|Home||Archived February 20, 2019||(i)|
Mechanical Arm + Internet = Realtime Profiles of Particles Near the Seafloor
We're not sure what to call it, but it moves, it sees, and it has a brain (of sorts). For a month, it lived 12 m deep about a mile off South Beach on Martha's Vineyard, Massachusetts. And it talked to us, because it was connected to power and Internet supplied by the Martha's Vineyard Coastal Observatory. The talking part was not all good…we liked getting the data and good reports, but because we could check in at all hours of the day and night, we felt like we had to. Kind of like texting your kids while they're out on a date. We were happy to find that it was well behaved.
"It" is a unique profiling system designed and built by a team of U.S. Geological Survey (USGS) engineers and technicians at the USGS Woods Hole Coastal and Marine Science Center in Woods Hole, Massachusetts. We deployed it to study how particles affect the attenuation of light in the water and to answer questions like: How deep does sunlight penetrate? Does enough light reach the bottom for seagrass or coral to grow? Will our laser bathymetry devices work? How far can a diver see? Can an ROV (remotely operated vehicle) see an oil plume? How much sediment is in the water? Are our sediment-transport models predicting the right amount? Do our optical measurements (based on light) agree with acoustic measurements (based on sound)?
To answer these questions, we needed to make both optical and acoustic measurements in the same place, at the same time.
The profiling system was deployed as part of the OASIS (Optics Acoustics and Stress In Situ) project, funded by the Office of Naval Research, with additional support from the USGS Coastal and Marine Geology Program. The profiler was designed to move a package of optical and acoustic sensors up and down through the 2 m of water just above the seafloor. After much debate, our final design uses a custom-built electric motor to screw a threaded rod through a nut on the end of a cantilevered arm. As the screw turns, the other end of the arm moves from about 2.2 m above the bottom to 0.3 m above the bottom at a rate of 11 to 12 cm per minute. The motor has incredible torque, considering that its power is limited to less than 4 amps at 24 volts, and the arm can easily lift the 50-kg instrument package. It starts profiling on the even hour and takes about 17 minutes to move from the top to the bottom position, where it remains until 20 minutes past the hour. Then it profiles to the top, waits until 40 minutes past the hour, and repeats another round-trip profile.
The recent deployment, from mid-September through mid-October 2011, was the first time we have had real-time data from sediment-transport instruments connected to a cabled observatory. The project has been managed by Marinna Martini, who designed the electronics of the arm-control system, programmed the controller, and oversaw all the other technical activities. Emile Bergeron designed (and redesigned) and built (and rebuilt) the mechanical components. He made most of the key parts by hand and specified the Technadyne Industries motor that has been so reliable. Jonathan Borden built the tripod, the housing for the electronics, and all the cables, and he runs all the logistics on deck during deployment and recovery. Ellyn Montgomery wrote the C programs and Linux scripts that actually control the arm and enable the various computers to communicate, and she modified code from Sam Laney of the Woods Hole Oceanographic Institution (WHOI) to make the computers log data in real time. Pat Dickhudt heroically managed to get the new holographic particle sizer running reliably and, with the help of Chris Sabens and Brandy Armstrong, set up most of the instruments on the tripod. Our summer Partnership Education Program (PEP) student Andy Klein helped us calibrate the motor speed and arm movement. Michael Casso, Sandy Baldwin, and Chuck Worley made several dives alongside WHOI divers led by Jay Sisson to check on the instruments and clean barnacles off the optics.
Optical profiles are difficult to make in the ocean because light is scattered and absorbed by water and particles in the water. Acoustics work much better, but sound and light respond differently to particles. Light responds to the area of particles in the water (proportional to the number of particles times the square of their diameters). Sound responds to the volume of particles in the water (proportional to the number of particles times the cube of their diameters). And, because particles can clump together to form aggregates that can later be torn apart by turbulence, the number and size of particles is always changing, even if no new sediment (or dead phytoplankton or other type of particle) is added or removed. In addition, the mass of particles near the bottom is always changing, as material settles from the productive surface layer, is transported from other places by currents, or is resuspended from the bottom by wave action.
Sound can travel much farther in the ocean than light, and so oceanographers rely on acoustics for many measurements, including profiles of particles. We'd like to be able to use those acoustic measurements to predict how light will behave (and answer the important questions listed at the beginning of this article). To do so, we need to compare the response of light and sound to clouds of particles at the same time, in the same place. From these measurements, physicists would like to be able to relate the scattering and absorption of light and sound to the size, number, and density of the particles. Geologists would like to be able to check our models of near-bed suspended-sediment transport. Ecologists want to know which particles are zooplankton and which are phytoplankton or detritus, and how they affect (or respond to) the amount of light that gets to the bottom. Divers would like to be able to predict near-bottom visibility from models, or at least from acoustic information. Acoustic profiles and point measurements have become easy to collect, but before we deployed our profiler, no one had made co-located profiles of both optical and acoustic properties in the bottom boundary layer (the part of the water flow that undergoes frictional slowing because of its proximity to the bed).
Our profiler has six optical sensors on the arm, three of which are standard for marine geologists: a 10-cm pathlength transmissometer (to measure how much light from a source is transmitted through the water; see http://www.whoi.edu/page.do?pid=8415&tid=282&cid=11489) and two single-wavelength infrared optical backscatter sensors (to measure how much light from a source is reflected back to its sensor by particles suspended in the water; see http://sfbay.wr.usgs.gov/access/wqdata/overview/measure/calib/Cal_tss.html). Two other sensors are more sophisticated (but nevertheless commercially available): a multiwavelength backscatter sensor with fluorometer (to measure chlorophyll), deployed by our University of Maine collaborator, Emmanuel Boss; and a laser-diffraction particle sizer. The sixth sensor is a prototype of a commercial laser holographic particle sizer. In addition to these optical sensors, we have also mounted on the arm an acoustic Doppler velocimeter (to measure currents and turbulence), an acoustic backscatter profiler, a CTD (conductivity-temperature-depth sensor), a dissolved-oxygen sensor, and an accelerometer (to monitor arm motion). Finally, Emmanuel Boss has put a water intake on the end of the arm; a pump on the tripod draws water through this intake and past an optical sensor that measures the amount of light absorbed and scattered at several wavelengths for both unfiltered and filtered (no particles) water. On the main tripod, we have six more optical sensors, four current meters to measure waves and near-bottom currents, and two more CTDs.
We deployed the profiler at the Martha's Vineyard Coastal Observatory on September 17, 2011, from the research vessel Connecticut, which has dynamic positioninga computer-controlled system that uses the vessel's propellers and thrusters to automatically maintain its position and heading. The Connecticut's dynamic positioning allowed us to place instruments on the bottom within approximately 3 m of a target. We needed that precision to ensure that our cable would reach the permanent underwater connection maintained as part of the Martha's Vineyard Coastal Observatory and to avoid the nearby instruments placed by our co-principal investigators, Norm Farr (WHOI), Paul Hill (Dalhousie University), and Tim Milligan (Bedford Institute of Oceanography).
Most of the instruments store data autonomously, and so we did not see the full dataset until after recovery of the profiling system on October 23, 2011. Three days before recovery, on Thursday, October 20, gale-force winds pummeled the Martha's Vineyard Coastal Observatory. We don't yet know how high the waves were because the observatory's wave gage had failed several days before. Sometime in the dark of Thursday night, the profiler's arm failed. As best we can reconstruct it, two failures took place—first, the coupling between the arm and the motor broke, and later the tripod was actually knocked over. On Sunday, October 23, we recovered everything under beautiful fall skies and on flat calm seas. Four of the optical sensors were broken, but all the data loggers apparently continued to work through the storm, and all the measurements were recovered.
We are currently analyzing these data, and what we have seen so far is exciting. Even before recovery, we were transferring about 1,600 images per day from the holographic camera and logging profiles of turbidity from one of the optical sensors. The turbidity profiles exhibit various shapes in response to upward mixing and downward settling of particles. By combining the images and turbidity profiles with the acoustic profiles and our measurements of bottom currents and stratification, we will be able to critically evaluate our models of resuspension and particle aggregation in the coastal ocean. These results, in turn, can be used in models for predicting coastal erosion, forecasting underwater visibility, modeling ecosystems, evaluating bottom habitats, studying transport of pollutants, and siting renewable-energy projects.
in this issue:
Realtime Profiles of Particles Near the Seafloor
|Home||Archived February 20, 2019|