Counting krill by sound

Hi! This is Wu-Jung reporting the acoustic side of our cruise. I am a grad student in the MIT/WHOI Joint Program working on the acoustic scattering from marine organisms. Coming out to Georges Bank in this season is quite exciting for me who came from tropical Taiwan!

So, we've been talking about sampling the water column using many different types of technology such as an optical device (VPR) and a net system (MOCNESS). However, the volume of water that we could look at using these methods is tiny compared to the size of the ocean, and we can only do this sampling at a handful of stations. How can we see the big picture of the distribution of krill as we drive over Georges Bank?

We use acoustics! Sound travels very well underwater compared to light. By transmitting sound into the ocean and analyzing the frequency, amplitude, and timing of the echoes, we can identify animal aggregations and estimate the number and composition of the aggregation. When we ‘ping’ (send short bursts of sound) along the way while doing the VPR and MOCNESS sampling, we can begin to ‘connect the dots (stations)’ between these locations where we have ground-truth photos and net samples of the animals. We can then start to put pieces together to restore the big picture of organism distribution in the entire area.

This time we have two acoustic devices onboard: the Greene Bomber and the Hammerhead. The Greene Bomber is a multi-frequency narrowband system that has 4 transducers (functioning as a combination of speakers and microphones) at 43, 120, 200, and 420 kHz. The Hammerhead is equipped with 5 broadband transducers encompassing a wide frequency band between 35 to 600 kHz.

People busy on the deck preparing to deploy the instruments. In the front, Patrick, Kaylyn, Gareth, and Oscar (left to right) working with the Greene Bomber. Under the A-frame, Dave (left) and Cindy (right) working on the MOCNESS. (Photo by Wu-Jung Lee)

Gareth (left) and Patrick (right) deploying the Hammerhead. (Photo by Wu-Jung Lee)

The narrowband acoustic measurement is a very well established method for studying the distribution of marine organisms. Different marine organisms scatter sound differently in different frequency bands: fish can be observed starting at lower frequency such as 43 kHz, but the zooplanktons most likely only show up at higher frequency bands like 120 kHz and above.

Echograms from the Greene Bomber. The zooplankton is more obvious on the higher frequency channel (120 kHz), while the fish show up on both channels. (Image courtesy of Dr. Gareth Lawson)

However, looking at data from a narrowband echosounder is like watching black-and-white TV. You see things, but it is hard to tell a red apple from a green one. Without sophisticated signal processing techniques that can be applied to broadband echoes, the resolution of the resulting picture (the ‘echogram’) is not very good either. Moving towards broadband echosounders is like upgrading your TV from black-and-white to modern color hi-def TV – not only do you have vivid color of the objects, detail of the objects are enhanced, too.

The development of broadband echosounder technology is just getting into a stage that it can be applied in the field. Therefore, we tried to deploy both systems simultaneously as much as we could, in order to compare the broadband system which is still in an experimental stage with the more established narrowband system. Now we’re coming back with the first ecological application of the broadband system!

Here’s more info about the recent development of broadband acoustic systems at WHOI:Now in Broadband: Acoustic Imaging of the Ocean