Yesterday we collected multi beam bathymetry data on the upper forearc slope south of Umnak Island using the ship’s EM302 multi beam echo sounder. The ship has a special hull-mounted acoustic transducer that sends out several beams of sound in a swath that is perpendicular to the length of the hull. This acoustic energy reflects off the seabed and is recorded by the same transducer. The system records the return arrival of acoustic energy as a function of angle along the transducer head, as well as its intensity and the two-way travel time of the acoustic energy (the time it takes to go from the ship, to the seafloor, and back to the ship). The two-way travel time and the angle allow the system to determine the distance to the part of the seafloor that reflected that energy. As the ship drives along a transect, a 3D model of the shape of the seafloor is determined. The swath is about three to five times wider than the ocean depth, allowing for a fairly wide portion of the seafloor to be mapped on a given transect. We can also use the intensity recordings to help determine the nature of the seabed (for example, sediments versus hard rock).
One key aspect of getting an accurate map is to use the sound velocity profile of the seawater when converting the acoustic travel times into distances. The sound velocity in the seawater is a function of both temperature and pressure. In the mixed layer of seawater in the upper few hundred meters of the ocean surface, temperature is the dominant factor. The temperature can vary quite a bit in the mixed layer and the thermocline below, and so it must be measured to get an accurate velocity model. This is done cheaply using what is known as an expendable bathythermograph, otherwise known as an XBT. This XBT consists of a small torpedo like weight that is dropped from the ship. The torpedo contains a thermocouple that measures the temperature. The torpedo has a very very long hair thin wire attached to it that sends the thermocouple voltage back to a computer on the ship. The torpedo has a well-calibrated descent rate, so the time since its launch is used to determine the depth of the probe at a given instance of time. Time is then converted to depth.
Here’s the data stream midway through the XBT launch. The horizontal scale is temperature (C) and the vertical scale is depth (m); here you can see the mixed layer of warm (well relatively speaking as few people would swim in 9C water) is only about 25 m thick here:
That XBT profile is then used to converted to sound velocity as a function of depth using some well determined relationship between temperature and sound velocity in seawater. The velocity profile is then used by the EM302 system when it converts the acoustic travel time to distances.
The image below shows Samer and the ship’s tech Darcy watching the bathymetry data come streaming in from the system. The upper monitor shows the swath returns from a single profile and the lower monitor shows the developing high resolution image of the rugged forearc bathymetry. This is mostly automated work so its not exactly fast paced, but despite that and despite Samer’s relaxed posture in the photo, I must have heard Samer say “I just saw the most amazing feature” or “this data is insane” at least one hundred times yesterday!