TURBULENCE @ OCEAN OBSERVATORIES

TURBULENCE @ OCEAN OBSERVATORIES

Research funded by

Browse the LEO-15 data set

Turbulence and observatories

Vertical turbulent transports of momentum, mass, chemical species and particles play major, often dominant, roles in a range of processes spanning all the sub-disciplines of oceanography – processes as fundamental and diverse as sediment resuspension, biological primary production, and particle/contaminant dispersion, to name but a few. Significant turbulent transfers are strongly variable in time, not only in estuarine and shelf environments, but also in offshore surface and bottom boundary layers, and in tidally-driven deep-sea regimes. Attempts to understand such fundamental questions as the response of marine ecosystems to anthropogenic forcing, whether imposed on local (ie changes in coastal environments caused by land-use changes) or global (ie climate change) spatial scales, require the characterization of turbulence over the full range of operative time scales. Ship-based observational programs, typically of order a few to several weeks, can't provide the information necessary to fully describe the modification of turbulence and its effects by seasonal or longer term variability of atmospheric forcing, much less by highly episodic, extreme events that may dominate such processes as sediment resuspension and on/offshelf tranports of bio-active material.

Present techniques for measuring ocean turbulence are often extremely labor-intensive (hence expensive), and use sensors that are highly sensitive to marine fouling. As a result, these techniques are not suitable for the long-term monitoring that is needed to support studies of the role of turbulence in both natural and anthropogenic time variability of marine systems. What is needed are turbulence tools designed for deployment at long-term ocean observatories, cabled or moored platforms that can provide required levels of power supply and data transmission rate. Ideally such tools would be relatively insensitive to biofouling, run continuously and cheaply with minimal human intervention, and provide estimates of turbulence parameters over a range of depths, rather than at a single point. A system consisting of a single 5-beam acoustic Doppler current profiler (VADCP) can potentially provide this essential ability, a potential inherent in a number of recently developed techniques which use standard acoustic Doppler measurements in non-standard ways. Our work aims to move these new techniques out of the low-noise, high-signal environments in which they have been developed into coastal and deep-water open ocean environments.

Want more information?

For the fundamental importance of turbulence to a wide variety of ocean processes, see Ocean Sciences at the New Millenium www.ofps.ucar.edu/joss_psg/publications/decadal ), a report of the National Science Foundation 2001.

For more information on ocean observatories, see Illuminating the Hidden Planet: The Future of Seafloor Observatory Science www.nap.edu/books/0309070767/html/), a report of the National Research Council, 2001.

Why cabled observatories first?

Our first step is deployment of a VADCP at a cabled observatory, a subsea junction box (node) that is connected to shore by an electro-optical cable buried under the sea floor. Cabled deployment is essential at this stage because of the power required and data rates produced by continuous operation of the Doppler system. Continuous operation over several months is highly desirable in order to provide multiple realizations of "normal" turbulence-generating mechanisms, such as storms, tidal flows etc, while even longer records may be necessary to illuminate the relative importance of episodic extreme events. Eventually, extensive experience with temporally well-resolved fields at cabled observatory sites will make it possible to suggest and evaluate statistically meaningful measures of turbulent and ecosystem quantities that can be made under the power and storage constraints associated with non-cabled systems.

Deployment at LEO-15

From April 25 through October 31 2003, the VADCP was deployed at the outermost (B) node of LEO-15 (http://marine.rutgers.edu/mrs/LEO/LEO15.html), a cabled observatory off the coast of New Jersey that is operated by NOAA's Mid-Atlantic Bight National Undersea Research Center and Rutgers University.

The shore station for LEO is the Rutgers University Marine Fieldstation (http://marine.rutgers.edu/rumfs), located in a refurbished Coast Guard Station, ......

......, near the mouth of Great Bay near Tuckerton, New Jersey .

The topography (or lack of it!) seen in these above-water photos continues offshore, so despite being at the end of ~ 7 km of cable, Node B is in only 15m of water. This water depth allows resolution of the entire water column with the range of a 1.2MHz ADCP.

The location of the LEO observatory is characteristic of the very gently sloping continental shelves that extend along most of the eastern seaboard of the United States, in places extending 100s of km offshore. LEO is also characteristic of strongly surface-wave-influenced shelf environments with significant supply of sediment from the bordering land. The gently undulating subsurface topography that surrounds Node B can undergo substantial re-organization as bottom sediments remobilize during storm events. For this reason, "bottom-mounted" instruments are actually attached to pipes driven ~ 2 m into the bottom.

A package containing a 1.2MHz 5-beam VADCP was mounted near the edge of the area (~ 25 m radius) around Node B that is protected by guard buoys, and connected to the node by a short underwater cable. At installation, the height of the transducers off the bottom was ~ 0.6m.

The water column at LEO is unstratified during winter, but becomes strongly stably stratified during the summer heating season. Proper interpretation of turbulence measurements in a stratified fluid requires a quantitative measure of stratification, which was provided by the combination of a nearby moored string of closely spaced thermistors, sampling frequently in time, and a profiling CTD at Node B, operated on an irregular schedule.

Ancillary data includes pressure from the node and a full suite of meteorological data from a tower located at RUMFS, directly on-shore from the LEO nodes.

By choosing a date during the deployment, you can view selected ancillary data and water column fields derived from the VADCP measurements, starting with the first full-depth VADCP file (Note 1) recorded on that date. Because LEO-15 is a wave-exposed location, the raw beam velocities (Note 2) from all 5 beams are dominated by the velocities associated with surface gravity waves. Fortunately, because surface waves pass the moored instrument at a phase speed that is much larger than the mean current which adverts turbulent structures, these unwanted velocities may be removed by simple low-pass filtering. The fields shown are “raw” surface-wave-removed fields resulting from application of a 9^th order Butterworth filter with frequency cutoff corresponding to ~ 1/3 the peak surface wave frequency. However data collected over a time interval T can be filtered only for those bins (depths) below the minimum water surface (including both tidal and surface wave displacements) that occurs during T. In order to avoid losing too much near-surface data, the filtered data shown use T = half the file length, so a discontinuity in the number of bins available for display may be visible at the middle of a file.

Note 1: The VADCP was normally configured to measure over the full depth of the water column. However during three periods ( June 03 through June 15, July 26 through August 2, September 26 through October 04), it was reconfigured remotely to obtain higher vertical resolution data (not shown) over the bottom half of the water column every second day. Switches between full and half water column measurements took place daily around 1000h EDT, so during these periods, full-depth data is available over only part of a calendar day.

Note 2: Beam velocity = an estimate of the water velocity parallel to a Doppler beam, positive if towards the transducer. Beam velocities can be put together in various ways to produce estimates of mean velocity components, as well as a number of turbulence characteristics.