The Instruments

As I mentioned in the last post, there are three legs to our survey: two along the forearc rifts and one along the spreading center north of them. The primary mapping instrument we are using is deep-towed sonar. There are two instruments: the IMI-30 and the IMI-120. They are both sonar systems that are towed on a cable relatively close to the seafloor. The IMI-30 is slightly lower resolution (~5 m) and towed ~500 m above the seafloor, but gives you slightly wider swaths of data. The IMI-120 is towed ~100 m above the seafloor, gives you narrower swaths, but higher resolution (~1-2 m). For comparison, ship-mounted systems have resolutions of 30-100 m, depending on water depth and the frequency of the instrument. As I explained in a previous post ("Sonar Mapping" in the "The Mission and Instruments" section of the blog on the previous cruise, 11/12/11), sonar systems emit sound waves that bounce off the seafloor and return to detectors on the instrument. There are two types of measurements: bathymetry and sidescan (aka backscatter). For bathymetry, the instrument measures the travel time of the sound beams and converts it to a distance to give a map of the topography of the seafloor. Sidescan (backscatter) is a measure of the intensity of the reflection, which is affected by both the type of material and the orientation of the seafloor relative to the instrument. Harder surfaces (such as fresh lava flows, or rock that is exposed along a fault scarp) give a stronger reflection while softer surfaces (such as sediment) give a weaker reflection. With an instrument mounted on a ship that is a few 1000 m above the seafloor, the type of material is the primary thing that is picked up in sidescan imagery, the orientation of the seafloor does not have much of an effect because it is so far away. You essentially end up with a black and white image of the seafloor that shows you where highly reflective recent volcanism and faults are, and less reflective sediments are. However, with a deep-towed instrument, since it is much closer to the seafloor, the topography of the seafloor becomes a much bigger factor. Imagine if the instrument was flying through a valley that is 200 m deep but it is only 100 m off the bottom. It will not be able to "see" over the edges of the valley and thus you will get no reflection from outside of the valley. Also, if the instrument is flying along a steep slope, it will get a very good return from the upslope side but will get very little reflection from the downslope portion. If there is a volcano on one side of the instrument, the side of the volcano facing the instrument will give a strong reflection, but there will be a "shadow" on the opposite side because the sound waves are not hitting that side. Deep-towed data also tends to degrade toward the edges of the swath and is usually not very good directly below the instrument because the beams are oriented toward each side not vertically (hence the "sidescan" name). However, you can see MUCH more detail than you can with instruments that are mounted on the ship, allowing you to identify individual small faults and lava flows and even see the texture of the surface of a lava flow so you can tell what kind of flow it is. In an image from a ship-mounted system, you would just see dark and light patches, and only large faults will show up. We are primarily using the IMI-30 because it allows us to cover a wider area while still giving great resolution and has less distortion than the IMI-120. We will use the IMI-120 if we want to examine a particular area in greater detail or if the IMI-30 craps out on us. We will be towing these instruments along the length of the rifts and the spreading center, providing much more detailed maps of the volcanic and tectonic structures than we currently have. It will eventually be my job to interpret these detailed maps and try to figure out what is going on in this region (I honestly don't know exactly what I'll be saying in the papers I'll be writing because I haven't seen any of the data yet). We are basically driving the ship in straight lines along the features, collecting 2 or 3 parallel overlapping swaths of data for each of the features.

A small (I couldn't get a larger version without signing in) schematic image showing the deep-towed sidescan sonar setup. You can see how the beams are oriented, the blank spot directly below the instrument, and the strong reflections/shadows on either side of the volcanoes. I'll post an actual image of what the data looks like later.

My job on this cruise is to do some of the initial processing (called "bottom detect editing") of this data as it is collected. I work a 9 hour shift from 6 pm-3 am. Fernando's (my advisor) other student, Regan, is doing another 9 hour shift from 3 am to noon, and one of the employees of the group that owns and operates the instrument (Hawaii Mapping and Research Group (HMRG)) is covering the noon to 6 pm shift. It is somewhat difficult to explain what exactly I'll be doing and I'm not fully comfortable with all of the technical details of how these instruments work, but basically I'm making sure that the instrument knows where the real bottom is. Because we're dealing with sound waves and they aren't as precise as something like a laser, there are a lot of echoes and reflections that confuse the instrument. The instrument has an array of transducers (the things that emit and detect the sound waves) on either side, roughly oriented 45 degrees from vertical. Sometimes the sound reflects off something in the water column that is not actually the seafloor, and sometimes the sound waves from one side of the instrument are picked up on the other side of the instrument. This can make the instrument confused as to where the real bottom is on a particular side. So I look at the raw data right after it's collected and try to find the reflection that represents the "true" bottom. However, because we don't actually know definitively where the bottom is, it can be very ambiguous sometimes, especially when the topography is complex and/or different on either side of the instrument. The guy who is in charge of HMRG calls it a "dark art" rather than a science, and this is an apt description. The basic idea is not to be 100% precise on getting the actual bottom, because you can't, but rather to make the image more believable and easier to interpret and to eliminate the obvious errors. Sometimes it is very clear where the true bottom is, sometimes it is not clear at all. So far, it has been going pretty well, there have definitely been some ambiguous swaths, but I'm getting more and more comfortable and efficient with the process. Unfortunately, the first survey area is along the spreading center, which is the simplest of the three survey areas, so it will likely be more difficult for the other two areas.

The other major part of the mission is sampling the volcanic features that we see in the sonar data. I will not be involved in this part of the operation, beyond checking out some of the rocks for fun. There is another group of grad students from University of Rhode Island and University of Texas Dallas who are covering this part of the mission. There are two main sampling methods: wax coring and dredging. I have described dredging in detail (see "The Mission and Instruments" post from the last cruise, 11/12/11) previously, so I won't talk much about that. Wax coring seems to be kind of a lame sampling method and doesn't really produce great samples, I think dredging will end up being the primary method. No dredging has occurred yet, we are waiting until we map the first area to identify good sites to dredge, but we did do a few wax cores when we had to pull up the deep-towed sonar for repairs two days ago. It's basically a cluster of small metal cylinders filled with wax that have a heavy metal weight on top and are attached to a wire extending from the ship. It's a pretty simple process, they basically just drop the thing overboard and let gravity slam it into the seafloor. Ideally little bits of rock will get stuck in the wax, which can then be removed and analyzed. To any non-geochemist, the samples are not impressive at all, they are literally tiny mm-sized bits of rock (or sediment if you're unlucky) that aren't much to look at unless you have a powerful microscope. They still can do chemical analyses of the tiny samples, so it is useful, but definitely not the best sampling method. Dredging may be imprecise, but at least you get lots of rock to look at. I'll post some pics once we actually get some dredged rocks. I think that covers the gist of what we are doing.