Martin Kramp SailingOne, 6 Place de la République, 14000 Caen, France
Fabienne Gaillard IFREMER Technopole Brest-Iroise 29280 Plouzané, France
Pierre Blouch Meteo France 13, Rue du Chatellier, CS 12804 29228 Brest Cedex 2, France
Peer Fietzek IFM-GEOMAR Düsternbrooker Weg 20 24105 Kiel, Germany

Abstract – The aim of the OceanoScientific® Programme and its SolOceans One-design Class is to collect and transmit scientific data from the ocean-atmosphere interface during regularly starting offshore sailing races. Data collected on board the first vessel proved to be of good quality. Thus the first important step towards the introduction of an accepted as well as highly valuable platform for ocean surface and atmospheric parameter acquisition has been taken and its serial production can begin.

Introduction

Ocean races lead fleets of sailing vessels all around the planet. Routes hardly change and there are regular starts every year. Major parts of those races take place between and south of the continental capes (Horn, Good Hope and Leeuwin) where data from the ocean-atmosphere interface are both, rare and crucial for scientific projects such as CLIVAR (International program on CLImate VARiability and predictability) and GOOS (Global Ocean Observing System). Federating the efforts of scientists from different French institutes (IRD, CNES, CNRS, IFREMER, INSU, IPEV, OMP), the Sea Surface Salinity Observation Service gives an overview of regular routes of Ships of Opportunity (SOOP programme) at [1].

Racing yachts have been equipped with scientific sensors before, but the possibilities were always very limited because of the competition and onboard conditions.

In 2006, the French Sailing Federation (FFVoile) launched the SolOceans race, which from the beginning combined the sportive aspects of a sailing race in the Southern Ocean with the scientific need for data from said area.
The SolOceans One-design was created for this challenge. Designed by the famous Finot-Conq Group fully in carbon, it is a 16 meters long high tech vessel for single-handed racing that allows for the deployment of various ocean and atmosphere sensors (see Fig. 2), following French Grenelle de la Mer’s commitments of a Future Vessel (Navire du Futur) [2].

At the fifth session of the Ship Observations Team (SOT-V) of the Joint Technical Commission for Oceanography and Marine Meteorology (JCOMM) of the World Meteorological Organization (WMO) and the Intergovernmental Oceanographic Commission (IOC), the OceanoScientific® Programme was introduced to the international scientific community [3].
The fleet of SolOceans One-designs forms its own class and will take part not only in the SolOceans, but also other major offshore races.

The mixture of scientific expeditions and an ocean race offers new opportunities to report on climate change and other environmental challenges beside the fact that high quality data is collected. It enhances public awareness for scientific questions, not only within the Race Villages at departure and arrival or during virtual races but also by educational programs.

Partners

The project already has various partners from both science and industry, who come together for general meetings twice a year. The number of partners and the constitution of the project are not restricted and it is hoped to welcome further partners in the future.

Science

In France IFREMER [4], INSU-CNRS and Meteo France are partners of the project, together with German IFM-GEOMAR and supported by the French and European space agencies CNES and ESA [5].

Together, these partners define the parameters of the OceanoScientific® Programme, share their expertise to design the equipment, validate the emerging data and contribute it to international data networks such as the Global Telecommunication System (GTS) of the WMO.

Industry

SailingOne, based in the French Lower Normandy Region, is specialized in ocean racing. SailingOne equips all SolOceans One-designs (hulls being built by nearby JMV Industries), manages all maintenance service and organizes the SolOceans races.

German SubCtech, a recently founded CONTROS spin-off, is specialized in complex flow-through-systems with years of experience in underway technology.

French Mer Agitée is the technical team of world-class sailor Michel Desjoyeaux and participates in the project by its new hydro generator, source of emission-free energy onboard the SolOceans One-design Class.

Parameters

The general idea guiding the selection of parameters by the scientists is: The availability of qualified, stable, compact and low power sensors and the possibility of integrating the OceanoScientific® data set in an existing or intended worldwide observing system. The system development is conceived of several steps of increasing complexity.

Step One – Initialization (in 2006)

• Wind direction and speed • Atmospheric pressure • Air temperature and humidity • Sea surface temperature and salinity • Sea surface partial pressure of carbon dioxide (_pCO₂)

Step Two – Additional parameters (Starting 2010)

• Solar radiation • Fluorescence • pH

Step Three – Additional parameters (Starting 2012)

• Plankton • Nutrients • The technical infrastructure is very flexible and further parameters can be added or exchanged with others

Technology

The ensemble of systems installed onboard the SolOceans One-designs was named OceanoScientific® Kit. It consists of the sensors and their necessary infrastructure. During the maintenance service of the vessel, which occurs at least once a year, all scientific instruments are regularly replaced or calibrated, which is crucial for constant high quality data.

Ocean

Realizing a suitable water intake for the SolOceans One-designs was the first challenge to meet. Similar measurements from buoys and floats are performed with the sensor submerged at the measurement depth, but most of the time SolOceans One-designs are moving very quickly with top speeds exceeding 25 nautical miles per hour and operate at the uppermost ocean layer. External buoy systems were found to be too vulnerable and penalizing for the ship’s speed.

On other SOOP or Research Vessels, the water is taken at a depth of several meters (2 to 20) and pumped to the sensor installed inside the ship. This internal solution was chosen, but the type and size of the SolOceans One-design imposes technical limits on weight, size and power-consumption, even if those are not an issue in terms of competition within the one-design concept (i.e. all starting vessels carry the same instrumentation package).

Taking water through the keel using the pressure generated by the speed was considered at first, but in order to reduce the list of the vessel to a minimum and optimize the performance, its canting keel can be moved up to 40° to the windward side. That results in the usual draft of more than 4 meters being reduced to almost zero in certain conditions with a hardly submerged intake for the seawater flow-through system.

Tests performed with IFREMER and INSU-DT in November 2008 invalidated this concept: There was either no flow at all or the rate was not stable enough. As soon as the vessel accelerated, listed or pitched in the waves, the circuit was taking in great amounts of air, which is fatal for the data quality of most ocean sensors.

An industrial partner finally realized the development of a pumped system that suits the SolOceans One-design. With a weight of less than 30 kg SubCtech’s Micro-OceanPack Racing comes with:

• Pump • Debubbler • Flowmeter • Datalogger • Plug & Play interface for various sensors • Sediment trap

Table I:
DATA COMPARISONS (S AND T): TIME AND POSITION (OCT. 26TH, 2009)
	Time	Latitude N	Longitude W	Description
1	14h15	48°21.5′	004°31.3′	Between naval base and Saint-Anne Lighthouse
2	14h37	48°19.9′	004°36.2′	Near Petit Minou
3	16h03	48°17.64′	004°38.00′	Near Toulinguet
4	16h36	48°20.99′	004°33.62′	Near Marel buoy

Transmission option (Iridium or Inmarsat).
The self-priming pump and a new debubbler design with bypass enable a debubbling-procedure without gas-exchange and a stable flow-rate (adjustable, up to 10 litres per minute). Even in very rough offshore conditions, lists of more than 20° and even while the vessel is surfing at full speed, these conditions can be maintained. With a second bypass, the flow-rate can be reduced for one or several sensors if necessary.
The first ocean sensors deployed are: Sea-Bird SBE45 (salinity and temperature) CONTROS HydroC/CO₂ (_pCO₂)

Additional sensors will be installed in 2010, such as a modified CARbon Interface Ocean Atmosphere system (CARIOCA, [6], [7]) for measurement redundancy and inter-comparison with the HydroC/CO₂ unit. Partly the new instruments are still under development, such as an outside sea temperature sensor designed especially for this application.

Atmosphere

Table II:
DATA COMPARISONS (S AND T): RESULTS (OCT. 26TH, 2009)
	SBE 45 T (°C)	TFX 392 T (°C)	∆T (°C)	SBE45 S(PSS)	LPO S(PSS)	∆S (PSS)
1	14.863	14.8	0.0<∆<0.1	35.267	35.288	-0.021
2	14.854	14.8	0.0<∆<0.1	35.269	35.295	-0.026
3	14.854	14.8	0.0<∆<0.1	35.338	35.350	-0.012
4	14.846	14.8	0.1<∆<0.2	35.268	35.287	-0.019

Meteo France’s department for ocean observing systems (DOS-OCE) tested positively that the Automated Weather Station (AWS) BATOS could be installed onboard the SolOceans One-designs in November 2008. Since then, Meteo France is the partner for the atmospheric part of the programme and responsible for the choice, installation and maintenance of the concerned sensors as well as for the acquisition, storage, transmission, monitoring and reporting of the data to the GTS. Within the OceanoScientific® Programme the idea to acquire and transmit salinity data with BATOS was also successfully implemented.

The BATOS system deployed today uses the following sensors:
Gill Windsonic (wind speed and direction)
Rotronic S3CO3 (air humidity and temperature)
Vaisala PTB 220 (atmospheric pressure)
Sea-Bird SBE45 (sea temperature and salinity)
A Young 41003 radiation shield protects the air humidity and temperature sensor.

The sailor can regularly add visual meteorological observations. Meteo France’s special training for sailors on Voluntary Observing Ships (VOS) and BATOS’s graphical user interface enables them to easily add the extra information on the sea state, clouds, precipitation and further observations.
The data are transferred to shore on an hourly basis by an Inmarsat-C data reporting system that also delivers GPS position and speed over ground data. This is necessary to calculate true wind data, in combination with the vessel’s fluxgate compass.

The data transmitted to shore are:

• Wind averages from h-10 to h+00 • Air humidity and temperature at h+00 • Atmospheric pressure at h+00 • Sea salinity median from h-02 to h+00 • Sea temperature corresponding to salinity index • Position, course and speed over ground

Optional: Visual meteorological observations.
In addition to the transmission, all atmospheric data are stored onboard with a flexible sample rate. Sea salinity and temperature data are stored every 6 seconds.

Evaluation and Validation

From October 16^th to December 20^th, 2009 the first OceanoScientific® Kit was put through a thorough testing period. The vessel sailed the French waters between Caen (Lower Normandy) and Brest (Brittany). After the first successful trial it was put to strain in violent storms and heavy seas during a challenging voyage to Portugal.

The Laboratoire de Physique des Océans (LPO) was in charge of the sea temperature (T) and salinity (S) validation and took part in the tests in Brest from October 23^rd to 28^th.

The validation included: A visual inspection of the Sea-Bird system, water sampling for laboratory analysis and checking of the temperature and salinity data transmission by BATOS (reduced hourly dataset) and storage (full dataset sampled at 6 seconds).

The additional water samples were not taken in via the water inflow, they were collected with a bucket on October 26^th at several locations and stored in OSIL type bottles. They were analyzed for salinity at LPO. Temperature was controlled on board with a thermometer Ebro TFX 392, it should be noted that the accuracy of this equipment is lower than the SBE45 and that the measurement is given only for information.
The comparison of the water samples with the corresponding OceanoScientific® SBE45 measurements (see Tables I and II) indicates that the two independently measured salinities were within 0.02 PSS (Practical Salinity Scale) of each other.

Given the rapid space and time variability and the non-optimal sampling conditions, the achieved accuracy is very good, and complies with the scientific requirements for Sea Surface Salinity. In order to monitor the possible drift of the salinity (conductivity) sensor, water samples are taken regularly (if possible every three days).

During the test the water temperature inside the thermosalinograph (jacket temperature) did not show any warming, but this point will be validated on a more representative voyage using an external temperature sensor.
The datasets both transmitted and stored have been analyzed over the period from December 12^th to 20^th (Fig. 3 and 4).

It appears that there is very little signal loss (much less than usually seen by SOOP), although the data were taken in very heavy seas and rough wind conditions, and the comparison with water samples taken underway, and by nearby Argo floats indicates little deviation. We can thus conclude that the OceanoScientific® Kit will gather high quality, near surface data that will be extremely valuable for the calibration of the recently launched SMOS (ESA) and future Aquarius (NASA/CONAE) satellites.

Table III:
DATA COMPARISONS BETWEEN SOLOCEANS ONE-DESIGN BATOS AND SWANSEA VALE MOORED BUOY (OCT. 26TH, 2009)
Parameter	Time	SolOceans One-design	Swansea Vale	Required uncertainty
Atm. Pressure	14:00	1019.8 hPa	1019.7 hPa	±0.1 hPa
	15:00	1019.6 hPa	1019.5 hPa
Sea temperature	14:00	14.9°C	14.8°C	±0.1°C
	15:00	14.9°C	14.8°C
Wind direction	14:00	180°	180°	±5°
	15:00	170°C	170°C
Wind speed (22m)	14:00	8.2 m/s	8.1. m/s	±0.4 m/s (10% for >5m/s)
	15:00	8.2 m/s	8.1 m/s
Air temperature	14:00	15.9°C	16.0°C	±0.1°C
	15:00	15.9°C	16.0°C
Air humidity	14:00	88%	88%	±1%
	15:00	88%	90%

For the validation of the atmospheric sensors the observations of the BATOS AWS were compared to those of other platforms – two moored buoys and a light vessel – as well as to analysis outputs of two weather models – French Arpège and ECMWF (European Centre for Medium-Range Weather Forecasting).
For instance, on October 26th, 2009, the SolOceans One-design kept sailing for a couple of hours in the vicinity of a navigation buoy (Swansea Vale – 48°19.3’N-4°38.7’W) equipped with an AWS. The data of the BATOS AWS were compared to those of the moored buoy. The wind velocities measured by the buoy at 3.50 m height were corrected to 22 m (height of the anemometer above the SolOceans One-design waterline), with

W = W_ref * ln(z/z₀)/ln(z_ref/z₀) (1)

which assumes the neutral atmospheric stability conditions are met [8]. In (1), W is the wind velocity at height z above the sea level, W_ref is the reference speed (i.e. measured by the buoy) at height z_ref and z₀ is the roughness length.
With a roughness length of 0.001 m currently used at the sea surface [8], (1) becomes

W = 1.225 * W_ref (2)

Table III shows the values of different parameters measured by the two platforms as well as the recommended measurement uncertainty requirements for general operational use in meteorology [9]. It must be noted that the resolution for wind direction is still in tens of degrees as previously recommended by the WMO.

Table III also shows the very high concordance between the measurements of the two stations. It clearly appears that the measurements carried out by the BATOS AWS of the SolOceans One-design meet the WMO requirements.
The observations of the SolOceans One-design have also been compared to model outputs during all navigations at sea. Fig. 5 shows the wind speeds and directions reported by the BATOS AWS, compared to co-located model outputs. The concordance confirms that the BATOS data of the SolOceans One-design are reliable.

Conclusion

The fully automatic acquisition and transmission of underway data from aboard the SolOceans One-design Class has been demonstrated as feasible and efficient. The OceanoScientific® Programme will provide full sets of information of the ocean-atmosphere interface in hardly explored sea areas.

Regular sensor maintenance and calibration guaranty stable data quality. Optional visual observations and additional water samples can be added to complement the automatic sampling.

The regular acquisition of data on identical routes by the whole fleet of the SolOceans One-design Class make this programme a new partner for the Global Ocean Surface Underway Data (GOSUD) project of the International Oceanographic Data and Information Exchange Programme (IODE) and JCOMM. The OceanoScientific® data will be made available free of charge to climate and ocean research, to operational oceanography, meteorology and to the public.

Acknowledgment

The OceanoScientific® Programme could not have been launched without the financial support of Veolia Propreté (Veolia Environnement group), the Lower Normandy region and Bostik (Total group). Fabienne Gaillard was funded by IFREMER program PG02: Dynamique, Bio-Géochimie de l’Océan et Climat and TOSCA-CNES project GLOSCAL. The project is further supported by Laurence Eymard, Jacqueline Boutin, Nathalie Levèvre, Gilles Reverdin, Nicolas Metzl, Jean-Claude Gascard (LOCEAN), Patrick Farcy, Pascale Pessy-Martineau, Michel Hamon, Pierre Brannelec (IFREMER), Emilie Brion (CNES/INSU/LPO), Jean-Baptiste Cohuet, Clément Testa, Louis Porhel, Vinciane Unger, Mahdi Belaid (Meteo France), Théodore Danguy, Laurence Beaumont, Antoine Guillot (INSU-DT), Martin Visbeck, Arne Körtzinger (IFM-GEOMAR), Eric Thouvenot, Danielle de Staerke, Pascale Faucher, Eliane Moreaux (CNES), Frédéric Adragna (MERCATOR OCEAN), Bernhard von der Weyhe and Robert Meixner (ESA).

References

[1] Sea Surface Salinity Observation Service, Monitoring Sea Surface Salinity in the Global Ocean from Ships of Opportunity, http://www.legos.obs-mip.fr/observations/sss/, May 2010.
[2] Grenelle de la Mer, Blue Book Commitments of the Oceans Round Table, http://www.legrenelle-mer.fr/spip.php?rubrique61, August 2009.
[3] JCOMM, Ship Observation Team Fifth Session – Final Report (MR 63), http://www.jcomm.info/sot-v, June 2009.
[4] Laboratoire de Physique des Oceans (LPO), OceanoScientific, http://wwz.ifremer.fr/lpo/observation/oceanoscientific, December 2009.
[5] European Space Agency, Sailors braving treacherous waters for science, http://www.esa.int/esaLP/SEMOOU49J2G_index_0.html, December 2009.
[6] N. Lefèvre, J. Ciabrini, G. Michard, B. Brient, M. DuChaffaut, and L. Merlivat, “A new optical sensor for _pCO₂ measurements in seawater” Mar. Chem. 42, pp. 189-198, 1993.
[7] J. Boutin, L. Merlivat, C. Hénocq,N. Martin, and J.B. Sallée, “Air–sea CO₂ flux variability in frontal regions of the southern ocean from CARIOCA drifters,” Limnol. Oceanogr. vol. 53, pp. 2062-2079, 2008.
[8] J. Holton, An Introduction to Dynamic Meteorology, 4^th ed., Elsevier Academic Press, 2004.
[9] WMO, Guide to Meteorological Instruments and Methods of Observations (CIMO Guide), 7^th ed., WMO-No8, 2008. http://www.wmo.int/pages/prog/www/IMOP/IMOP-home.html