The ABCs of the ADCP
Posted by sogasex on March 22, 2008
By Dave Hebert, University of Rhode Island
“What’s the ADCP showing?” This is one of the questions I get everyday when I get up, especially from co-Chief Scientist David Ho. He either wants to know where to inject the tracer (not where there are strong currents) or where the injected tracer has been carried (or, in oceanographic terms, advected) by the ocean currents.
One of the difficulties of doing oceanographic research is to know what is happening beneath the ocean surface. Unlike the atmosphere, we cannot see very far underwater. The ocean does not let electromagnetic waves (of which light is a type) travel very far in it. So, we are limited to sampling the near surface of the ocean, either through optics (see Richard’s blog on Mar 9th) or by taking surface water samples though the underway system. To get an idea of what is happening at depth, the ship normally has to stop and make a CTD cast (see Pete’s blog on Mar 18th). We would love to make measurements of ocean remotely but the ocean is opaque to most of our measurement techniques. However, there is one technique that we can use, and has been used by mammals for millennia, that is acoustics or sound.
In an earlier blog (Mar 16th), Juan mentioned the mapping of the sea floor topography using a system called multibeam. Like radar, the multibeam sends out an acoustic pulse that gets reflected off a target, the sea floor, and returns to the ship. The time that it takes for the sound to travel to ocean bottom and back, tells us how far the bottom is away from the ship. Thus, we can produce a map of sea floor as we steamed back and forth over an area.
Another use of radars is the ability to tell how fast an object, like a speeding car, is traveling. As the light (or sound) bounces off a moving target, the frequency of the light (or sound) changes. This is why the pitch of a train whistle increases as it approaches you and decreases as it moved away. This Doppler shift is also used to estimate how fast the universe is expanding by astonomers. In the ocean, we can use this principle to estimate how fast the ocean is moving at different depths. If we look at the sound returning to the ship after a certain time after we send a sound pulse out, we are probing the ocean at a specified distance (depth) away from the ship. Now, if the pulse is reflected off particles (believed to be zooplankton, tiny little animals) moving with the ocean current, the observed Doppler shift will tell us the relative motion of those particles to the ship. Using several sonic beams in different directions, we can obtain the ocean current in all three directions. However, given the small vertical motions and uncertainties in the calculations, only the horizontal velocities are reliable.
On the Ronald H Brown, we have a Teledyne RD Instrument’s Ocean Surveyor Acoustic Doppler Current Profiler, commonly referred to just as the ADCP. This ADCP can work in two modes: ‘broadband’ which provides higher vertical resolution data at the expense of depth of penetration and a ‘narrowband’ mode which provides velocity data down to approximately 1000 m. We are using a data acquisition system developed at the University of Hawaii that allows us to use the ADCP in both modes at the same time by alternating between broadband and narrowband pulses. The software collects and stores the data. Computer routines calculate the ship’s speed and direction from GPS data and process the ADCP data to provide absolute ocean currents averaged over 5 minute periods. This data is processed every 30 minutes (we could do it more often but the ocean doesn’t change that quickly) and provide plots of east-west and north-south currents of the last 36 hours (see below). We can also use this data to predict where the injected patch has moved assuming that the currents we observed represent the currents felt by the tracer. For example, the trajectory of the injected tracer (51.15°S, 38.48°W) is predicted to follow the path shown by the red line below starting at the asterisk over the time period shown for the ADCP data.
Timeseries of currents from the ADCP (broadband)
Timeseries of currents from the ADCP (narrowband)
Trajectory based on ADCP