OLI Grant: Sensors for Velocity and Heart Rate Acquisition on Wild Marine Mammals


Project Funded: 2001

Proposed Research

Marine mammals use the underwater environment for foraging, communication, and in some cases, mating, and have developed a number of adaptations to enable this. However, the sub-surface behavior of marine mammals, especially deep divers such as sperm and beaked whales remains a mystery because direct observations are not possible. Understanding these behaviors is not only of biological interest, but is necessary to understand the effect on marine mammals of human activities such as fishing, oil exploration, and naval operations.

Johnson and Tyack previously have developed a multi-sensor tag (DTAG) which records simultaneously the sounds heard and produced by the host animal, the orientation of the animal, and its depth. This non-invasive tag has been deployed on northern right whales and sperm whales providing critical new insights into their sub-surface behavior and response to sound. Having a method for detecting a response, the next goal is to be able to understanding the biological significance of such changes in behavior. A prerequisite to answering this question is to characterize the host animal's energy use, so that changes in behavior can be understood by how they affect the energy budget of the animal.

Progress Report

The objective of this project is to develop two new sensors for a multi-sensor marine mammal tag, called the DTAG. The DTAG, developed by the authors at WHOI, currently records sound and orientation of the tagged whale. With the new heartbeat and swim-speed sensors we are designing, it will be possible to estimate the energetic balance of natural activities such as swimming and foraging for whales in the wild as well as the cost of their responses to human activities. This information is vital for the effective management of human impact on coastal dwelling cetaceans.

There are a number of methods for measuring heartbeat and swim-speed, each with inherent advantages and disadvantages. Our approach has been to develop prototype versions of the most promising sensors and evaluate these using test facilities at WHOI and, where appropriate, on captive marine mammals at nearby aquaria. As the sensors mature, we are incorporating them into the DTAG for deployment on wild marine mammals during scheduled field efforts, permitting evaluation under fully realistic conditions.

Although an intense field program for the authors in 2003 has slowed down progress on the new sensor project, we have made several important steps. We have been evaluating three methods for heartbeat acquisition: electrical (EKG), passive acoustic (AKG), and Doppler ultrasound. We have now tested all three methods on captive marine mammals including manatees, belugas and dolphins with varying results. Fig. 1 shows a heartbeat signal recorded using our EKG sensor on a captive dolphin exemplifying the high quality measurements possible with carefully-placed EKG electrodes. Results on a wild animal would be significantly poorer due to the limited electrode spacing consistent with a suction-cup attached tag. More recently have obtained a very promising result with an AKG sensor on a wild Blainsville's beaked whale (Mesoplodon densirostris), the first time this little-known species has been tagged successfully. Blainsville's beaked whale is one of two species of beaked whale that have stranded on a number of occasions coincident with the use of navy sonars making this tagging success very significant. The photograph in Fig. 2 shows Blainsville's beaked whale at our study site in the Canary Islands and shown in Fig. 3 is an example of the heartbeat data obtained from the DTAG. AKG is adversely affected by flow noise around the sensor and we are currently only able to detect heartbeat acoustically when the tagged whale is resting. It appears now that a combination of sensors, for example EKG and AKG, will ultimately provide the best performance on free-swimming animals and work is continuing to integrate both sensors into the tag.

For swim-speed sensing, we have focused on an active acoustic sensor using the Doppler principle. Doppler velocity sensors are widely used in oceanographic instruments for measuring water currents and have the important benefits of (i) being able to measure the direction as well as the magnitude of flow, and (ii) being able to measure the flow at a distance from the sensor rather than the disturbed flow right next to the sensor. These benefits make a Doppler sensor an excellent choice for use on marine mammals and the primary difficulty to be overcome is that of miniaturizing the electronics for integration into the DTAG. This year, a group of senior electronic engineering students at the University of Massachusetts at Dartmouth, undertook to design and test a miniature single-axis version of the Doppler sensor as a final-year project. The group succeeded in producing a small low-power prototype and tested it using the WHOI flume facility. Preliminary tests results, such as those shown in Fig. 4, are encouraging. In this figure, the Doppler shift due to movement of the sensor against still water in the flume is clear and can be translated into an accurate velocity estimate. Work is now continuing to extend the prototype sensor to measure in 3-axes and to integrate the device into the DTAG.

Aknowledgements: Tom Hurst, Alex Shorter, Douglas Nowacek, Jennifer Miksis, David Brown, Mystic Aquarium, Disney Corporation, Mote Marine Laboratory, University of Massachusetts at Dartmouth, University of La Laguna, Field team in El Hierro (Canary Islands).