Return to D'Asaro Home

OCEANOGRAPHIC FLOATS

Subsurface drifting floats have been one of the major types of oceanographic instruments since Swallows's pioneering work. These devices float within the ocean, like a balloon, and follow the motion of the water at their level. Their motion can therefore map out ocean currents. Float technology has advanced on two fronts. Floats have become cheaper and easier to use so that more of them can be deployed. For example, project ARGO aims to place thousands of floats throughout the world ocean in order to provide continuous long-term monitoring. Floats have also gained to ability to follow water more accurately and measure more quantities along their drift track. We have emphasized this second approach, working to develop new ways of using floats.

Lagrangian Floats

At APL/UW, we have pioneered the use of floats to measure ocean mixing rates particularly in regions of very strong mixing such as the near-surface boundary layers, under storms and hurricanes, in deep convection in the polar regions and in strong coastal flows. The mixing is crucial in coupling the atmospheric and oceanic circulations and transferring quantities between them. In such regions, water parcels repeatedly cycle vertically across turbulent layers tens to hundreds of meters thick, working to homogenize them and carrying heat, salt and other properties across them. The floats aim to follow the motion of this water while measuring its properties, thus measuring both the intensity of the mixing and fluxes that it carries. We call these instruments "Lagrangian Floats" after the French mathematician Lagrange who first described the equations of fluid motion in a water-following frame. Our floats are not perfectly Lagrangian, but we try to make them as Lagrangian as possible.

The key to constructing such floats is to ensure that the float's density accurately matches that of the seawater in which they float. A mass of water, like other masses, moves at a constant velocity unless acted upon by external forces. The forces are due to pressure gradients from the surrounding water and the force of gravity acting on the parcel. Forces due to friction are very small unless the parcel is very small. If we make a float with the same density as water, it will experience the same gravity and pressure forces as the water that it displaces and thus accelerate in the same way. The trick is to maintain the float's density to match that of seawater as the float moves. The most important factor, as realized by Tom Rossby years ago, is to match the compressibility of the float to that of seawater. We do this by carefully designing the float's hull for the correct compressibility and measuring the compressibility in a ballasting tank. See D'Asaro et. al, 1996.

Real floats can match the density of seawater only imperfectly. Our floats are typically a gram light or heavy. (That's not bad shooting given that the total weight of some floats is about 50 kg!) This will cause them to rise or fall relative to the water. We minimize this velocity by adding an approximately 1 square meter drogue to the float. This reduces the float's motion to well below the typical vertical velocity of the water in turbulent regions (~ 1 cm/s) allowing useful measurements to be made.

Species of Lagrangian Floats and Some Scientific Results

Our first float (image at right) was made in 1981 in cooperation with David Farmer of IOS/BC and with the generous support of ONR for this novel and probably crazy idea. The float was acoustically tracked from a floating acoustic platform, which also imaged bubble clouds. You can see the tracking hydrophone at the bottom of the float. The drogue was made of perforated aluminum. It, and a weight, fell off to make the float surface. The drogue was retained on a light wire.The idea was to use the bubble clouds to define the structures of flow and the float to show the velocities of these structures. The float measured pressure and temperature and acoustically telemetered these to the platform. This wasn't very reliable! We made measurements in wind forced ocean boundary layers and in strong turbulence in tidal channels in Haro Strait, BC (Farmer et. al, 1995).

MLF

Improved versions of this float, the MLF or Mixed Layer Float, (image at left: CLICK on the image for a larger figure with annotation) was built and used in a variety of locations from 1993-1995 including Knight Inlet. Pressure, temperature on both the top and bottom, and spin rate were measured every 5 seconds and logged internally. The float was acoustically tracked using GPS equipped, acoustic buoys designed by David Farmer's group. This allowed us to measure three-dimensional Lagrangian velocities, heat fluxes and vorticity (from the spin rate) in turbulent flows (Lien et al,1998) and in stratified flows (D'Asaro and Lien 2000) and connect the observed spectra to classical ideas about turbulent spectra and internal waves. We infer the vertical velocity of the water from the rate of change of pressure measured at the float. This measurement is insensitive to the very large vertical velocities due to surface waves. Floats are thus a good platform for measuring turbulent properties near the ocean surface where surface waves usually corrupt the measurements. In wind forced mixed layers, we find a remarkably good correlation between vertical kinetic energy and wind stress (D'Asaro, 2001). The figure on the top of this page shows the simultaneous launch of three MLF's in a wind-forced mixed layer to study the rate of horizontal dispersion in the mixed layer.

DLF

The major shortcoming of the above floats was that they needed to be tended by a ship. This greatly limited the amount of data that could be taken and made operating the floats people intensive and therefore expensive. The Labrador Sea Deep Convection Experiment offered an opportunity to build fully autonomous floats which built upon our experience with the MLF. The result was the DLF or Deep Lagrangian Float, shown on the right. This is small float, with only 15L of displacement. Temperature and pressure data is recorded internally. The float is positioned a few times a day using the long range RAFOS system (the big black cylinder on the top of the float is the hydrophone). A small piston extruded from the bottom of the float allows it to change its volume and thus ballast itself to the density of the surrounding water. The (white) cloth drogue folds at deployment and for autoballasting. The drogue and mechanism falls off at the end of the mission, causing the float to surface and send its data via the ARGOS satellite system, a very slow, but effective data telemetry system. The spikey thing on top of the float is the ARGOS antenna.

The major innovation in DLF is the hull. The float needed to match the compressibility of seawater to pressures of 2000 db. Simple cylindrical hulls with a large enough compressibility will fail at pressures far below this. One solution is to use ring-stiffened cylindrical hulls, which have rings placed periodically along the hull to prevent buckling (i.e. the sides will collapse inward but the metal does not fail. This is an instability and depends on the geometry of the hull and the stiffness of the metal, but not on its failure strength). The rings effectively turn one long cylindrical hull into several shorter ones. DLF used a variant on this, machining an aluminum tube into a series of rings connected by cylindrical arched bays. The rings prevent buckling of the hull and the arches transfer the pressure forces acting on the hull to the rings while maintaining a nearly constant stress within the material. The resulting hull is strong, compressible and very light since most of the unnecessary material has been removed. If the distance between the rings is small enough, the hull fails when the stress in the arches exceeds the strength of the material rather than by buckling. This is desirable because material failure, unlike buckling, is easy to model and predict. The hull was therefore constructed of 7075 Aluminum which is about twice as strong as the more common 6061-T6 alloy. It is, however, difficult to obtain in tube stock. A stock of 7075 tubing was salvaged from vintage 1970's deep sea pressure cases donated by numerous colleagues. The final hull design was achieved through a combination of numerical modeling and destructive testing. This same hull concept has more recently been used in the UW Seaglider, now a commercial product.

DLF's were used sucessfully in the Labrador Sea Convection of experiment. These data are described in more detail here. Many of them crossed the North Atlantic and washed up on the beaches and rocky shores of Europe and were returned to us. These are tough instruments! Three of these were retrofurbished and deployed in Hurricane Dennis. Others were deployed in the Equatorial undercurrent to measure mixing there. Others are being used in studies of Puget Sound.

MLF2

Our most recent float is MLF2, a second (or third?) generation mixed layer float. Our experience with MLF and DLF showed that although we could learn a lot from temperature and pressure measurements, these alone were insufficient to diagnose the dynamics of oceanic turbulent boundary layers. Measurements of shear, stratification and surface waves were needed. Furthermore, accurate ballasting of the floats under conditions of strong mixing in general requires knowledge of the density of the seawater, so salinity measurements were needed. Also, there seemed to be a great potential for studying biological phenomena in the upper using optical sensors. Finally, advances in satellite communications offered the possibility of two-way data transmission so that floats could be controlled and their mission parameters modified remotely. All these factors pointed toward the construction of a larger float with lots of power and the capability of easily accommodating a variety of sensors.

MLF2 is about the same size as MLF (1 m), but with a larger displacement (50L). It has much more limited depth capability than DLF (300 db, extendable to 500 db), but has the ability to surface repeatedly using 750cc of active volume control. The buoyancy control is again accomplished using an extruding piston. Like DLF, it has a cloth drogue, but this drogue can be folded and unfolded as many times as desired. The MLF2 mission consists of periods of Lagrangian drift typically 12-24 hours long. Between these, the float surfaces, uses GPS to determine its position and transmits its position and some of its data using the Iridium system. It then waits for instructions before beginning another drift period. MLFII can carry a large instrument suite, including a Doppler sonar, altimeter, CTD's, accelerometers, PAR sensor and fluorometer. Storage of this data requires several hundred megabytes, so the float must be recovered to retrieve the bulk of the data. Early MLFII depoyments were used in a study of the wintertime North Pacific mixed layer in 2000 and in studies of upwelling off Oregon during 2000 and 2001. An air-deployment package has been used for further deployments in hurricanes in 2002, 2003 and, most successfully in 2004 during CBLAST. We have resumed Hurricane work in 2008 and 2009 with deployments in Typhoons expected in 2010.

The sensors on MLFII allow a much better understanding of the behavior and accuracy of floats. For example, in a stratified fluid, the dominant cause of drag on the float is the radiation of internal waves. This internal wave drag is much larger than the normal form drag at low speeds and causes the floats to be Lagrangian for frequencies higher than about N/30, where N is the buoyancy frequency. Since most mixing events in a stratified fluid happen on time scales of about N, this means that floats can be good instruments to measure mixing rates in the ocean interior. These results are described in D'Asaro, 2003.

This understanding has allowed us to develop a variety of techniques to measure mixing rates simply and for long periods of time from floats. The first idea is that turbulence moves the floats vertically at high frequencies; we can measure this using pressure and deduce the energy flux through the turbulence and thus the mixing that it causes. Similarly, rapid fluctuations in temperature, salinity or density measured by the float's CTD allow us to measure the rates at which these quantities are mixing. Another method is to look at the combination of velocity and temperature to compute the vertical heat transport (D'Asaro, 2004).

Chemical and Biological Applications

Floats, particularly MLFII, is well suited to carry many of the newly developed biological and chemical sensors and thereby begin to address the highly complicated biogeochemical interactions with circulation and mixing. I plan to devote considerable effort to this in coming years.

Since floats move with the water, they offer the intriguing possibility of measuring biological and chemical rates directly from the local rate of change. Some measurements off Oregon were used to explore this possibility followed by work in Carr Inlet, Puget Sound as part of the MIXED experiment which were analyzed by graduate student Eric Rehm.

The first major sucess has been the measurement of air-sea gas transfer rates at hurricane force winds in Hurricane Frances in cooperation with Dr. Craig McNeil.

Accidental biology was sampled during equatorial measurements in April 2007. We added a camera to a float to figure out what was resting on it at night and making it heavy. The answer was fish! This picture was taken with a camera on the float and sent to us via satellite.

The next big step in this area was the 2008 North Atlantic Bloom Experiment, in which 4 Seagliders navigated around 2 Lagrangian floats. The ML2 float carried a heavy payload of physical, biological and chemical sensors in order to redundantly measure a wide suite of variables in this complex system. It was a difficult task to get all of these sensors working well! Click on the image to see an annoated image.