Behavioral Ecology and Sociobiology (2018) 72:78 https://doi.org/10.1007/s00265-018-2492-8

ORIGINAL ARTICLE

Night-life of Bryde’s whales: ecological implications of resting in a baleen whale Sahar Izadi 1

&

Mark Johnson 2,3 & Natacha Aguilar de Soto 4 & Rochelle Constantine 1,5

Received: 28 July 2017 / Revised: 8 April 2018 / Accepted: 17 April 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract Many animals require intervals of rest or sleep in which their vigilance level is reduced. For marine fauna, including large baleen whales, resting potentially increases the risk of predation and vessel-strike. However, there is scarce information about how, and how often, whales rest which makes it difficult to assess the severity of this risk. Here we examine resting patterns of Bryde’s whales (Baleaenoptera edeni/brydei), using data collected by sound and movement archival tags (DTAGs) deployed on four whales in the Hauraki Gulf, New Zealand. To identify low activity levels associated with resting, we used RMS jerk and mean flow noise (as proxies for activity and speed, respectively), as well as changes in dive patterns (dive depth and shape), fluking, and respiration rates. The tagged whales showed strong diel differences in behavior with long periods of low activity consistent with resting occurring exclusively during the night. This pattern indicates that either (i) Bryde’s whales rely on senses that are less effective in the dark to locate prey, or (ii) that prey aggregate less densely at night, making foraging less efficient. Thus, Bryde’s whales conserve energy through rest during times when the net benefit of foraging effort is low. However, by reducing their interaction level with their environment, night-time resting also makes Bryde’s whales more vulnerable to vessel strikes, an important source of mortality for cetaceans. Significant statement All mammals need to rest periodically and whales are no exception. But while resting land mammals can be observed directly, little is known about when and how whales rest; even though lower vigilance levels during resting could make them more vulnerable to threats such as collisions with boat traffic. We used sound and movement logging tags on resident Bryde’s whales in a busy gulf to study their daily activity patterns. We found that, while whales were active during daytime making energetic lunges to capture tonnes of plankton, they dedicated much of the night to rest. This suggests that whales may rely on vision to find prey or that prey are less densely aggregated at night making foraging less efficient. However, this near-surface resting behavior which may also be shared by the other giant baleen whales increases the risk of ship strikes. Keywords Rest . Sleep . Whales . Diel behavior . Accelerometry . Biologging

Communicated by L. Rendell * Sahar Izadi [email protected] 1

Institute of Marine Science, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand

2

Sea Mammal Research Unit, University of St Andrews, Fife, Scotland KY16 8LB, UK

3

Department of Bioscience, Aarhus University, Aarhus, Denmark

4

BIOECOMAC, La Laguna University, Tenerife, Canary Islands, Spain

5

School of Biological Sciences, University of Auckland, Private Bag 92019, Auckland 1142, New Zealand

Introduction Many animals periodically enter a phase of inactivity. This behavior may combine resting with increased vigilance for threats or could form part of a strategy to minimize transmitting cues to predators (Lima et al. 2005; Siegel 2011). However, inactive intervals may also be sleeping periods when animals can be most vulnerable due to the adoption of less-responsive brain states (Lima et al. 2005). Animals use a number of strategies to decrease their vulnerability to predation when resting or sleeping: monkeys and apes rest at night (to avoid producing acoustic cues when their feline predators are active) and in safe places (Anderson 2000). Some social

78

Page 2 of 12

animals, such as green-winged teals (Anas crecca crecca) (Gauthier-Clerc et al. 1998) and harbor seals (Phoca vitulina concolor) (da Silva and Terhune 1998), aggregate in larger groups to increase overall vigilance while they sleep. Some species of mammals, birds, and reptiles have developed a remarkable solution to maintaining vigilance during resting intervals by sleeping with one cerebral hemisphere while the other hemisphere is awake (Rattenborg et al. 2000); this is known as unihemispheric sleep. Cetaceans, pinnipeds (except for true seals), and sirenians are the only mammals known to sleep unihemispherically. For these voluntary breathers, unihemispheric sleep is essential to ensure continued and safe respiration in the water, and to maintain body position and thermogenesis. Unihemispheric sleeping may also allow animals to keep watch on offspring and group members (Lyamin et al. 2008). Irrespective of its benefits, resting and sleeping also come at the expense of other critical behaviors such as socializing and foraging. The timing and duration of inactive states for both active hunting species and their prey are often driven by ecological interactions, with food availability, presence of predators and competitors, and environmental conditions (such as light levels or temperature) being the most important ecological factors influencing circadian rhythms (KronfeldSchor and Dayan 2003). For example, the timing of resting bouts of predators may provide an indication of prey diel behavior with inactivity timed to coincide with periods in which prey, or the cues used to hunt prey, are unavailable, e.g., during the night for most visual predators. Detailed data on the behavior, physiology, and ecological correlates associated with resting are difficult to collect in the wild and, as a consequence, this behavior has mainly been studied in captive animals (Amlaner and Ball 1983; Campbell and Tobler 1984; Roth et al. 2006; Siegel 2008; Lesku et al. 2009). Studies on captive dolphins and porpoises h a v e p r o v i d e d el e c t r o p h y s i o l o g i c al e vi d e n c e o f unihemispheric sleep (e.g., Serafetinides et al. 1972; Mukhametov et al. 1977; Mukhametov 1987; Lyamin et al. 2002) but logistical difficulties currently preclude measurements of electroencephalographic (EEG) activity in wild marine mammals. Behavioral observations of some toothed whales both in captivity and in the wild have revealed sleeplike behavior (e.g., Goley 1999; Watkins et al. 1999; Constantine et al. 2004; Sekiguchi et al. 2006; Miller et al. 2008; Ford 2009; Shpak et al. 2009), which has been interpreted as unihemispheric sleep. For the largest mammals, the baleen whales, our knowledge of resting is limited to occasional behavioral observations with no EEG record as they are more difficult to keep in captivity. Lyamin et al. (2001) observed resting in a 1-year old gray whale (Eschrichtius robustus) calf briefly kept in captivity and concluded that baleen whales can sleep unihemispherically, both at the surface and underwater. Field studies of some baleen whale

Behav Ecol Sociobiol (2018) 72:78

species have also reported behavioral periods in which animals breathe slowly, log at the surface, or move very slowly near the surface; all of which are interpreted as resting (e.g., Dolphin 1987; Christiansen et al. 2015). Here we investigate the resting behavior of Bryde’s whale in an area of dense vessel traffic, the Hauraki Gulf, where resting behavior may enhance risk of ship strike for these whales (Constantine et al. 2015). The Hauraki Gulf is a large bay with depths around 50 m, located on the north east coast of New Zealand (Paul 1968). The inner Hauraki Gulf is the primary habitat for a small and endangered year-round resident population of Bryde’s whales (Baker et al. 2010). Unlike most baleen whales, Bryde’s whales do not undertake long seasonal migrations linked to distinct habitat requirements for breeding and feeding. Instead, they remain in warmtemperate waters year-round (Kato and Perrin 2009). While the pelagic populations of Bryde’s whales perform limited migrations between equatorial and temperate latitudes, the inshore populations stay in coastal waters year-round (Tershy 1992; Zerbini et al. 1997; Best 2001). As a consequence of not migrating, the foraging behavior of inshore Bryde’s whales differs from that of migratory whales: they feed constantly throughout the year taking advantage of year-round populations of prey, or switching seasonally between prey types, rather than exploiting the seasonal abundance seen in the summer polar feeding grounds (Murase et al. 2007; Kato and Perrin 2009; Penry et al. 2011). Bryde’s whales in the Gulf are most commonly observed singly or in small groups of two to three whales, and have a variable diet, feeding on zooplankton and small schooling fishes (e.g., pilchards, Sardinops sagax) (Jarman et al. 2006; Baker and Madon 2007; Wiseman 2008). Despite their cosmopolitan distribution, relatively little is known about the behavior of Bryde’s whales. As with other rorqual whales, Bryde’s lunge-feed on aggregations of prey at the surface (Zerbini et al. 1997; Baker and Madon 2007; Penry et al. 2011). However, when not feeding at the surface, the inconspicuous and brief surfacings of Bryde’s whales make observational studies challenging. By excluding sub-surface and nighttime activities, visual observations are able to sample only a small proportion of Bryde’s whale behavior. Time-depth recording tags (TDRs) deployed on Bryde’s whales in deep waters off Madeira, Portugal, showed regular diving to depths of 40–300 m, with some foraging in deep waters. Also, whales showed a diel pattern in diving behavior and this was interpreted as possibly correlated with zooplankton diel distribution (Alves et al. 2010). However, the limited sensor suite on TDR tags makes it challenging to distinguish activities such as slow travel or prey search from resting. The broader array of sensors on sound and movement archival tags (DTAGs) (Johnson and Tyack 2003; Johnson et al. 2009) provides data on orientation, locomotion, lunging, and sound production that can help to define active and inactive phases

Behav Ecol Sociobiol (2018) 72:78

(e.g., Nowacek et al. 2004; Goldbogen et al. 2006; Aguilar Soto et al. 2008; Simon et al. 2012; Ydesen et al. 2014). Importantly, these data can be obtained irrespective of the time of day enabling studies of diel patterns in whale behavior. In this study, we applied multi-sensor DTAGs to characterize the resting behavior and its diel patterns of occurrence in Bryde’s whales using accelerometer and acoustic data. The results delimit the environmental factors driving resting behavior of Bryde’s whales in the Hauraki Gulf, and contribute to our understanding of the ship strike risk for this species, with implications for the design of management measures to reduce ship strike mortality of the Bryde’s whale.

Methods We deployed suction-cup attached multi-sensor tags (DTAG version 2, Johnson and Tyack 2003) on six Bryde’s whales in the Hauraki Gulf, New Zealand (36°10′-37°10′S, 175°30′E). These tags have a frontal area of 30 cm2, which is << 1% of Bryde’s whale frontal area, and so likely have little impact on the drag of the animal. The tags are equipped with triaxial magnetometers and accelerometers, as well as pressure and temperature sensors and two hydrophones. The sensors and hydrophones are sampled synchronously at 50 Hz and 96 kHz, respectively, with 16 bit resolution and all data are recorded to a 16-GB solid-state memory using loss-less compression (Johnson et al. 2013). Whales were located while conducting surveys on a 15-m research vessel, the RV Hawere. After a whale was spotted, a 5-m inflatable boat with 60 hp four-stroke engine was launched to approach the whale for tagging. Tags were deployed on the dorsum of the whales as they surfaced to breathe using a 6-m hand-held carbon-fiber pole. After deploying a tag, the small boat returned to the RV Hawere which then maintained a distance of > 300 m from the whale during the remaining daylight hours to record its surface behavior and position on a computer time-synchronized with the DTAG. The tag was retrieved using VHF radio tracking after it detached from the whale. The recorded data were then downloaded for analysis and the tag was recharged for redeployment. Blind data recording was not possible because our study involved focal animals in the field.

Data preparation Analyses were restricted to four of the six tag deployments, which had more than 2 h of recording before and after sunset to allow characterization of diel patterns in whale behavior. Custom software (www.soundtags.org) in MATLAB (Version 8, MathWorks) was used to calibrate and analyze DTAG data following the methods in Johnson and Tyack (2003). Sensor data were first decimated to a common sampling rate of 25 Hz and the

Page 3 of 12 78

triaxial magnetometer and accelerometer data were then corrected for the initial tag position on the animal and for any subsequent orientation changes due to tag movement during the deployment. Histograms of dive depth showed a clear separation between shallow submergences between respirations (< 5 m depth) and deeper departures from the surface. Dives were accordingly defined as a submergence deeper than 5 m, which corresponds to 30– 50% of the body length of a Bryde’s whale. The accelerometer and magnetometer data were used to determine the orientation of the whale as a function of time, parameterized by the Euler angles pitch, roll, and heading. The acceleration data were also differentiated to calculate acceleration rate or jerk (unit: ms−3) generated by sudden movements of the whales. As the magnitude of acceleration signals depends on the position of the tag on the body, which varies among deployments, jerk levels cannot be compared across animals and are used here to provide a relative measure of time-varying activity levels within individuals. The RMS of the norm of jerk was calculated over contiguous 5-min blocks (Ydesen et al. 2014). This averaging time was selected as a trade-off between reducing serial correlation in the measurements while still keeping enough time resolution to capture short-term changes in behavior. Since water movement over the tag at the surface can produce high accelerations, jerk values within 1 m of the surface were excluded from these calculations. The same 5-min blocks were used to measure a number of other parameters to evaluate the behavior of the whales. Low frequency sound recorded by the tag was used as a proxy for flow noise generated during active whale behaviors such as traveling and foraging (Nowacek et al. 2004; Goldbogen et al. 2006). To calculate this proxy, we applied a 500-Hz low-pass filter to the sound data and then computed the RMS power in the filtered signal over contiguous 40-ms blocks. This resulted in a noise power time sequence at the same sampling rate as the sensors. Like jerk data, noise values within 1 m of the surface were excluded from the analysis. The mean of the remaining flow noise values over 5-min blocks was used as a relative indicator of the whale’s speed.

Activity levels When whales are actively foraging or traveling, we expect high but variable levels of jerk and flow noise (e.g., Goldbogen et al. 2006; Aguilar Soto et al. 2008; Simon et al. 2012), while in resting intervals distinctly lower and more constant values of these parameters are expected. We used histograms of the log-transformed jerk and flow noise to identify switches in activity level and to determine thresholds for separating the data into high and low activity intervals. Since jerk and flow noise levels are dependent on the tag position on the whale, these thresholds varied slightly for each

78

Behav Ecol Sociobiol (2018) 72:78

Page 4 of 12

individual. Although tag movement during a deployment can lead to a varying relationship between jerk or flow noise levels and activity for each animal, we did not attempt to correct for this effect due to a lack of a reference behavior in low activity intervals. The impact of such tag movement on our results will be discussed later. Tail strokes cause cyclic variations in the body posture, specifically the pitch angle for a cetacean, and these can be sensed with an animal-attached accelerometer (Martin Lopez 2016). As the accelerometer x-axis (i.e., caudorostral axis) signal varies with the sine of the pitch angle (Johnson and Tyack 2003), tail strokes give rise to oscillations in this signal at the stroking rate. Fluking rates (tail strokes per second) were therefore calculated by first applying a delay-free high-pass filter (cut-off frequency of 0.075 Hz, approximately one half of the nominal stroking rate) to the accelerometer x-axis accelerometer data and then counting the positive zero crossings in the filtered signal using a hysteretic detector (Simon et al. 2012). Mean fluking rates were calculated over 5-min blocks. Individual respirations produce distinctive breathing sounds in the tag audio recording when the surfacing noise is low. Respirations are also associated with brief surfacing events, which are evident in the dive profile and this was a more robust way to detect breathing. Here we characterized respirations as intervals in which the whale approaching within 1 m of the surface and then submerged below 2 m shortly afterwards. Respiration rates were calculated for the 5-min analysis blocks. To characterize the behavior of whales during high and low activity phases, dive depth and respiration rate over the 5-min intervals were examined as response variables in linear mixed models with random intercept using the package lme4 (Bates et al. 2015) in R (R Core Team 2018). Mixed m o d e l s w e r e c h o s e n t o m i n i m i z e t h e e ff e c t s o f pseudoreplication on the results. Activity level (a binary variable indicating low or high activity) and individual whale were considered as fixed and random factors, respectively. For each model, the F statistic and p value were obtained using the Anova function in package car (Fox and Weisberg 2011). Since the distribution of the maximum depth was rightskewed and so violated the assumption of normal distribution of the residuals, a log transform was applied to this response variable; after fitting, the results were back transformed, and ratios were used for interpretation. To explore the timing of active and inactive intervals, activity level was compared to light level by dividing the 24-h day into daytime, nighttime, and dusk with the latter defined as an hour window around local sunset and sunrise times. Data availability statement The datasets during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Results The six tags remained attached to the whales for a total of 61 h, with four deployments including more than 2 h of both day and night recordings. These four recordings lasted from 6.5 to 19.7 h, summing 56 h (Table 1) of which 57% was at night. Examination of the RMS jerk and flow noise in these deployments suggested that whales switched between two distinct levels of activity (Fig. 1). This was confirmed by plotting histograms of log-transformed RMS jerk and flow noise for each individual (Fig. 2). For three of the four whales, these histograms were bimodal, with clear thresholds separating intervals of high and low activity for each individual. The fourth whale (221b) spent a large proportion of the tag deployment in an apparent low activity state and this led to an indistinct high activity mode in the histograms of jerk and flow noise for this whale. The average of the thresholds determined for the other whales was used to separate activity levels for this whale. Visual assessment of the resulting activity state designations indicated that this was a reasonable approach. Applying the RMS jerk and flow noise thresholds for each whale, we labeled each 5-min analysis interval as low activity if both parameters were lower than their respective individual thresholds, and as high activity otherwise. The detected low activity intervals were almost continuous, covering many hours in all whales (Fig. 3, Table 1). As well as RMS jerk and mean flow noise, the switch in activity levels was associated with changes in dive shape and depth, posture, and fluking and respiration rates (Figs 4, 5). Dive depths during active intervals were widely variable and dives frequently contained foraging U-shape dives (Goldbogen et al. 2013) or V-shaped transitions likely due to foraging or traveling (Alves et al. 2010; Parks et al. 2011). In comparison, dives during inactive periods were generally smoothly U-shaped with extensive bouts of dives to the same general depth. The maximum dive depth was significantly related to activity level (F1,1128.7 = 11.09, p < 0.001) albeit with a small effect size: active dives were 1.12 times deeper than inactive ones (Table 2). Body posture during low activity period dives was horizontal with little variation except during brief dive descents and ascents (Fig. 5). This contrasted with body posture in active intervals, which varied widely due primarily to sudden orientation changes in feeding lunges. Periods of reduced activity were accompanied by a reduction in the whales’ respiration rate. Inactive respiration rates dropped significantly, 0.48 breaths-per-minute (bpm), compared to when the whales were active (F 1,611.55 = 401, p < 0.0001) (Table 2). The mean fluking rate also decreased from 0.07 (SD = 0.04) per second during active intervals to 0.02 (SD = 0.01) per second during low activity periods, although this change was not modeled and tested for significance as fluking rate is linked to both flow noise and jerk, and so cannot be considered an independent variable.

Behav Ecol Sociobiol (2018) 72:78 Table 1

Page 5 of 12 78

Summary of four Bryde’s whale DTAG deployments and the percentage of time spent resting throughout the day in the Hauraki Gulf

Date

09 Aug 11 22 Aug 11 26 Aug 11 28 Sept 11

Tag ID

221b 234a 238a 271a

Tag on

14:51 11:18 10:54 15:24

Tag off

11:05 23:58 06:13 21:58

Total time (hh:mm)

17:30 12:40 19:41 06:32

Day

Dusk

Night

TT (hh:mm)

% RT

TT (hh:mm)

%RT

TT (hh:mm)

%RT

3:00 6:05 6:52 3:28

11 0 0 0

2:00 1:00 1:00 1:00

75 0 0 1

12:30 5:35 11:49 2:04

100 75 81 92

Total time represents the time period analyzed. TT is the total recorded time and %RT is the percentage of recorded time spent resting

However, the low fluking rate confirms that the inactive intervals involve little horizontal movement. The co-occurrence of substantial reduction in movement, respiration rate, and postural variation led us to interpret low activity periods primarily as resting behavior. Rest occurred almost entirely at night, with 94% of the detected resting intervals occurring during the hours from 30 min after sunset to 30 min before sunrise, and 5% of low activity intervals happening at dusk. The remaining 1% of resting intervals were recorded during the day in only one of the tagged whales (221b). Whales spent 75–100% of the recorded night hours in the low activity state (Fig. 3, Table 1), making only occasional short bouts of active dives in between rest periods during the night.

Discussion In the past 50 years, the application of biologging tools has helped researchers to answer questions about the behavior of wild animals that could not be investigated by direct observation (Rutz and Hays 2009). Nonetheless, many aspects of the behavior of the world’s largest mammals, the baleen whales, remain to be described. Here we used data from suction-cup Fig. 1 A section of one whale’s activities (234a). The upper graph shows the dive profile over a 7-h period covering day, dusk, and night time, and the lower graph shows the corresponding RMS jerk and mean flow noise level (relative to the clipping level of the recorder). Dark gray and light gray areas represent night and dusk, respectively

attached sound and movement logging tags to investigate the resting behavior and diel activity levels of one of the least studied baleen whales, the Bryde’s whale, in an area with frequent ship traffic. Rest or sleep is essential for all mammals and we found that Bryde’s whales in the Hauraki Gulf are no exception. The tag data revealed a strong diel activity pattern with active behaviors such as traveling and foraging during the day, and a less active state indicative of rest during nighttime. Although tag durations in this study were too short to estimate the overall activity time budget of the whales reliably, the apparent close link between resting and nighttime suggests that Bryde’s whales in this location may spend a high proportion of the night resting. A major challenge in biologging is to infer behavioral modes and physiological states reliably from the sensors available in a tag. As there is often no independent observation of activities that occur below the surface, alternative explanations for patterns in the sensor data must be considered carefully. Here we used RMS jerk and low frequency noise as proxies for rapid movement and swimming speed, respectively, to infer activity level. Flow noise, jerk, and other high-pass filtered acceleration measures such as the dynamic acceleration have been used in other studies to detect behavioral states such as foraging and traveling (e.g., Acevedo-Gutierrez et al.

78

Page 6 of 12

Fig. 2 Log histograms of RMS jerk and mean flow noise level from four Bryde’s whales. Dashed vertical lines indicate the detection thresholds used to separate active and inactive behaviors

Fig. 3 Dive profiles and activity levels for four Bryde’s whales. Detected inactive intervals are in red. Shaded areas represent night (black) and dusk (gray)

Behav Ecol Sociobiol (2018) 72:78

Behav Ecol Sociobiol (2018) 72:78

Page 7 of 12 78

Fig. 4 Movement parameters of four Bryde’s whales: (a) RMS jerk (m/s3), (b) mean flow noise level, (c) fluking rate (stroke/s), and (d) respiration rate (bpm). Horizontal dashed lines in (a) and (b) indicate

the thresholds used to distinguish active and inactive behaviors. Shaded areas represent night (black) and dusk (gray)

2002; Johnson et al. 2004; Goldbogen et al. 2008; Alves et al. 2010; Tyson et al. 2012; Friedlaender et al. 2013). Several factors influence the reliability of these as activity proxies. Jerk and flow noise are influenced by tag position on the whales’ body which prevents both standardization of these parameters as well as inter-individual comparisons. Thus, it is not immediately clear if a lower level of jerk on one animal as compared to another is an indication of a difference in behavior or simply a consequence of a different tag position. Also, rear-ward sliding of tags during a deployment can lead to step changes in jerk and flow noise for the same activity level. Finally, noise from ship traffic near a tagged whale can be mistaken for increased flow noise. None of these factors explain the observed patterns in RMS jerk and flow noise in our study: All animals showed intervals of high and low activity, with large changes in jerk and flow noise making differences in tag location across and within animals an

implausible explanation. Suction-cup attached tags can occasionally slide along the body during deployments, but such movements are usually small (large tag movements on the body generally lead to rapid detachment) and are almost always caudal thus causing an increase in the apparent activity (jerk and flow noise tend to increase as the tag approaches the caudal peduncle due to the larger magnitude propulsive movements). As the low activity intervals occurred after tags had been attached for some hours and showed consistency throughout the deployments, they could not be a result of sliding. Finally, ship traffic is present day and night in the Hauraki Gulf but is not dense enough to greatly impact the flow noise measure. To help interpret the behavioral mode, we used several parameters that are independent of the activity proxies, specifically, dive shape and depth, and respiration rate. We also calculated fluking rate which is derived from the same

78

Behav Ecol Sociobiol (2018) 72:78

Page 8 of 12

Fig. 5 Detailed view of dive profile and accelerometer data in one whale (234a). The upper graph shows the dive profile with detected low activity intervals in red. The lower graphs show an expanded view of the dive profile, pitch, and roll during 30 min in daylight (1300–1330, left) and night time (2300–2330, right). Shaded areas represent night (black) and dusk (gray)

Table 2 Linear mixed model and ANOVA test outputs for two response variables, respiration rate (breaths per minute—bpm), and maximum depth (m) in response to activity levels (active vs inactive) and with tags as random effect. Ratios for back transformed results are presented in parentheses Response variable

Respiration rate (bpm) Maximum depth (m)

Random effect

Tag residual Tag residual

Fixed effect

Variance

SD

0.13 0.07 0.03 0.17

0.36 0.27 0.18 0.41

SD standard deviation, SE standard error *Values are compared to the base level which is active

Intercept Activity level* Intercept Activity level*

ANOVA Estimate (ratio)

SE

t value

df

df.residual

1.08 − 0.48 2.27 (0.010) − 0.11 (1.12)

0.18 0.02 0.09 0.03

5.95 − 20.03 24.87 − 3.34

1

611.55

1

1128.7

F statistic

p value

401

< 0.0001

11.09

< 0.001

Behav Ecol Sociobiol (2018) 72:78

accelerometer signals, as is jerk, but which measures a different aspect of movement. These parameters are independent of tag location and so can help diagnose changes in jerk and flow noise due to tag movement. For example, in whales 234a and 238a, a few peaks in jerk and flow noise detected during low activity periods were not accompanied by elevated fluking and respiration rates. These peaks were most likely therefore due to contact of the tag with another animal or with material in the water, which may provoke a tag movement rather than brief bursts of activity. Although Bryde’s whales in the Hauraki Gulf have occasionally been observed in low activity modes during daytime (unpublished data), there is no indication from our data that this is common (only 1% of low activity periods were found during the day). Instead whales spent most of the daytime with high fluking and respiration rates consistent with traveling and foraging. Conversely, the nighttime drop in activity level is associated with very low rates of fluking and respiration precluding travel and lunge feeding. It is also unlikely to be explained by less active feeding methods such as continuous ram feeding used by right (Eubalaena spp.) and bowhead whales (Balaena mysticetus). Even though ram feeding occurs at low speeds, strong fluking is required to combat the increased drag of the open mouth (Simon et al. 2009). The baleen shape of rorquals and anatomy are ill-suited to this type of feeding (Pivorunas 1979; Croll et al. 2009). Although the literature is sparse, a number of factors appear to influence the timing and duration of rest in wild cetaceans. Daytime resting has been observed in Hawaiian spinner dolphins (Stenella longirostris) (Norris and Dohl 1980), bottlenose dolphins (Tursiops truncatus) (Constantine et al. 2004), and humpback whales (Megaptera novaeangliae) around Antarctica (Friedlaender et al. 2013). Nighttime resting has been reported for dusky dolphins (Lagenorhynchus obscurus) off Gulf of San Jose (Würsig and Würsig 1980) and minke whales (Balaenoptera acutorostrata) off the coast of northern Norway (Blix and Folkow 1995). Blix and Folkow (1995) also observed that minke whales off Svalbard did not show resting behavior in August when there was light 24 h a day whereas rest was observed in minke whales at lower latitudes during the same month. However, all of these visual observations were limited to surface periods and so likely underestimate resting. More recently, biologging tags recorded changes in diving and feeding behavior of blue whales at night (Calambokidis et al. 2007; Oleson et al. 2007; Goldbogen et al. 2011) and these were related in some cases with visual observations of surface resting (Oleson et al. 2007). By using multi-sensor tags in our study, we were able to measure activity levels equally well during day and night. Bryde’s whales in the Hauraki Gulf exhibited clear differences in day- and night-time activity. Their inactivity during nighttime may indicate a dependence on senses that are not as

Page 9 of 12 78

efficient in the dark. Although vision in cetaceans is generally well adapted to low levels of light underwater, cetaceans have also developed non-visual capabilities for prey detection (Mobley and Helweg 1990; Peichl et al. 2001; Griebel and Peichl 2003). A combination of olfaction, vision, and hearing could be used by whales to detect their patchily distributed preferred prey: zooplankton and schooling fishes. Bryde’s whales in the Gulf feed on fish aggregations that are simultaneously targeted by common dolphins (Delphinus delphis), Australasian gannets (Morus serrator), and shearwaters (Puffinus spp.) (Baker and Madon 2007). These highly active multi-species foraging events, termed Bwork-ups,^ generate noise over a range of frequencies from dolphin vocalizations, bird plunges, and fish jumps (Szczucka 2009; Benoit-Bird et al. 2011). Based on their inner ear morphology and vocalization frequencies, baleen whales are presumed to hear best at lower frequencies (< 5 kHz) (Ketten 2000; Nummela 2009). Since sounds at these frequencies can travel distances of several kilometers, work-up noise might be one of the factors that help Bryde’s whales to find their prey at distances greater than they can visually detect them. Birds are very dependent on vision (Hart 2001), so fewer work-ups form at night. Thus, if work-up noise is an important cue for Bryde’s whales (particularly when they feed on fish), it makes sense that they conserve energy through rest during times when this cue is unavailable. Another factor influencing the activity level of predators is the availability of prey. Rest times may coincide with times of the day in which favored prey are less easily accessible. Many zooplankton species undertake diel vertical migration, descending in daylight and ascending at night time or vice versa (Ohman et al. 1983; Lampert 1989; Andersen and Nival 1991; De Robertis 2000). Some cetaceans have been reported to follow the prey’s vertical migrations (Fiedler et al. 1998; Panigada et al. 2003; Calambokidis et al. 2007; Oleson et al. 2007; Goldbogen et al. 2011); however, the Hauraki Gulf, with an average depth of 40–45 m (Booth and Søndergaard 1989), is too shallow for significant zooplankton vertical migration. But, it is possible that some prey species of Bryde’s whales have diel patterns of movement, such as scattering and aggregating, which influences the activity levels of the whales. In either case, the long rest intervals observed here may not have a great energetic cost. As a non-migratory species, Bryde’s whales do not need to accumulate large energy reserves so as to survive several months of fasting as migratory rorquals do. Instead, Bryde’s whales are generalist foragers living in warm-temperate waters and this may allow them to rest on a daily basis with little consequence on body condition. Sleep appears to be a critical behavioral state for all mammals (Rechtschaffen 1998; Rattenborg et al. 2000; Siegel 2005, 2008), with prolonged deprivation leading to compromised function, loss of alertness and, under extreme

78

Page 10 of 12

situations, death (Rechtschaffen 1998). Our data did not allow us to investigate sleep stages and types in Bryde’s whales, but it seems likely that the tagged animals spent part of their prolonged rest intervals asleep. If so, it is most likely that Bryde’s whales sleep unihemispherically as found for smaller cetaceans in captivity (Rattenborg et al. 2000) so as to control respiration (which is voluntary in cetaceans) and maintain body position in the water column. But regardless of whether Bryde’s whales rest or sleep, this behavior changes their interaction level with their environment. As it affects their level of consciousness, it makes them more vulnerable to predators and human-related threats such as vessel strikes. This may explain the high level of ship strike mortality found in the Hauraki Gulf (Constantine et al. 2015) prior to voluntary mitigation measures were established. Thus, having a better understanding of Bryde’s whale resting behavior enables a better understanding of their ecology and effective conservation action planning. Acknowledgments We thank the Marine Alliance for Science and Technology Scotland, the field assistants from the Marine Mammal Ecology Group, RV Hawere and Dolphin Explorer crews, Phil Brown, Vivian Ward, Tim Higham, Kevin Chang, and tāngata whenua of the Gulf. We also acknowledge the anonymous reviewers for careful and thorough review of our manuscript and for their useful comments. Funding This work was supported by the Auckland Council, the Department of Conservation, the University of Auckland including a Doctoral Scholarship to SI, and a Marie Curie Post-Doctoral Fellowship to NAS.

Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. Ethical approval The research was conducted in accordance with permits from the University of Auckland Animal Ethics Committee (Permit #AEC03/2008/R636 and 910) and the New Zealand Department of Conservation (Permits #PER02/2009/01) issued to RC.

References Acevedo-Gutierrez A, Caroll DA, Tershy BR (2002) High feeding costs limit dive time in the largest whales. J Exp Biol 205:1747–1753 Aguilar Soto N, Johnson MP, Madsen PT, Diaz F, Dominguez I, Brito A, Tyack P (2008) Cheetahs of the deep sea: deep foraging sprints in short-finned pilot whales off Tenerife (Canary Islands). J Anim Ecol 77:936–947 Alves F, Dinis A, Cascão I, Freitas L (2010) Bryde’s whale (Balaenoptera brydei) stable associations and dive profiles: new insights into foraging behavior. Mar Mammal Sci 26:202–212 Amlaner CJ, Ball NJ (1983) A synthesis of sleep in wild birds. Behaviour 87:85–119 Andersen V, Nival P (1991) A model of the diel vertical migration of zooplankton based on euphausiids. J Mar Res 49:153–175

Behav Ecol Sociobiol (2018) 72:78 Anderson JR (2000) Sleep-related behavioural adaptations in freeranging anthropoid primates. Sleep Med Rev 4:355–373 Baker AN, Madon B (2007) Bryde’s whales (Balaenoptera cf. brydei Olsen 1913) in the Hauraki Gulf and northeastern New Zealand waters, Science for Conservation, vol 272. Department of Conservation, Wellington Baker CS, Chilvers BL, Constantine R, DuFresne S, Mattlin RH, van Helden A, Hitchmough R (2010) Conservation status of New Zealand marine mammals (suborders Cetacea and Pinnipedia). New Zeal J Mar Freshw Res 44:1–15 Bates D, Maechler M, Bolker B, Walker S (2015) Fitting linear mixedeffects models using lme4. J Stat Softw 67:1–48 Benoit-Bird KJ, Kuletz K, Heppell S, Jones N, Hoover B (2011) Active acoustic examination of the diving behavior of murres foraging on patchy prey. Mar Ecol-Prog Ser 443:217–235 Best PB (2001) Distribution and population separation of Bryde’s whale Balaenoptera edeni off southern Africa. Mar Ecol-Prog Ser 220: 277–289 Blix S, Folkow L (1995) Daily energy expenditure in free living minke whales. Acta Physiol Scand 153:61–66 Booth WE, Søndergaard M (1989) Picophytoplankton in the Hauraki gulf, New Zealand. New Zeal J Mar Freshw Res 23:69–78 Calambokidis J, Schorr GS, Steiger GH, Francis J, Bakhtiari M, Marshall G, Oleson EM, Gendron D, Robertson K (2007) Insights into the underwater diving, feeding, and calling behavior of blue whales from a suction-cup-attached video-imaging tag (Crittercam). Mar Technol Soc J 41:19–29 Campbell SS, Tobler I (1984) Animal sleep: a review of sleep duration across phylogeny. Neurosci Biobehav Rev 8:269–300 Christiansen F, Lynas NM, Lusseau D, Tscherter U (2015) Structure and dynamics of minke whale surfacing patterns in the Gulf of St. Lawrence, Canada. PLoS One 10:e0126396 Constantine R, Brunton DH, Dennis T (2004) Dolphin-watching tour boats change bottlenose dolphin (Tursiops truncatus) behaviour. Biol Conserv 117:299–307 Constantine R, Johnson M, Riekkola L, Jervis S, Kozmian-Ledward L, Dennis T, Torres LG, Aguilar de Soto N (2015) Mitigation of vesselstrike mortality of endangered Bryde’s whales in the Hauraki Gulf, New Zealand. Biol Conserv 186:149–157 Croll DA, Tershy BR, Newton KM (2009) Filter feeding. In: Perrin WF, Würsig B, Thewissen JGM (eds) Encyclopedia of marine mammals, 2nd edn. Thewissen, Academic Press, London, pp 421–425 De Robertis A (2000) Size-dependent visual predation risk and the timing of vertical migration: an optimization model. Limnol Oceanogr 47: 925–933 Dolphin WF (1987) Ventilation and dive patterns of humpback whales, Megaptera novaeangliae, on their Alaskan feeding grounds. Can J Zool 65:83–90 Fiedler PC, Reilly SB, Hewitt RP, Demer D, Philbrick VA, Smith S, Armstrong W, Croll DA, Tershy BR, Mate BR (1998) Blue whale habitat and prey in the Channel Islands. Deep-Sea Res Pt II 45: 1781–1801 Ford J (2009) Killer whale. In: Perrin WF, Würsig B, Thewissen JGM (eds) Encyclopedia of marine mammals, 2nd edn. Thewissen, Academic Press, London, pp 650–657 Fox J, Weisberg S (2011) An {R} companion to applied regression, 2nd edn. Sage, Thousand Oaks Friedlaender AS, Tyson RB, Stimpert AK, Read AJ, Nowacek DP (2013) Extreme diel variation in the feeding behavior of humpback whales along the western Antarctic Peninsula during autumn. Mar EcolProg Ser 494:281–289 Gauthier-Clerc M, Tamisier A, Cezilly F (1998) Sleep-vigilance trade-off in green-winged teals (Anas crecca crecca). Can J Zool 76:2214– 2218

Behav Ecol Sociobiol (2018) 72:78 Goldbogen JA, Calambokidis J, Shadwick RE, Oleson EM, McDonald MA, Hildebrand JA (2006) Kinematics of foraging dives and lungefeeding in fin whales. J Exp Biol 209:1231–1244 Goldbogen JA, Calambokidis J, Croll DA, Harvey JT, Newton KM, Oleson EM, Schorr G, Shadwick RE (2008) Foraging behavior of humpback whales: kinematic and respiratory patterns suggest a high cost for a lunge. J Exp Biol 211:3712–3719 Goldbogen JA, Calambokidis J, Oleson E, Potvin J, Pyenson ND, Schorr G, Shadwick RE (2011) Mechanics, hydrodynamics and energetics of blue whale lunge feeding: efficiency dependence on krill density. J Exp Biol 214:131–146 Goldbogen JA, Friedlaender AS, Calambokidis J, McKenna MF, Simon M (2013) Integrative approaches to the study of baleen whale diving behavior, feeding performance, and foraging ecology. Bioscience 63:90–100 Goley PD (1999) Behavioral aspects of sleep in Pacific white-sided dolphins. Mar Mammal Sci 15:1054–1064 Griebel U, Peichl L (2003) Colour vision in aquatic mammals—facts and open questions. Aquat Mamm 29:18–30 Hart NS (2001) Variations in cone photoreceptor abundance and the visual ecology of birds. J Comp Physiol 187:685–697 Jarman SN, Wiseman N, Baker CS, Gales NJ (2006) Incidence of prey DNA types in Bryde’s whale scats. SC/58/E29. Scientific Committee of the International Whaling Commission, St Kitts and Nevis Johnson MP, Tyack PL (2003) A digital acoustic recording tag for measuring the response of wild marine mammals to sound. IEEE J Ocean Eng 28:3–12 Johnson M, Madsen PT, Zimmer WMX, Aguilar de Soto NA, Tyack PL (2004) Beaked whales echolocate on prey. Proc R Soc Lond B 271: S383–S386 Johnson M, Aguilar de Soto N, Madsen P (2009) Studying the behaviour and sensory ecology of marine mammals using acoustic recording tags: a review. Mar Ecol-Prog Ser 395:55–73 Johnson M, Partan J, Hurst T (2013) Low complexity lossless compression of underwater sound recordings. J Acoust Soc Am 133:1387– 1398 Kato H, Perrin WF (2009) Bryde’s whales (Balaenoptera edeni/brydei). In: Perrin WF, Würsig B, Thewissen JGM (eds) Encyclopedia of marine mammals, 2nd edn. Thewissen, Academic Press, London, pp 158–163 Ketten DR (2000) Cetacean ears. In: WWL A, Popper AN, Fay RR (eds) Hearing by whales and dolphins. Springer, New York, pp 43–108 Kronfeld-Schor N, Dayan T (2003) Partitioning of time as an ecological resource. Annu Rev Ecol Evol Sci 34:153–181 Lampert W (1989) The adaptive significance of diel vertical migration of zooplankton. Funct Ecol 3:21–27 Lesku JA, Roth TC II, Rattenborg NC, Amlaner CJ, Lima SL (2009) History and future of comparative analyses in sleep research. Neurosci Biobehav Rev 33:1024–1036 Lima SL, Rattenborg NC, Lesku J, Amlaner CJ (2005) Sleeping under the risk of predation. Anim Behav 70:723–736 Lyamin OI, Mukhametov LM, Siegel JM, Manger PR, Shpak OV (2001) Resting behavior in a rehabilitating gray whale calf. Aquat Mamm 27:256–266 Lyamin OI, Mukhametov LM, Siegel JM, Nazarenko EA, Pollyakova LG, Shpak OV (2002) Unihemispheric slow wave sleep and the state of eyes in a white whale. Behav Brain Res 129:125–129 Lyamin OI, Manger PR, Ridgway SH, Mukhametov LM, Siegel JM (2008) Cetacean sleep: an unusual form of mammalian sleep. Neurosci Biobehav Rev 32:1451–1484 Martin Lopez ML (2016) Fine-scale kinematics and energetics: a detailed insight into the biomechanical strategies of cetacean locomotion. PhD thesis, The University of St Andrews, Scotland, UK Miller PJO, Aoki K, Rendell LE, Amano M (2008) Stereotypical resting behaviour of the sperm whale. Curr Biol 18:21–23

Page 11 of 12 78 Mobley JRJ, Helweg DA (1990) Visual ecology and cognition in cetaceans. In: Thomas JA, Kastelein RA (eds) Sensory abilities of cetaceans, laboratory and field evidence. Springer, New York, pp 1689– 1999 Mukhametov LM (1987) Unihemispheric slow-wave sleep in the Amazonian dolphin, Inia geoffrensis. Neurosci Lett 79:128–132 Mukhametov LM, Supin AY, Polyakova IG (1977) Interhemispheric asymmetry of the electroencephalographic sleep patterns in dolphins. Brain Res 134:581–584 Murase H, Tamura T, Kiwada H, Fujise Y, Watanabe H, Ohizumi H, Yonezaki S, Okamura H, Kawahara S (2007) Prey selection of common minke (Balaenoptera acutorostrata) and Bryde’s (Balaenoptera edeni) whales in the western North Pacific in 2000 and 2001. Fish Oceanogr 16:186–201 Norris KS, Dohl TP (1980) Behavior of the Hawaiian spinner dolphin, Stenella longirostris. Fish Bull 77:821–849 Nowacek DP, Johnson MP, Tyack PL (2004) North Atlantic right whales (Eubalaena glacialis) ignore ships but respond to alerting stimuli. Proc R Soc Lond B 271:227–231 Nummela S (2009) Hearing. In: Perrin WF, Würsig B, Thewissen JGM (eds) Encyclopedia of marine mammals, 2nd edn. Thewissen, Academic Press, London, pp 553–562 Ohman MD, Frost BW, Cohen EB (1983) Reverse diel vertical migration: an escape from invertebrate predators. Science 220:1404–1407 Oleson EM, Calambokidis J, Burgess WC, McDonalds MA, LeDuc CA, Hildebrand JA (2007) Behavioral context of call production by eastern North Pacific blue whales. Mar Ecol-Prog Ser 330:269–284 Panigada S, Pesante G, Zanardelli M, Oehen S (2003) Day and night-time behaviour of fin whales in the Western Ligurian Sea. In: Proceedings of the oceans 2003, San Diego, CA, pp 466–471 Parks SE, Searby A, Célérier A, Johnson MP, Nowacek DP, Tyack PL (2011) Sound production behavior of individual North Atlantic right whales: implications for passive acoustic monitoring. Endanger Species Res 15:63–76 Paul LJ (1968) Some seasonal water temperature patterns in the Hauraki Gulf, New Zealand. New Zeal J Mar Freshw Res 2:535–258 Peichl L, Behrmann G, Kröger RHH (2001) For whales and seals the ocean is not blue: a visual pigment loss in marine mammals. Eur J Neurosci 13:1520–1528 Penry G, Cockcroft V, Hammond P (2011) Seasonal fluctuations in occurrence of inshore Bryde’s whales in Plettenberg Bay, South Africa, with notes on feeding and multispecies associations. Afr J Mar Sci 33:403–414 Pivorunas A (1979) The feeding mechanisms of baleen whales. Am Sci 67:432–440 R Core Team (2018) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ Rattenborg NC, Amlaner CJ, Lima SL (2000) Behavioral, neurophysiological and evolutionary perspectives on unihemispheric sleep. Neurosci Biobehav Rev 24:817–842 Rechtschaffen A (1998) Current perspectives on the function of sleep. Perspect Biol Med 41:359–390 Roth TC II, Lesku JA, Amlaner CJ, Lima S (2006) A phylogenetic analysis of the correlates of sleep in birds. J Sleep Res 15:395–402 Rutz C, Hays GC (2009) New frontiers in biologging science. Biol Lett 5: 289–292 Sekiguchi Y, Arai K, Kohshima S (2006) Sleep behaviour: sleep in continuously active dolphins. Nature 441:E9–E10 Serafetinides EA, Shurley JT, Brooks RE (1972) Electroencephalogram of the pilot whale, Globicephala scammoni, in wakefulness and sleep: lateralization aspects. Int J Psychobiol 2:129–135 Shpak OV, Liamin OI, Manger PR, Siegel JM, Mukhametov LM (2009) Rest and activity states in the Commerson’s dolphin (Cephalorhynchus commersonii). J Evol Biochem Phys 45:97–104

78

Page 12 of 12

Siegel JM (2005) Clues to the functions of mammalian sleep. Nature 437: 1264–1271 Siegel JM (2008) Do all animals sleep? Trends Neurosci 31:208–213 Siegel JM (2011) Sleep in animals: a state of adaptive inactivity. In: Kryger MH, Roth T, Dement WC (eds) Principles and practice of sleep medicine, 5th edn. WB Saunders, Philadelphia, pp 126–138 da Silva J, Terhune JM (1998) Harbour seal grouping as an anti-predator strategy. Anim Behav 36:309–316 Simon M, Johnson M, Tyack P, Madsen PT (2009) Behaviour and kinematics of continuous ram filtration in bowhead whales (Balaena mysticetus). Proc R Soc Lond B 276:3819–3828 Simon M, Johnson M, Madsen P (2012) Keeping momentum with a mouthful of water: behavior and kinematics of humpback whale lunge feeding. J Exp Biol 215:3786–3798 Szczucka J (2009) Acoustic studies of diving birds in the Arctic. In: 3rd International Conference on Underwater Acoustic Measurements: Technologies and Results, Nafplion, Greece, pp 1181–1188 Tershy BR (1992) Body size, diet, habitat use, and social behavior of Balaenoptera whales in the Gulf of California. Am Soc Mammal 73:477–486 Tyson RB, Friedlaender AS, Ware C, Stimpert AK, Nowacek DP (2012) Synchronous mother and calf foraging behaviour in humpback

Behav Ecol Sociobiol (2018) 72:78 whales Megaptera novaeangliae: insights from multi-sensor suction cup tags. Mar Ecol-Prog Ser 457:209–220 Watkins WA, Daher MA, DiMarzio NA (1999) Sperm whale surface activity from tracking by radio and satellite tags. Mar Mammal Sci 15:1158–1180 Wiseman N (2008) Genetic identity and ecology of Bryde’s whales in the Hauraki Gulf, New Zealand. PhD thesis, The University of Auckland, New Zealand Würsig B, Würsig M (1980) Behavior and ecology of the dusky dolphin, Lagenorhynchus obscurus, in the South Atlantic. Fish Bull 77:871– 890 Ydesen KS, Wisniewska DM, Hansen JD, Beedholm K, Johnson M, Madsen PT (2014) What a jerk: prey engulfment revealed by high-rate, super-cranial accelerometry on a harbour seal (Phoca vitulina). J Exp Biol 217:2239–2243 Zerbini AN, Secchi ER, Siciliano S, Simoes-Lopes PC (1997) A review of the occurrence and distribution of whales of the genus Balaenoptera along the Brazilian Coast. SC/48/SH4. Scientific Committee of the International Whaling Commission, Monte Carlo, Monaco