals and the fact that observed habituated animals during commercial approaching animals were found among all age operations. Habituation to commercial swimmer classes rather indicates that seal behavior is repre- approaches has been shown to increase avoidance sentative of the population. among dolphins (Constantine, 2001). Furthermore, Interactive behaviors were observed during all swimmers have been reported to disturb seals dur- seal-swims with grey seals, and nonrisky behav- ing commercial operations at established sites iors occurred at higher rates than risky ones. In that by splashing water, encircling and chasing seals, sense, the nature of interactions could be described or attempting to touch them (Boren et al., 2009). to be benign. However, during 73 % of seal-swims, Grey seals on Dune islet are not habituated to in- grey seals at some point displayed risky behaviors, water approaches. Furthermore, swimmers during and it is noteworthy that “mouth grabs” occurred this study behaved passively and did not actively along a continuum of bite strengths. While most approach the animals. These differences might “mouth grabs” were gentle and seemingly affili- explain the absence of avoidance and the high rates ative, such grabs quickly can become a bite. It of interactive behavior observed during this study. remains unclear whether this was aggression or Next to habituation effects, species-specific a form of play behavior, but seals are robust, and life history and age-related factors might explain injury could easily occur. Grey seals often exhib- similarities and differences between species and ited such behaviors during intraspecific play–fight locations. Cowling et al. (2014) found for New behavior (Wilson, 1974; personal observations), Zealand fur seals that swimmers were preferably suggesting that this context is the same perceived approached by subadult animals rather than adults by the seals when interacting with humans. How- (see also Boren et al., 2009). They further found ever, for both cases, humans are at health risk that interaction with swimmers tended to be higher because they might experience physical inju- during the breeding season than did during the ries and subsequent infections (Hunt et al., 2008; postbreeding period. Boren et al. (2009) reported Waltzek et al., 2012). Several researchers described that swims with the same species at a different risky and nonrisky behaviors of seals during swim- location in the early breeding season resulted in with activities (Boren et al., 2009; Kirkwood et al., low interactions and high neutral response because 2003; Martinez, 2003; Muir et al., 2006; Osterrie- mainly adult males were in the water, which move der et al., 2017; Stafford-Bell et al., 2012). How- between territories. The present study was con- ever, these studies focused on the impacts on the ducted outside the breeding season, and age-related seals, were not specific to grey seals, and did not Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. 168 SCHEER

evaluate risky versus nonrisky behaviors initiated not commercialized. Though professional divers by seals towards humans. So far, only Dans et al. and snorkelers sometimes visit grey seals in the (2017) studied interactive behaviors during seal- water in summertime, seal-swims are mostly a by- swims with South American sea and found product of bathing. However, the increase of media high rates of bites and other contact behaviors. reports during the last years may contribute to an To foster predictability and management of increase in tourists seeking to swim with grey seals human–seal interaction, it is essential that swim- on site. Dune islet provides stable and safe habitat mers know behaviors displayed by seals and conditions for females to give birth, with a notice- addressed towards them. Swimmers with little or able increase in seal pups over the recent years (Abt no experience should be prepared before swims. & Engler, 2009). Thus, grey seal abundance in the On the other hand, swimmers should behave in area may increase, which also increases the poten- a noninvasive way as inappropriate behaviors tial of conflict with tourists. Grey seals in the area (i.e., chasing and encircling seals, inappropriate were recently found to prey on ( touching attempts) and noise disturbance could vitulina) pups during the summer months and to provoke harassment and avoidance among seals. be engaged in interspecific sexual aggression (van Overall, the code of conduct applied during this Neer, Jensen, & Siebert, 2015). Though the authors study seemed to establish good practice, as it did assumed that it is unlikely that grey seals will pose not elicit avoidance and let the animals choose to a serious risk to humans, it cannot be ruled out that, interact or ignore swimmers on their own. How- for example, breeding season frustration of sub- ever, physical contact between seals and swimmers adult males might elicit risky behaviors towards should be discouraged. Among marine , swimmers. touching or touching attempts by humans during Managers are confronted with a general para- close encounters in the water might elicit undesir- dox: some customers want to be as close as pos- able animal behaviors (Dans et al., 2017; Scheer, sible while others expect that the following 2010, for a review on cetaceans) and could provoke of regulations will protect them from any risk grabs or other risky behaviors in response. It is rec- (Knight, 2009; Wiener, 2013). As for all living ommended that swimmers should disengage from resources targeted by tourism, management of physical contact with grey seals by slowly moving swims with grey seals on Heligoland should find away from them. a compromise between the continued well-being Furthermore, grey seals were shown to respond of the resource and obtaining an economic benefit with rapid submerges when swimmers lifted their from it (Harwood, 2010). To ensure a potential bodies out of shallow water. Thus, swimmers tourism venture based on the grey seals, voluntary should be stipulated to remain low in the water codes of conduct and governmental regulations are not to startle seals and retreat to the beach slowly essential tools for its management. Research on when finished swimming. The small swimmer compliance, consumer satisfaction, and potential group sizes here (1–3), compliance with the code negative impacts on the resource are further ele- of conduct and the absence of inappropriate behav- ments to ensure sustainability (i.e., Curtin et al., iors and noise disturbance could explain the lack of 2009; Kirkwood et al., 2003). avoidance among grey seals. Swimming with marine mammals has been Similar to grey seals at UK watching sites (see discussed controversially. On the one hand, free- Curtin et al., 2009), the Heligoland animals are ranging animals have the choice to maintain or ter- predictable in time and space, easily approach- minate encounters. On the other hand, they might able and viewable, and are tolerant of human become habituated to human contact and nega- approaches. Only some disadvantages exist such tively impacted by encounters or indirect effects as limited underwater visibility (mostly <5 m dur- (see Allen, 2014, for a discussion). The risk of ing the study period) and low water temperatures injury to both humans and animals has been pointed (11–17°C for the period June to August). While out as a key problem for swim activities. As most land-based seal watching during wintertime has swimmers on site have no previous experience grown during the last years on site, seal-swims are with seals during swims (personal observation) Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. GREY SEAL BEHAVIOR TOWARD SWIMMERS 169

and seal-swims are not regulated or guided so far, of age/sex classes and individuals being involved swimmers may become overwhelmed by the close- in swim encounters would shed light on the number ness as well as by the diversity of behaviors they of animals being involved and on possible life his- may find themselves exposed to. Next to the risks tory and age-related aspects. A comparison of swim for swimmers, this might result in inappropriate encounters with grey seals at other locations might human behaviors (e.g., feeding attempts, harass- reveal site-specific effects. ment) and could increase the risks for seals as well. To reduce the risk of injury to seal and human, it is recommended to implement recently proven Conclusion signage messages and placements to promote pub- Grey seals off Heligoland are highly interac- lic awareness of the behaviors and capabilities tive with human swimmers. Next to nonaggressive of the seals (Barton, Booth, Ward, Simmons, & and affiliative behaviors, they display high rates Fairweather, 1998; Martin, Momtaz, Jordan, & of risky behaviors during in-water encounters. Moltschaniwskyi, 2015) and to communicate a vol- Though seal swims are not commercialized to this untary code of conduct (Öqvist et al., 2018). Physi- date and swim encounters are in most cases a by- cal contact between swimmers and seals should be product of bathing, the growing birth rates among discouraged. The onset of guides has been shown local seals, and the rising numbers of tourists on to reduce disturbance as negative behaviors of Dune islet might introduce severe wildlife con- tourists towards seals can be stopped and avoided flicts. It is highly recommended to promote public (Acevedo-Gutiérrez, Acevedo, & Boren, 2011; awareness on site and to regulate seal swims before Boren et al., 2009). Guidance on Dune islet could commercial operations emerge. be ensured by lifeguards, seal rangers, or volunteers (see Le Boeuf & Campagna, 2013). It is further rec- ommended to implement governmental regulations Acknowledgments to control swimmer behavior among common tour- The field study was supported by the community ists and specifically among participants of com- of Heligoland and conducted in collaboration with mercial programs. The introduction of a licensing the NGO Jordsand e.V. Thanks to Jörg Singer, Klaus system for commercial operators, the prohibition of Furtmeyer, Michael Janßen, Dr. Eckart Schrey, and feeding, and limited access for swimmers to interact Rainer Borcherding. Ralf Jonas, Anke Marth, Leon with animals could be elements for governmental Marth, Ira Marth, Abbo van Neer, and Angela Ljil- control (Gales, 1995; Newsome & Rodger, 2008). janic supported fieldwork and data sampling. Dr. It has been shown that compliance with regulations Jaume Forcada, Dr. Roger Kirkwood, Fabian Ritter, is critical (Scarpaci, Nugegoda, & Corkeron, 2004; and Chantal Pagel improved the quality of early Strong & Morris, 2010). Thus, effective enforce- manuscript versions. Thanks to two anonymous ment and the sanctioning of undesirable human readers for peer reviewing this publication. behaviors should be further essential elements for governmental control. Biographical Note This study was conducted during a single 10-day Michael Scheer graduated as a biologist at the University of period at the onset of the primary tourist season in Bremen where he studied cetacean bioacoustics and behav- June. Additional data during the summer months ior. Michael’s research focused on the behavior of marine in July and August would be valuable as there mammals during touristic wildlife encounters. He recently might be more swimmers in the water due to higher worked as an independent scholar for the Institute for Terres- water temperatures. Future research should moni- trial and Aquatic Wildlife Research (ITAW) and the Depart- ment of Agriculture, Environment and Rural Areas (LLUR) tor swim encounters of tourists with grey seals on of Schleswig-Holstein on the management of grey seals and site such as the potential increase of risky behav- tourism. iors addressed towards tourists, the emergence of injuries among swimmers, and the possible onset or escalation of avoidance or other habituation ORCID effects among seals. The systematic identification Michael Scheer: https://orcid.org/0000-0001-5244-5042 Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. 170 SCHEER

References experiences, sustainability, management (pp. 93–113). Abt, K., & Engler, J. (2009). Rapid increase of the grey seal London, UK: Elsevier. (Halichoerus grypus) breeding stock at Heligoland. Hel- Curtin, S., Richards, S., & Westcott, S. (2009). Tourism and igoland Marine Research, 63, 177–180. grey seals in South Devon: Management strategies, vol- Abt, K., Hoyer, N., Koch, L., & Adelung, D. (2002). The untary controls and tourists’ perceptions of disturbance. dynamics of grey seals (Halichoerus grypus) off Current Issues in Tourism, 12, 59–81. in the south-eastern North Sea—Evidence of an open Dans, S. L., Crespo, E. A., & Coscarella, M. A. (2017). population. Journal of Sea Research, 47, 55–67. Wildlife tourism: Underwater behavioral responses of Acevedo-Gutiérrez, A., Acevedo, L., & Boren, L. (2011). South American sea lions to swimmers. Applied Animal Effects of the presence of official-looking volunteers on Behaviour Science, 188, 91–96. harassment of New Zealand fur seals. Conservation Biol- Gales, N. (1995). Hooker’s sea recovery plan (Pho- ogy, 25(3), 623–627. carctos hookeri) (Threatened Species Recovery Plan Allen, S. J. (2014). From exploitation to adoration. The his- Series). Wellington, New Zealand: Department of torical and contemporary contexts of human-cetacean Conservation. interactions. In J. Higham, L. Bejder, & R. Williams Gemeinde Heligoland. (2011). REK Heligoland 2010–2025: (Eds.), Whale-watching. Sustainable tourism and eco- Bestandsaufnahme und -analyse (Unpublished report). logical management (pp. 31–47). Cambridge, UK: Cam- Pinneberg, Germany. bridge University Press. Hall, A., & Thompson, D. (2009). Gray Seal - Halichoerus Altmann, J. (1974). Observational study of behavior: Sam- grypus. In W. F. Perrin, B. Würsig, & J. G. M. Thewissen pling methods. Behaviour, 49, 227–265. (Eds.), Encyclopedia of marine mammals (pp. 500–503). Barton, K., Booth, K. L., Ward, J. C., Simmons, D. G., & London, UK: Academic Press. Fairweather, J. R. (1998). Visitor and New Zealand fur Härkönen, T., Brasseur, S., Teilmann, J., Vincent, C., Dietz, seal interactions along the Kaikoura coast. Canterbury, R., Abt, K., & Reijnders, P. (2007). Status of grey seals New Zealand: Tourism Research and Education Centre along mainland Europe from the Southwestern Baltic to (TREC). France. NAMMCO Scientific Publications, 6, 57–68. Boren, L. J. (2001). Assessing the impacts of tourism on New Harwood, J. (2010). Approaches to management. In I. Boyd, Zealand fur seals (Arctocephalus forsteri) (Unpublished W. D. Bowen, & S. J. Iverson (Eds.), Marine master thesis). University of Canterbury, Christchurch, ecology and conservation (pp. 325–339). , NY: New Zealand. Oxford University Press. Boren, L. J., Gemmell, N. J., & Barton, K. J. (2002). Tour- Henry, E., & Hammill, M. O. (2001). Impact of small boats ist disturbance on New Zealand fur seals Arctocephalus on the haulout activity of harbour seals (Phoca vitulina) forsteri. Australian Mammalogy, 24, 85–95. in Métis Bay, Saint Lawrence Estuary, Québec, . Boren, L. J., Gemmell, N. J., & Barton, K. J. (2009). The Aquatic Mammals, 27, 140–148. role and presence of a guide: preliminary findings from Holcomb, K., Young, J. K., & Gerber, L. R. (2009). The swim with seal programs and land-based seal viewing influence of human disturbance on California sea lions in New Zealand. Tourism in Marine Environments, 5, during the breeding season. Animal Conservation, 12, 187–199. 592–598. Bosetti, V., & Pearce, D. (2003). A study of environmental Hunt, T. D., Ziccardi, M. H., Gulland, F. M. D., Yochem, conflict: The economic value of grey seals in southwest P. K., Hird, D. W., Rowles, T., & Mazet, J. A. K. (2008). England. Biodiversity & Conservation, 12, 2361–2392. Health risks for marine mammal workers. Diseases of Cassini, M. H. (2001). Behavioural responses of South Aquatic Organisms, 81, 81–92. American fur seals to approach by tourists—A brief Kirkwood, R., Boren, L., Shaughnessy, P., Szeren, D., Maw- report. Applied Animal Behaviour Science, 71, 341–346. son, P. R., Hückstädt, L. A., Hofmeyr, G., Oosthuizen, Constantine, R. (1999). Effects of tourism on marine mam- H., Schiavini, A., Campagna, C., & Berris, M. (2003). mals in New Zealand (Science for Conservation No. 106). Pinniped-focused tourism in the Southern Hemisphere: Wellington, New Zealand: Department of Conservation. A review of the industry. In N. Gales, M. Hindell, & R. Constantine, R. (2001). Increased avoidance of swimmers Kirkwood (Eds.), Marine mammals: Fisheries, tourism by wild bottlenose dolphins (Tursiops truncatus) due and management issues (pp. 245–264). Collingwood, to long-term exposure to swim-with-dolphin tourism. Australia: CSIRO Publishing. Marine Mammal Science, 17, 689–702. Knight, J. (2009). Making wildlife viewable: Habituation Cowling, M., Kirkwood, R., Boren, L. J., & Scarpaci, C. and attraction. Society & Animals, 17, 167–184. (2014). The effects of seal-swim activities on the New Kovacs, K. M., & Inness, S. (1990). The impact of tourism Zealand ( australis fosteri) in the Bay on harp seals (Phoca groenlandica) in the Gulf of St. of Plenty, New Zealand, and recommendations for a sus- Lawrence, Canada. Applied Animal Behaviour Science, tainable tourism industry. Marine Policy, 45, 39–44. 26, 15–26. Curtin, S., & Garrod, B. (2008). Vulnerability of marine Le Boeuf, B., & Campagna, C. (2013). Wildlife viewing mammals to diving tourism activities. In B. Garrod & D. spectacles: Best practices from (Mirounga Gossling (Eds.), New frontiers in marine tourism: Diving sp.) colonies. Aquatic Mammals, 39, 2, 132–146. Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location. GREY SEAL BEHAVIOR TOWARD SWIMMERS 171

Lehner, P. N. (1987). Design and execution of animal behav- Osterrieder, S. K., Salgado Kent, C., & Robinson, R. W. ior research: An overview. Journal of Animal Science, (2017). Responses of Australian sea lions, 65, 1213–1219. cinerea, to anthropogenic activities in the Perth metro- Lovasz, T., Croft, D. B., & Banks, P. (2008). Establishing politan area, Western Australia. Aquatic Conservation: tourism guidelines for viewing Australian sea lions Neo- Marine and Freshwater Ecosystems, 27, 414–435. phoca cinerea at Seal Bay Conservation Park, South Pomeroy, P. P., Anderson, S. S., Twiss, S. D., & McCon- Australia. In D. Lunney A. Munn, & W. Meikle (Eds.), nell, B. J. (1994). Dispersion and site fidelity of breeding Too close for comfort: Contentious issues in human- female grey seals seal (Halichoerus grypus) on North wildlife encounters (pp. 225–232). Mosman, Australia: Rona, . Journal of Zoology, 233, 429–447. Royal Zoological Society of New South . Scarpaci, C., Nugegoda, D., & Corkeron, P. J. (2004). No Martin, P., & Bateson, P. (1993). Measuring behaviour: An detectable improvement in compliance to regulations introductory guide. Cambridge, UK: Cambridge Univer- by “swim-with-dolphin” operators in Port Phillip Bay, sity Press. Victoria, Australia. Tourism in Marine Environments, 1, Martin, C. L., Momtaz, S., Jordan, A., & Moltschaniwskyj, 41–48. N. A. (2015). An assessment of the effectiveness of in- Scarpaci, C., Nugegoda, D., & Corkeron, P. J. (2005). Tour- situ signage in multiple-use marine protected areas in ists swimming with Australian fur seals (Arctocephalus providing information to different recreational users. pusillus) in Port Phillip Bay, Victoria, Australia: Are Marine Policy, 56, 78–85. tourists at risk? Tourism in Marine Environments, 1, Martinez, A. (2003). Swimming with sea lions: Friend or 89–95. foe? Impacts of tourism on Australian sea lions, Neo- Scheer, M. (2010). Review of self-initiated behaviors of phoca cinerea, at Baird Bay, SA (Unpublished master free-ranging cetaceans directed towards human swim- thesis). Flinders University of South Australia, Adelaide, mers and waders during open water encounters. Interac- Australia. tion Studies, 11, 442–466. McLaren, I. A. (1993). Growth in pinnipeds. Biological Stafford-Bell, R., Scarr, M., & Scarpaci, C. (2012). Behav- Reviews, 68, 1–79. ioural responses of the Australian fur seal (Arctocepha- Moscardo, G., Taverner, M., & Woods, B. (2006). When lus pusillus doriferus) to vessel traffic and presence wildlife encounters go wrong: Tourist safety issues asso- of swimmers in Port Phillip Bay, Victoria, Australia. ciated with threatening wildlife. In Y. Mansfeld & A. Aquatic Mammals, 38, 241–249. Pizam (Eds.), Tourism, security and safety: From theory Strong, P., & Morris, S. R. (2010). Grey seal (Halichoerus to practice (pp. 209–227). London, UK: Elsevier. grypus) disturbance, ecotourism and the Pembrokeshire Muir, S., Barnes, D. K. A., & Reid, K. (2006). Interactions Marine Code around Ramsey Island. Journal of Ecotour- between humans and leopard seals. Antarctic Science, ism, 9, 117–132. 18, 61–74. van Neer, A., Jensen, L. F., & Siebert, U. (2015). Grey seal Newsome, D., & Rodger, K. (2008). Impacts of tourism on (Halichoerus grypus) predation on harbour seals (Phoca pinnipeds and implications for tourism management. In vitulina) on the island of Heligoland, Germany. Journal J. Higham & M. Lück (Eds.), Marine wildlife and tour- of Sea Research, 97, 1–4. ism management—Insights from the natural and social Vauk, G. (1978). Erstnachweis der Kegelrobbe (Halichoerus sciences (pp. 182–205). , UK: Biddles Ltd. grypus) für Heligoland. Zeitschrift für Jagdwissenschaft, O’Connor, S., Campbell, R., Cortez, H., & Knowles, T. 24, 44–45. (2009). worldwide: Tourism numbers, Waltzek, T. B., Cortés-Hinojosa, G., Wellehan Jr., J. F. X., & expenditures and expanding economic benefits. Yar- Gray, G. C. (2012). Marine mammal zoonoses: A review mouth, MA: International Fund for Animal Welfare. of disease manifestations. Zoonoses and Public Health, Öqvist, E. L., Granquist, S. M., Burns, G. L., & Angerbjörn, 59, 521–535. A. (2018). Seal watching: An investigation of codes of Wiener, C. (2013). Friendly or dangerous waters? Under- conduct. Tourism in Marine Environments, 13, 1–15. standing dolphin swim tourism encounters. Annals of Orsini, J. P., Shaughnessy, P. D., & Newsome, D. (2006). Leisure Research, 16, 55–71. Impacts of human visitors on Australian sea lions (Neo- Wilson, S. (1974). Juvenile play of the common seal Phoca phoca cinerea) at Carnac Island, Western Australia. vitulina with comparative notes on the grey seal Hali- Tourism in Marine Environments, 3, 101–115. choerus grypus. Behaviour, 48, 37–60.

Delivered by Ingenta IP: 80.134.236.247 On: Sat, 17 Oct 2020 11:41:13 Article(s) and/or figure(s) cannot be used for resale. Please use proper citation format when citing this article including the DOI, publisher reference, volume number and page location.