Page 14 - Acoustics Today

Fig. 8. Track of Slocum glider in the Gulf of Mexico and locations of detected fish sounds and suspected fish sounds. The toadfish found more offshore is likely the leopard toadfish, which had previously been unknown. Illustration by Carrie Wall.
Fig. 9. A deep-sea giant cusk eel on Davidson Seamount at 2,677 m depth. So far, no deep-sea fish sounds have been positively identified. Photo: National Oceanic and Atmospheric Administration (NOAA).
distributions. The glider recordings
also demonstrate how little we know
about the sound producers, literally
in our backyards. New methods are
needed to identify which species
make which sounds. Video cameras
only work where there is light, and
often rely on intimate knowledge of
the location and reproductive activ-
ity of a species to be able to capture
these special moments. Marine
mammal acoustic recording tags,
like the Dtag and Acousonde, have
yielded tremendous insights into
the use of sound by dolphins and
whales. But, they are too large for
most fishes. One solution may be
miniature, accelerometer recording
tags to store the vibrations associat-
ed with sound production in fishes.
Because most fish acoustic energy is
below 1,000 Hz and relatively high
level, an accelerometer should be
capable of measuring the near-field
particle acceleration component of
sounds produced by a tagged fish.
We need a renewed “Fish-and-Mowbray-like” effort to cat- alog sound production by marine animals in their natural habitats to realize the full potential of passive acoustic remote sensing of fish populations.
While it is has become easy to collect terabytes of acoustic recordings, analyzing those data is a critical chal- lenge. It is impossible for a small laboratory to listen to all of the recordings they can make. Automated signal pro- cessing algorithms are a high priority to stem the data del- uge. We have had success with automated routines for fixed locations where we had characterized all of the sounds
from that site. These algorithms, however, have fallen apart when applied to glider recordings, or even to new locations where other sound producing species cause false alarms. We are at the dawn of a new age of both discovery and fish- eries science with the integration of passive acoustics in ocean observatories. To fully develop this field requires the collaboration of engineers, experts in signal processing, and fisheries scientists. There is nothing like the thrill of being the first person to hear and identify a new sound, and there are plenty of fish in the sea waiting to be heard (See Fig. 9). AT
Remote Sensing of Fish 13