Tolotti et al.: Vertical movements of oceanic whitetip sharks {Carcharhinus longimanus) 
385 
Frequency 
Figure 2 
Periodogram generated with fast Fourier transforms of the continuous depth data 
from an oceanic whitetip shark ( Carcharhinus longimanus), AOCS3, tagged in the 
Atlantic Ocean in 2011. The cycles are expressed by l/(frequencyOe)). Spectral den- 
sity represents the relative magnitude of frequency. 
tagged on the same day at similar locations, and the 
duration of their monitoring period was the same. This 
unique situation facilitated the incorporation of a spa- 
tial aspect in the discussion concerning shifts between 
behavior types. The proportions of the 3 behavior types 
were linked across the tracks of these individuals (Fig. 
5). Considering that individuals that are simultane- 
ously at the same location typically also experience 
the same environment, this analysis revealed moments 
where environmental factors could have been the driv- 
er for the behavior of these sharks. At the beginning 
of May 2012, for example, sharks AOCS5 and AOCS7 
were in the same square of latitude and longitude and 
had a similar behavioral pattern. Shark AOCS6, on the 
other hand, was in another location and had a differ- 
ent behavioral pattern. In fact, this shark remained in 
the same area for most of May, but its behavior pat- 
tern shifted completely between the first and second 
half of the month. Again, an environmental factor may 
have been the driver. However, during the second half 
of April, sharks AOCS5 and AOCS7 were at the same 
location exhibiting completely different behavioral pat- 
terns. This trend appeared again during the second 
half of May, indicating that the physical environment 
might not be the only driver of behavior types. 
Spike dives While examining the time series, we noted 
that all individuals stayed primarily within the top 150 
m of water but descended on rare occasions to depths 
below 150 m. By looking closely at these rare deep diving 
events, which accounted for only 0.15% of the monitor- 
ing periods (Tolotti et al., 2015b), we identified common 
features. These features, referred to as spike dives, were 
characterized by rapid descents to depths greater than 
150 m, followed by considerably slower ascents. Two 
examples of spike dives can be seen in Supplementary 
Figure 3 (online only). All individuals performed spike 
dives during their monitoring period, and estimated 
descent rates varied from 0.14 to 1.05 m/s and ascent 
rates varied from 0.08 to 0.26 m/s (Table 2). Most of the 
spike dives lasted from 30 to 45 min, but spike dives of 
more than 1 h also were noted. With the exception of 
movements of shark AOCS7, spike dives occurred pri- 
marily during the day (Suppl. Fig. 4 (online only)). Besides 
performing the great majority of its spike dives during 
nighttime, shark AOCS7 also exhibited this behavior 
much more frequently than the other individuals. 
Vertical movements and the environment 
Daily average depths and SDs were plotted on top of 
temperature profiles to identify possible links between 
vertical movements and the thermal structure of the 
water column. Although some thermal changes were 
observed, the daily average depths were stable for all 
sharks (Fig. 6). The daily SD of depth records, on the 
other hand, indicated some variation. For sharks IOCS1 
and AOCS6, for example, higher SD values were ob- 
