Use of an acceleration data logger to measure diel activity patterns in captive whitetip reef sharks, Triaenodon obesus. (web version)
Nick Whitney, University of Hawaii at Manoa
Traditional telemetry methods can tell us a shark's approximate location, but cannot tell us about the physical movements of the shark's body. This study used a prototype acceleration data logger to quantify the body movements of three captive whitetip reef sharks for periods of 6 to 16 days at a time. All three sharks showed clear patterns of heightened activity at night, followed by long periods of resting during the day. These results were in agreement with results of respirometry experiments from two newborn whitetips which showed higher metabolic rates at night than during the day. Together, these results suggest that whitetips are nocturnally active and that this activity pattern is "hardwired" from birth. Future experiments should seek to directly correlate body acceleration with metabolic rate using accelerometry and respirometry simultaneously. This would require smaller accelerometers and/or larger respirometers than were available for this study.
Unlike many shark species, whitetip reef sharks do not have to swim constantly in order to breathe. They're able to actively pump water into their mouths and out over their gills while lying in place on the sea floor. But how much time do they actually spend resting compared to swimming? Are they more active during some parts of the day than others? Answering these questions can provide key insights into whitetip movement patterns and energetics.
Traditional telemetry methods can tell us a shark's approximate location, but cannot tell us about the physical movements of the shark's body. For instance, a shark that was resting on the bottom and a shark that was actively feeding in a small area (trying to flush a reef fish out of a hole for instance) would both look the same (showing zero displacement) to a scientist tracking those sharks from a boat using traditional telemetry. This study used a prototype acceleration data logger to quantify the body movements of three captive whitetip reef sharks for periods of 6 to 16 days at a time.
What is an acceleration data logger (ADL)? image © Vemco
Acceleration data loggers represent a new tool for measuring animal behavior and energetics, because they can tell us when and how the animal's body is actually moving. Contained within the casing of a traditional acoustic tracking tag, Vemco (Amirix Systems Inc.) acceleration data loggers have three sensors, each of which measures acceleration in a different dimension. When the logger is attached to an animal, the speed and direction of that animal's body movements are constantly recorded for the duration of the experiment (often a week or longer, depending on sampling rate/memory and battery life). When the logger is recovered, it is downloaded to a computer and the data can be analyzed to answer various questions about the animal's movements.
These loggers have been used to study stroke velocity and energetics in a variety of species (birds, fish, mammals) but have not been previously applied to a shark.
How were ADLs used in this study?
This study used a prototype acceleration logger from Vemco to measure the diel (day and night) swimming activity of captive whitetip reef sharks for continuous periods of 6 to 16 days at a time. Loggers were either inserted gastrically (force-fed) or attached externally to the shark using a single Hallprint-type tag barb attached to one end of the logger. Animals were visually monitored and/or video recorded for brief periods (10-30 min) at least every other day to look for signs that the tag had been regurgitated/shed, and to record the physical movement of the shark for later corroboration with acceleration data from the logger. Once tags were shed or their battery life/memory had expired, they were recovered and their data downloaded and analyzed.
What were the results?
Both attachment techniques produced data showing clear differences between resting and active swimming behavior (see figure at right). Data from all three captive sharks showed long periods of rest during daylight hours, interrupted by short intervals of active swimming. In contrast, nighttime behavior was marked by long periods of active swimming with only shorter and less frequent stops to rest on the bottom. The difference between day and night activity levels was statistically significant for all sharks.
Overall, the whitetip reef sharks in this study spent an average of only 35.2 + 11% of their time swimming. They spent 10-24% of their time swimming during the day, and 42-67% swimming at night. These results were in agreement with respirometry experiments conducted on two newborn (approx. 3 week-old) whitetip reef sharks that showed significantly higher metabolic rates at night compared to daytime.
Gastric insertion provided a natural means of logger recovery (regurgitation) but led to shorter retention times than external tag attachment.
Why does it matter?
These "proof of concept" data represent the first use of an accelerometer to measure activity in an elasmobranch and show increased nocturnal activity (relative to daytime) in captive whitetip reef sharks. This pattern was seen despite the fact that animals were fed to satiation every other day (during daylight hours) and therefore had no need to hunt at night. Similarly, the fact that newborn whitetips (born in captivity with no hunting experience) showed an increase in nighttime metabolic rate indicates that nocturnal activity is "hard-wired" in this species and that free-living whitetip reef sharks are likely to show the same general pattern. Knowing the diel (day/night) activity cycles of an organism is important for understanding larger patterns of energetics, movement and dispersal.
Future experiments should seek to directly correlate body acceleration with metabolic rate using accelerometry and respirometry simultaneously. This would require smaller accelerometers (that could be applied to smaller animals) and/or larger respirometers (that could house larger animals) than were available for this study.
Acceleration data loggers also represent a potentially invaluable tool for quantifying specific behaviors that are difficult to observe directly in elasmobranchs (e.g. feeding, agonistic interactions, courtship and mating). Acceleration signatures from these behaviors would need to be corroborated with direct observations initially (perhaps using captive animals), but may then be applicable across species. Mating in particular involves a suite of movements that are similar among different species and likely to produce highly specific acceleration signatures relative to other behaviors.
Where can I read more about this study?
Download the full PDF from the scientific publications section of our "About Us" page.
Who helped with this study?
We thank Tom TinHan, J. Henly, E. Rechisky, and others for help with experiments, and D. Webber (VEMCO) for technical support. Thanks also to J. Dale and A. Taylor for manuscript review and analytical advice. This project was partially funded by a National Science Foundation Pre-Doctoral Fellowship and a NFWF Budweiser Conservation Scholarship to Nick Whitney.