When studying an animal as poorly understood as the coelacanth, it is important to build up as complete a picture of its life as possible. We generally only see coelacanths in the day, which is their resting time (they are nocturnal animals). Therefore, we learn little of their actual lives whilst observing them hiding in their caves during the day. It is only at night that coelacanths become truly active. Coelacanths are nocturnal hunters that seem to drift over their habitat or swim slowly at night, waiting for prey to come within range of a short lunging attack. It is difficult and somewhat intrusive to follow these hunting forays from a submersible. Instead, we use a system comprising of a small ultrasonic tag and a reciever unit with a directional hydrophone.

A directional hydrophone fitted to Jago

.

In practice, the system works much like tracking radiotagged land animals, with further complications added by the fluid environment and working from boats - wind, waves, current, working at night and very complex bottom topography all add to the challenge.

The pinger unit, with an integrated depth sensor is attached to the animal from Jago.

Kerry Sink follows a tagged coelacanth
from aboard Algoa. The large metal pole is
connected to the hydrophone and allows the operator
to swivel it around to hunt for a strong signal.
A much smaller version of this device
is used aboard the inflatable boat
used to track the fish

This unit sends out intermittent pulses of ultrasound. Encoded in these pulses is the current depth of the tag. These pulses of sound travel through the water, where we can "listen" for them with a hydrophone (underwater microphone).

The hydrophone is directional; this means it can only pick up sound from a relatively narrow arc, much like a TV aerial has to be precisely aligned to the transmitter tower to pick up a good signal. In contrast, a normal hydrophone picks up sound from every direction.

Differences between normal (omnidirectional) (left)
and directional (right) hydrophones

The tag sends out "pings" of ultrasound. These cannot pass through rock, so signals emanating from caves can be very weak and detectable only in a small area. You can see in the illustration that rock is blocking the signal in certain directions; normally the signal in unobstructed water would travel in all directions. The red lines represent the ultrasound waves passing through the water. The white lines indicate the extent of the signal, which is cut down by the rocks in the canyon. Coelacanths deep in a cave may end up giving no detectable signal at all from the surface, but Jago is able to get into the Canyons and look around to relocate an animal if we lose it from the surface. We can track three ways; from a small inflatable boat (the most manoueverable), from Jago or from Algoa

Some of the skippers who are assisting us with tracking the tagged coelacanth at night
from their boats. The man with no shirt is Peter Timm, one of the divers who first
discovered coelacanths at Sodwana Bay

The VR-60 surface based tracking device. A hydrophone plugs into this unit, and it listens for and amplifies signals. If a good strong signal is recieved, then the unit will be able to decode the depth information encoded in the ping. You can listen for weak pings using the speaker; they have a very distinctive "ping" noise. You turn the hydrophone around in the water, listening for these weak pings. Once you detect one, you carefully line up the hydrophone in the direction of strongest signal; this can be determined by the strength (loudness) of the ping, but more accurately using the signal strength indicator. Once several good pings have been recieved, the unit will display the depth on the LCD screen. The gain knob changes the amount of amplification; if you have a very weak signal, you need to use a lot of gain, but this also amplifies background noises; you try to keep the gain as low as possible. Once you have the strongest signal coming from directly underneath you, a GPS and depth reading is taken, and noted down. This can then be used together with other GPS positions and depths collected in the same way to track and plot the movements of coelacanths at night

We have another slightly different tagging system which allows us to monitor coelacanths for some time remotely by deploying VR2 receivers on the bottom. These nondescript grey devices are fixed to the bottom or floated off the bottom on rope, and they listen for the pings given off by the tags on the fish. When a tagged fish comes into range, the unit sends a message to the tag to ask it to send the depth data it has collected since it was last downloaded. This will give us some indication of the vertical (depth) movements coelacanths undergo over longer periods of time; if several VR2s are deployed in an "array", you can also track their horizontal movements, although not with the resolution you get from surface or Jago tracking.

A VR-2 Listening Station; the orange part
at the top is the omnidirectional hydrophone

Mike Roberts intends to deploy several of these "listening stations" around Jesser and Wright Canyons so we can track their movements. He also intends to attach a thermistor chain to some of them to monitor the temperature structure of the water column. A thermistor is an electronic circuit that changes its resistance according to temperature; this makes an effective thermometer, and with some recording equipment attached, you can follow the temperature structure of a section of the ocean over an extended period of time - water temperature is very important in oceanography, and can tell us a lot about the oceanographic processes occurring around Sodwana Bay.

A VR-2 tag. These tags record depths and relay them to a listening station every time the animal brings the tag within range

When we are next in the area, we collect the VR2s and thermistor chains by activating a device called an "acoustic release", which listens for a particular (loud!) frequency before activating a small motor that releases a bolt attached to the mooring, allowing the scientific equipment (which has buoys at the top) to float to the surface. The data can then be downloaded and analysed on a PC.

A VR2 listening station with thermistor chain. The instrumentation is moored to the bottom (black square) in up to 200m of water. The yellow rectangle is an acoustic release so we can retrieve the device from the ship. The white lines represent thermistors, and you can see the buoys at the top (orange) which keep the instrumentation hanging upright in the water from the mooring; it also raises the package to the surface when the acoustic release is activated

An "array" of listening stations (small grey circles). The white lines indicate the detection range of the VR2 listening station. The orange block is a tag, and the red circles indicate the acoustic "pings". The pinger is in range of the bottom left hand listening station, and is transferring data (arrow)

The pingers, hydrophones and receivers are manufactured by Vemco in Canada.