Tuesday, 9 September 2008

Acoustic surveys

We're currently on an acoustic survey line, where the ship navigates a course along the coastline of Madagascar.

To the left, you can see "mission control" - part of the acoustics lab. Starting at the computer on the desk on the left, from where the CTD is monitored and controlled, you then have a laptop to the right, which is used to monitor and record data from the thermosalinograph and fluorometer which sample water continuously, 24 hours a day, from about 5m under the surface (a pump sucks water in through a seacock, passes it through the instrumentation, located in the same room the CTD is housed in, and then the water is dumped back overboard). This tells you how warm the water is, what the salinity is, and gives you an idea what the levels of chlorophyll in the water are like - telling you roughly how productive that water is. None of that is acoustics gear, of course (it does nothing with sound!).

Then there are 3 racks of gear. The lefthand most rack contains (from the top) some UPS monitoring instrumentation, then a display that tells you how quickly the CTD is going up or down and how much wire has been let out; below that is a display of the ship's heading; below that there is the CTD deck unit that translates between simple electrical signals the CTD sends and can understand to computer-ese; below that is the ADCP (Acoustic Doppler Current Profiler) which measures currents in the upper few hundred meters 24 hours a day.

The middle rack contains an echosounder display, the ship's log (think of it as the mile-o-meter in your car), below that the multibeam echosounder display (dark screen [off!]) and rack mount PC.

The third rack has more echosounder displays at the top, then there is the Seapath 200 deck unit; this is basically a very accurate inertial navigation system which tracks the vessel's motion in 3 dimensions (particularly pitch and roll); data from there is used to calibrate sensors, particuarly the multibeam echosounder, that are sensitive to the vessel's precise position in the water. Below this is a remote screen and mouse which displays and can control the Olex navigation machine on the bridge. Behind the camera position is a powerful workstation with two enourmous screens used to post-process the echogram data - more on that a bit later.

Never mind the funky-looking toys, what exactly is this acoustic survey business anyway?!

As you may know, many fish species show up on echosounders, and indeed, particular species reflect different sonar frequencies in an often characteristic way (important characters that contribute to this include the size and shape of the swimbladder as well as behaviour - the shape the fish take within their schools and where in the water column they hang out). This allows an experienced observer, with the right tools, to estimate the abundance of particular species of fish without even catching any of them - a pretty ideal situation!

However, it is of course important to calibrate the system; for this reason, trawls of various types are deployed to target characteristic signals in order to calibrate the system (i.e. understand what the various echoes represent) and gather additional data you can't get from echoes alone. The signatures of many major northern hemisphere fishery species are very well known, to the point one can often automate the process; down here in the southern hemisphere, things are a bit more complex and less well studied. So, it's trawling time!

How do we get this magical acoustic data in the first place? Well, the Dr. Fridtjof Nansen is equipped with a Simrad ER60 echosounder system, using 18, 38, 120 and 200kHz frequency transducers. As you're probably well aware, sonar sends out a series of "pings" and measures how long it takes for these to reflect back; if you know the speed of sound through the water, you can work out how far away the object that reflected the sound is. Incidentally, the speed of sound in water changes depending on density - and this may change through the water depth; we use accurate density information from the CTD to create a detailed sound profile of the water column for all the acoustic instrumentation on board. Of course, with the growth of technology, other information other than depth/distance can also be gleaned from not only the time(s) of the echo(es), but subsequent reflected echoes and even the quality ("loudness") of the echo. Different frequencies reflect off different types of objects in different ways, so, once you know what bottom type X looks like on the sonar, you can know what bottom type you're passing over; objects in the water column, like fish, also produce particular echoes.

Recorded echogram data from this system is then post-processed using the Large Scale Survey System (LSSS). This software allows us to reduce the "clutter" in the water column caused by plankton, and focus on the fish. It also allows us to identify fish shoals and allocate them to specific groups, such as "anchovies" or "horse mackerel" or "scads", for example. Even the live view of the echosounder gives us an idea of how many fish there are out there, and a view of the bottom, so we can decide if the bottom looks trawlable or not. Take a look at the two screenshots below:

If you open them up full-size by clicking on them, you'll see a lot more detail. The main part of the window is the sonar echogram. The wavy dark red line is basically the bottom; the little peaks and valleys indicate it's pretty rough (and not good bottom trawling ground). Most of that random-looking blue speckling above the red line is plankton in the water column. On the left side image above, below the wavy bottom line, there is a straight dark red line, which is also the bottom (with the kinks smoothed out digitally), and represents an "expanded view" of the part of the echogram just above the bottom - where fish often lurk! The diagonal line across the screen is a plot of the total power of all the echoes received; when there is a sudden jump, that generally indicates a school of something. Jens-Otto thinks that the pelagic fish (those that live in the water, not near the bottom) are staying much closer to the bottom than normal, and he thinks whales are responsible (the fish are hiding from them) - it's apparently very noticeable on the echogram when whales are around. The little graphs indicate (amongst other things) how the particular target you're examining reflects in the different frequencies; on the right at the bottom one can assign the section of the echogram you've selected to a particular category you define (anchovies, scad, horse mackerel etc.). To the left, I've cropped out what a small school of fish near the bottom looks like; the raised blue-green hump rising up from the reddish bottom represents the fish! The red dot on the map in the bottom left hand corner represents the location of the section of the acoustic readings you're currently looking at.

This software, when correctly calibrated (i.e. you know what the different species look like) allows you to estimate the total biomass of various fish species that you pass over when you survey - and it can do so over any terrain, unlike trawls, and does so without killing anything or trampling all over the seabed like a bottom trawl does. Acoustic surveys are being increasingly used in fisheries management as they allow much more rapid and wide-spread surveying to be done (and you don't smell of fish at the end of the day) - but trawling can give you other useful information, so it is unlikely to ever be totally replaced.

There are, of course better tools than basic echosounders for studying what the bottom looks like; the Nansen is equipped with a Kongsberg EM710 multibeam echosounder system. The user interface is a program called Seafloor Information System (SIS), which gives a real-time view of the seabed in three dimensions - very handy for getting an idea of what the bottom really looks like - and allows control of the system. Data from this system is additionally fed into Olex, the navigation system on board, which can also be viewed in 3D. (Data from the ER60 system is also fed into Olex, and gives an indication of bottom hardness, as well as updating the depth information on the chart in real-time). If you're prepared to spend the time, you can build up really nice 3D pictures of large areas of the seafloor. I'll try and get an image of an area of seabed they surveyed last year off Mozambique using this method at some stage and go into more detail on the multibeam - they're very awesome tools.

Anyway, hopefully this has given you a little bit about what is going on in acoustic surveying; if you want more info, make a comment using the link below and let us know what you'd like to hear more about - or if anything was unclear!

At the time I started this post, we were about to do a trawl around here.

Comments: Post a comment

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]