Book Review ocean engineering technology

OceanTech: profiling the sub-surface via Argo floats

This theme week on has focused on ocean technology with posts about cataloging biodiversity with ROVs, AUVs, and HOVsevolution of underwater photographydrilling technologies, and old school ocean technology (#throwbackthursday). The last post of theme week introduces a technology that enables continuous, real-time data collection of the ocean sub-surface called the Argo global float array. You may have heard about Argo data in previous Oceanbites posts: One person’s noise, is another person’s dataCan the ocean take the heat?Sailing the Southern Ocean for scienceTracking the movements of a heavily fished Fijian sharkSea Ice Plays a Key Role in Ocean Circulation, and Sensors probe oceans for answers!

What is the Argo Array?

Figure 1: an army of Argo floats in storage room (

The Argo array is a global distribution of free-floating instruments (Figure 1) that measure temperature (T), salinity (S), and velocity of the ocean’s sub-surface.  Argo is great because it provides almost real-time data (within 24 hours) of the ocean’s upper profile (<2000m).  It is a means to collect data that  cannot be obtained from satellites alone and that require substantial time and money to measure entirly from shipboard casts.

The Argo array effort is distributed among more than 20 nations that include 50 research and operational agencies, each with varying degrees of involvement.

It is an extremely advantageous program because of the international collaboration that enables global coverage, the fact that all of the data is available to the public, and the shear volume of data collected.  To put it in perspective, about 6000 shibboard-CTD casts are made in year, where as with Argo, an average of 87,500 water column T and S profiles were measured annually between 2004 and 2008.

How does Argo work?

The Argo array is designed so that floats are deployed in a 3 degree by 3 degree grid between 60 degrees N and 60 degrees S.   The original program was designated a 10 year evaluation period and the initial goal was to deploy 800 floats per year.

Figure 2: Argo float Cross section (

The floats are designed to measure T, S, and velocity in the water column.  They are cylindrical, ~1.3 meter tall, and equipped with components such as an antenna, temperature and salinity probes, a battery, and a bladder (figure 2). The antenna is used to position the float, the probes are used for data collection, and the bladder is used to raise and lower the instrument in the water column.  The depth that Argo floats descend to is a function of density and pressure.

Floats are deployed (video of deployment) off multiple boat types based on convenience, including research vessels and commercial ships. For areas that are isolated special missions of either boat of aircraft are planned.

Figure 3: Argo operation (

Once a float is deployed it is designed to complete 150 profiling cycles similar to the one illustrated in figure 3.  To summarize a profiling cycle:

1) After a float is deployed and has positioned itself, it sinks to a given pressure level; the decent takes about 6 hours.

2) After drifting for 10 days in the sub-surface the float ascends by filling the bladder with water (decreasing the density of the float), and collects T and S data during the 6 hour ascent.

3) At the surface the float positions itself with satellites and transmits the data.

4) Data is transmitted via the Global Telecommunications System and is monitored by the Argo Information center. All data is available on the Global Data Assembly Centers and about 90% of data is available to the public within 24 hours of transmission (figure 4).

Figure 4: Real time data flow (

Design obstacles:

Designing an instrument that will be subject to harsh ocean conditions comes with a set of challenges.  First off, the floats are exposed to a variety of temperatures (30 degree C range) and pressures (200 atm range), so they need to be rugged.  A second challenge to engineering is minimizing the need for sensor calibration over the floats life time so that they are reliable and the data they collect has little to no error or drift.

At a 2003 meeting to discuss preliminary progress of the array, an initial issue identified with the floats was a battery life of less than 2 years, which was much less than the target of 4-5 years (~150 cycles) (Gould et al., 2004).  The problem was addressed, and in 2008 68% of floats deployed in 2004 by the United States were still active (Roemmich et al., 2009).  In addition to improvements in design, the number of cycles a float was able to achieve was prolonged by positioning to the Iridium satellite network as oppose to the Jason satellite network (Roemmich et al., 2009) .  This switch reduced surface time form 10 hours to just minutes.  The reduction is surface time lessens the potential impacts of vandalism by organisms and the risk of the float drifting into shallow areas.

Regardless of latitude, the Argo array floats have encountered problems.  Floats in low latitudes were limited to profiling the upper 1000m of the subsurface because they were not able to overcome the stratification of the water column. An improvement to the design to reach depths of 2000m was accomplished by creating a buoyancy controlled float by the use of a gas cylinder and carbon-fiber casing (Roemmich et al., 2009).  A challenge in high latitudes is the floats becoming trapped underneath ice.  A solution to this problem has been to design program that will allow data to be stored until satellite transmission can be achieved (Roemmich et al., 2009).

Key dates in the Argo timeline:

1990-1998: Argo was developed as part of the World Ocean Circulation Experiment.

1999: The first floats were deployed by Australia into the Indian ocean.

2001: 294 floats were deployed.

2003: International meeting of 200 scientists to examine preliminary results.

2004: Goal of 1000 deployed floats was met (on May 7, 2004 there were 1171 floats).  100,000 water column profiles were made that year.

2004-2008: 350,000 sub-surface profiles were collected via Argo floats (Roemmich et al., 2009).

2007: The final goal of 3000 deployed floats was reached.

2014: First deployments of Deep Argo floats in the Pacific Ocean (these reach 6000m!).

May 18, 2017: 3936 floats are deployed globally (Figure 5).

Figure 5: Argo floats May 18, 2017 (3936) (

Future of Argo:

As technology continues to improve so will our use of it.  Advancements to expect include secondary CTD’s that can measure profiles between the sea surface and 5m (Roemmich et al., 2009), higher precession and quality data with minimal errors and standardization between nations (Roemmich et al., 2009), wider spatial coverage, and additional sensors (Gould et al., 2004). Additionally, we will hopefully see a greater abundance of Deep Argo floats.

Figure 6: Argo projects 2017, 10 papers per page and over 25 pages for 2017 alone.

Why is the Argo array important?

Argo floats are important for a multitude of reasons including the sub-surface information they collect, their spatial coverage, the accessibility of the data, and the quality of the data.  Data collected with the Argo array has been utilzed to better understand the air-sea interactions that control climate and heat distribution, changes in the hydrologic cycle, the evolution of surface conditions during cyclones, water mass formation, and circulation (Gould et al., 2004; Roemmich et al., 2009).  A major advantage of the Argo array providing a long-term data set is that dynamics can be investigated on seasonal and decadal time scales (Roemmich et al., 2009).  An advantage of the Argo array providing spatially distributed sub-surface profiles is that information can be coupled with global scale satellite data related to sea surface temperature and height (Roemmich et al., 2009), for examples, to better understand whole system interactions between the atmosphere and hydrosphere.   On a smaller scale the profiles collected with the Argo floats can be combined with in-situ data sets to provide a more complete picture of a selected region (Roemmich et al., 2009).   The advantages of the public accessibility of the Argo array data is evident with a basic Google Scholar search for Argo array, over 2500 hits of peer-reivewed publications came back for just the year 2017 (Figure 6).

A fun way to access the data:

Figure 7a: Google Earth layer enables the selection of individual floats, the track can be displayed and additional float information can be accessed.
Figure 7b: float information available when a Google Earth target is selected, the information panel for float 4901733 off the coast of California is displayed

If you have Google Earth, then you may have a bit of fun playing around with the Argo float data. You can download the file from the UCSD website.  With the layer employed, select the float of your favor. You can choose to view the trajectory (figure 7a), or you can link to the data page (Figure 7b) to download a CVS. and make your own plots (Figure 7c).


Figure 7c: Data from the Internet, or the Google Earth layer as done here, can be downloaded as a .CVS and data plots can be made in jiffy! Data for float 4901733 off the coast of California is displayed.


Gould, J., Roemmich, D., Wijffels, S., Freeland, H., Ignaszewsky, M., Jianping, X., … & Takeuchi, K. (2004). Argo profiling floats bring new era of in situ ocean observations. Eos, Transactions American Geophysical Union, 85(19), 185-191.

Roemmich, D., Johnson, G. C., Riser, S. C., Davis, R. E., Gilson, J., Owens, W. B., … & Ignaszewski, M. (2009). The Argo Program: Observing the global ocean with profiling floats.

Leave a Reply

Your email address will not be published.