A broad-scale global array of temperature/salinity profiling floats, known as Argo, is planned as a major component of the ocean observing system. Deployment began in 2000. Conceptually, Argo builds on the existing upper-ocean thermal networks, extending their spatial and temporal coverage, depth range and accuracy, and enhancing them through ad dition of salinity and velocity measurements. The name Argo is chosen to emphasize the strong complementary relationship of the global float array with the Jason altimeter mission. For the first time, the physical state of the upper ocean will be systematically measured and assimilated in near real-time.

   Objectives of Argo fall into several categories. Argo will provide a quantitative description of the evolving state of the upper ocean and the patterns of ocean climate variability, including heat and freshwater storage and transport. The data will enhance the value of the Jason altimeter through measurement of subsurface vertical structure (T(z), S(z)) and reference velocity, with sufficient coverage and resolution for interpretation of altimetric sea surface height variability. Argo data will be used for initialization of ocean and coupled forecast models, data assimilation and dynamical model testing. A primary focus of Argo is seasonal to decadal climate variability and predictability, but a wide range of applications for high-quality global ocean analyses is anticipated.

   The initial design of the Argo network is based on experience from the present observing system, on newly gained knowledge of variability from the TOPEX/Poseidon altimeter, and on estimated requirements for climate and high-resolution ocean models. Argo will provide 100,000 T/S profiles and reference velocity measurements per year from about 3000 floats distributed over the global oceans at 3-degree spacing. Floats will cycle to 2000 m depth every 10 days, with a 4-5 year lifetime for individual instruments. All Argo data will be publicly available in near real-time via the GTS , and in scientifically quality-controlled form with a few months delay. Global coverage should be achieved during the Global Ocean Data Assimilation Experiment ( GODAE ), which together with CLIVAR and GCOS / GOOS , provide the major scientific and operational impetus for Argo. The design emphasizes the need to integrate Argo within the overall framework of the global ocean observing system.

 At present there are three models of profiling float used extensively in Argo. All work in a similar fashion but differ somewhat in their design characteristics. All spend most of their life at depth where they are stabilised by being neutrally buoyant at the "parking depth" pressure by having a density equal to the ambient pressure and a compressibility that is less than that of sea water. All ascend and descend by pumping fluid in and out of an extenal bladder. The floats are battery powered and are considerd expendable.

The three float models in use are the PROVOR built by MARTEC in France in close collaboration with IFREMER, the APEX float produced by Webb Research Corporation, USA and the SOLO float designed and built by Scripps Institution of Oceanography, USA.

Two temperature/salinity sensor suites are used - SBE , and FSI . The temperature data are accurate to a few millidegrees over the float lifetime. For discussion of salinity data accuracy please see the section on the Argo data system.

   As the float ascends a series of typically about 200 pressure, temperature, salinity measurements are made and stored on board the float. These are transmitted to satellites when the float reaches the surface.

   For most floats in the Argo array the data are transmitted from the ocean surface via the System Argos location and data transmission system (see also Service Argos Inc. ). The data transmission rates are such that to guarantee error free data reception and location in all weather conditions the float must spend between 6 and 12 hrs at the surface. Positions are accurate to ~100m depending on the number of satellites within range and the geometry of their distribution.

   An alternative system to Argos has been tested using positions from the Global Positioning System (GPS) and data communication using the Iridium and Orbcomm satellites. This allows more detailed profiles to be transmitted with a shorter period at the surface and even two-way communication.

   GODAE and CLIVAR UOP have agreed to form an Argo Science Team with the following Terms of Reference:

   1.Develop an Implementation Plan for a global network of profiling (temperature and salinity) floats, using the GODAE/UOP Prospectus and Workshop Report as representative of the CLIVAR and GODAE requirements.

   2.Provide scientific guidance to, and receive advice from, the Upper Ocean Panel of CLIVAR and the International GODAE Steering Team on the scientific and technical issues associated with the implementation of the profiling float contribution to the sustained (ocean) observing system of CLIVAR and the global ocean climate observing system of GODAE and GOOS/GCOS.

   3.Develop an international consortium, to undertake the implementation and maintenance of the global network, and provide advice to the consortium as necessary.

   4.Promote and evaluate observing system studies to guide the initial Argo sampling design and to guide the long-term development and evolution.

   5.Provide advice and guidance technical innovations relevant to the float array.

   6.Liaise as appropriate with other groups associated with the (sustained) global ocean observing system, including the ship-of-opportunity program, the tropical atmosphere-ocean array, and remote sensing program such as Topex/Poseidon and Jason.

   7.Provide regular reports on progress to the GODAE and CLIVAR International Project Offices.